ALBERT

Beschreibung: 〈p〉Publication date: January 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 220〈/p〉 〈p〉Author(s): Xiyu Chen, Lin Liu, Annett Bartsch〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Microwave remote sensing, both active or passive, can provide useful information about the freeze/thaw (F/T) state of soil near the surface. Here we apply an edge detection algorithm on time series of indicators derived from measurements of SMAP L-band radiometer and ASCAT C-band scatterometer to detect the freeze/thaw onsets of surface soil. Comparing these results against the onsets derived from in situ measurements in Alaska, we demonstrate that this algorithm is an effective approach to detect onsets of the soil F/T transition. More specifically, our results show that the thawing onsets estimated from the SMAP data occurred 5 to 13 days earlier than the onsets estimated from the in situ measurements, which is likely due to the influence of snowmelt on the radiometer signal. The thawing onsets estimated from the ASCAT data were about 6 days later than the in situ onsets. Our estimated freezing onsets from each microwave remote sensing dataset were close to the in situ onsets (1–5 days). We also compare our estimated onsets with those from the SMAP Level 3 F/T product and the mean biases for thawing and freezing onsets are 1 ± 2 and 1 ± 3 days, respectively. Furthermore, we illustrate the complementary nature of the SMAP and ASCAT measurements and the potential for combining these two to differentiate snowmelt from soil thawing events.〈/p〉〈/div〉

Beschreibung: 〈p〉Publication date: January 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 220〈/p〉 〈p〉Author(s): Mathilde Cancet, David Griffin, Madeleine Cahill, Bertrand Chapron, Johnny Johannessen, Craig Donlon〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉Australia's marginal seas include a wide range of ocean current regimes ranging from tide-dominated in the North-West where the continental shelf is wide, to boundary current- and eddy-dominated in the South-East, where the shelf is narrow. Here, we take the opportunity to test the GlobCurrent surface ocean current products against in-situ observations in these two contrasting regimes. Observations by Acoustic Doppler Current Profiler (ADCP) instruments of Australia's Integrated Marine Observing System (IMOS) and drifting buoys of the Global Drifting Programme (GDP) are used.〈/p〉 〈p〉The monthly-timescale variability of the GlobCurrent alongshore current component is in moderately good agreement with the observations on the continental shelf in the South-East but neither the shorter period variability nor the long-term mean are representative of the ADCP observations. While the observed tidal currents are negligibly small, the wind-driven signals are not. But these are evidently too transient to be adequately sampled by altimetry. The inclusion of an Ekman component does not represent these signals because the interaction of the Ekman transport with the coastal boundary condition is not included. Similarly, the error of the time-mean velocity, which is the dominant error, is because that product is not designed to represent the highly anisotropic nature of the sea level gradients over the continental shelf nor the constraints on the flow field that are imposed by the topography. We thus conclude that the GlobCurrent product needs improvements before it can be described as very suitable for applications on this, and probably other, narrow continental shelf. Off the continental shelf, in contrast, the GlobCurrent products compare quite well with the trajectories of drifting buoys, confirming that the products are quite suitable for blue-water applications.〈/p〉 〈p〉In contrast to the South-East, the tides are very strong in the North-West region of Australia. The sub-tidal variability is weak, in both relative and absolute senses. Consequently, the removal of the tidal signal from the sea level observations needs to be very complete for the residual error to be smaller than the true sub-tidal signal. Transient wind forced signals are also occasionally large so this step of the de-aliasing also needs to be very accurate. Unfortunately, it appears that more work is required before accurate estimates of sub-tidal variability are available from GlobCurrent: the magnitude of the GlobCurrent estimates of sub-tidal current variability far exceed the magnitude of, and are uncorrelated with, the detided ADCP data.〈/p〉 〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: January 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 220〈/p〉 〈p〉Author(s): F.J.A. van Ruitenbeek, H.M.A. van der Werff, W.H. Bakker, F.D. van der Meer, K.A.A. Hein〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉A novel method is presented to measure rock microstructure in hyperspectral mineral maps of rock specimens. Shape parameters were calculated from rock objects in segmented mineral maps. Object area, object perimeter, object hull perimeter and fitted ellipses were used to calculate shape parameters such as compactness, convexity and a cookie-cutter parameter. Shape parameters were used to describe a variety of microstructures and microstructural elements. The parameters were tested on microstructures in artificial imagery and subsequently applied to hyperspectral mineral maps of rocks.〈/p〉 〈p〉Analyses of parameters calculated on artificial imagery showed that object shapes could be measured by the flattening of fitted ellipses as a measure of sphericity and elongation, together with the cookie-cutter parameters that measured angularity. Compactness and convexity could differentiate between euhedral, subhedral and anhedral crystal shapes. Aphanitic, phaneritic and porphyritic igneous microstructures could be identified and differentiated by homogeneity and relative object size parameters. The degree of sorting of sedimentary rocks was measured by the distribution of object sizes and statistical parameters describing the distribution. Orientation of single objects was measured by the angle between the major axis of a fitted ellipse and the vertical of the image. Preferred orientations in the rock microstructure were determined by calculation of a standardized resultant of orientation vectors and a mean angle. Layering and banding of the rock was identified by the length of major axes of fitted ellipses relative to the image dimension.〈/p〉 〈p〉The shape parameters calculated on objects in segmented hyperspectral mineral maps of rock specimens were able to discriminate between sedimentary and volcanic microstructures using the size distribution of mineral objects, the presence of a preferred orientation of the rock and a layered microstructure. The volcanic microstructures could be differentiated by the size distribution of amygdales, phenocrysts and xenocrysts in the rock. Shape parameters could be used to differentiate between xenocrysts and phenocrysts, the latter being more elongated in the studied samples.〈/p〉 〈p〉The study shows that object shape parameters can be used to measure microstructure and microstructural elements in mineral maps, and subsequently discriminate between different rock types and microstructures. The expression of microstructure into numeric parameters is a first step towards quantification of microstructures in mineral maps of rocks. Further development of the methodology could contribute to the creation of unbiased classification scheme of rocks, improved statistical modeling of compositional rock parameters such as mineral ore grades, and the automated recognition of microstructures in large image databases of rocks and drill-core.〈/p〉 〈/div〉 〈/div〉 〈div xml:lang="en"〉 〈h5〉Graphical abstract〈/h5〉 〈div〉〈p〉〈figure〉〈img src="https://ars.els-cdn.com/content/image/1-s2.0-S003442571830484X-ga1.jpg" width="314" alt="Unlabelled Image" title="Unlabelled Image"〉〈/figure〉〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): 〈/p〉

Beschreibung: 〈p〉Publication date: January 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 220〈/p〉 〈p〉Author(s): Jianzhi Dong, Wade T. Crow〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Due to their shallow penetration of the soil column, remote-sensing (RS) soil moisture retrievals are often considered ill-suited for measuring the strength of soil moisture-air temperature coupling. Consequently, soil moisture proxies derived from antecedent rainfall considerations are commonly applied in their place. However, the suitably of satellite RS products has not yet been examined for newer soil moisture products derived from L-band microwave radiometry. This study globally compares correlations between monthly soil moisture and the monthly number of summertime hot days (NHD) for the case of three separate RS-based soil moisture products and a fourth soil moisture proxy derived from the standard precipitation index (SPI). Compared with SPI-based estimates, C- and X-band RS soil moisture products demonstrate a significantly (at 〈em〉p〈/em〉 = 0.05 [-] confidence) weaker correlation with NHD. However, 2010–2018 L-band Soil Moisture and Ocean Salinity (SMOS) based soil moisture-NHD correlation is generally comparable to the SPI case. Furthermore, utilizing higher-precision 2015–2018 soil moisture products from the L-band Soil Moisture Active and Passive (SMAP) mission further strengthens soil moisture-NHD correlation and leads to stronger sampled correlations than SPI over global hot-spot regions (significant at 〈em〉p〈/em〉 = 0.05 [-] confidence). Combined with the general equivalence of monthly surface and root zone soil moisture anomalies, these results suggest that the signal-to-noise ratio (SNR, i.e. the relative size of soil moisture signal and random observation error variances) of RS-based surface soil moisture product, instead of their vertical measurement depth, is the key limiting factor determining their ability to quantify land-atmosphere coupling strengths. Based on this, we argue that L-band soil moisture products have reached a sufficient level of SNR to be of value for the study of land-atmosphere coupling.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): J. Antonio Guzmán Q., Benoit Rivard, G. Arturo Sánchez-Azofeifa〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Increases in liana abundance in tropical forests are pervasive threats to the current and future forest carbon stocks. Never before has the need been more evident for new approaches to detect the presence of liana in ecosystems, given their significance as fingerprints of global environmental change. In this study, we explore the use of longwave infrared reflectance (LWIR, 8–11 μm) as a wavelength region for the classification of liana and tree leaves and compare classification results with those obtained using visible-near infrared reflectance data (VIS-NIR, 0.45–0.95 μm). Twenty sun leaves were collected from each of 14 liana species and 21 tree species located at the canopy or forest edge (〈em〉n〈/em〉 = 700) in Santa Rosa National Park, Costa Rica. LWIR and VIS-NIR reflectance measurements were performed on these leaves using a portable calibrated Fourier Transform Infrared Spectroscopy (FTIR) Agilent ExoScan 4100 and a UniSpec spectral analysis system, respectively. The VIS-NIR and LWIR data were first resampled. Then these two spectral libraries were pre-processed for noise reduction and spectral feature enhancement resulting in three datasets for each spectral region as follows: filtered only, filtered followed by extraction of the first derivative, and continuous wavelet transformation (CWT). Data reduction was then applied to these data sets using principal components analysis (PCA). The outputs obtained from the PCA were used to conduct the supervised classification of liana and tree leaves. In total, 21 classifiers were applied to datasets of training and testing to extract the classification accuracy and agreement for liana and tree leaves. The results suggest that the classification of leaves based on LWIR data can reach accuracy values between 66 and 96% and agreement values between 32 and 92%, regardless of the type of classifier. In contrast, the classification based on VIS-NIR data shows accuracy values between 50 and 70% and agreement values between 0.01 and 40%. The highest classification rates of liana and tree leaves were obtained from datasets pre-processed using the CWT or from the extraction of the first derivative and classified using either random forest, 〈em〉k〈/em〉-nearest neighbor, or support vector machine with radial kernel. The results using the LWIR reflectance highlight the potential of this spectral region for the accurate detection of liana extent in tropical ecosystems. Future studies should consider this potential and test the regional monitoring of lianas.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Adam J. Purdy, Joshua B. Fisher, Michael L. Goulden, Andreas Colliander, Gregory Halverson, Kevin Tu, James S. Famiglietti〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Accurate estimation of global evapotranspiration (ET) is essential to understand water cycle and land-atmosphere feedbacks in the Earth system. Satellite-driven ET models provide global estimates, but many of the ET algorithms have been designed independently of soil moisture observations. As water for ET is sourced from the soil, incorporating soil moisture into global remote sensing algorithms of ET should, in theory, improve performance, especially in water-limited regions. This paper presents an update to the widely-used Priestley Taylor-Jet Propulsion Laboratory (PT-JPL) ET algorithm to incorporate spatially explicit daily surface soil moisture control on soil evaporation and canopy transpiration. The updated algorithm is evaluated using 14 AmeriFlux eddy covariance towers co-located with COsmic-ray Soil Moisture Observing System (COSMOS) soil moisture observations. The new PT-JPL〈sub〉SM〈/sub〉 model shows reduced errors and increased explanation of variance, with the greatest improvements in water-limited regions. Soil moisture incorporation into soil evaporation improves ET estimates by reducing bias and RMSE by 29.9% and 22.7% respectively, while soil moisture incorporation into transpiration improves ET estimates by reducing bias by 30.2%, RMSE by 16.9%. We apply the algorithm globally using soil moisture observations from the Soil Moisture Active Passive Mission (SMAP). These new global estimates of ET show reduced error at finer spatial resolutions and provide a rich dataset to evaluate land surface and climate models, vegetation response to changes in water availability and environmental conditions, and anthropogenic perturbations to the water cycle.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: Available online 19 September 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment〈/p〉 〈p〉Author(s): Sylvain Jay, Frédéric Baret, Dan Dutartre, Ghislain Malatesta, Stéphanie Héno, Alexis Comar, Marie Weiss, Fabienne Maupas〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉The recent emergence of unmanned aerial vehicles (UAV) has opened a new horizon in vegetation remote sensing, especially for agricultural applications. However, the benefits of UAV centimeter-scale imagery are still unclear compared to coarser resolution data acquired from satellites or aircrafts. This study aims (i) to propose novel methods for retrieving canopy variables from UAV multispectral observations, and (ii) to investigate to what extent the use of such centimeter-scale imagery makes it possible to improve the estimation of leaf and canopy variables in sugar beet crops (〈em〉Beta vulgaris〈/em〉 L.). Five important structural and biochemical plant traits are considered: green fraction (GF), green area index (GAI), leaf chlorophyll content (C〈sub〉ab〈/sub〉), as well as canopy chlorophyll (CCC) and nitrogen (CNC) contents.〈/p〉 〈p〉Based on a comprehensive data set encompassing a large variability in canopy structure and biochemistry, the results obtained for every targeted trait demonstrate the superiority of centimeter-resolution methods over two standard remote-sensing approaches (i.e., vegetation indices and PROSAIL inversion) applied to average canopy reflectances. Two variables (denoted GF〈sub〉GREENPIX〈/sub〉 and VI〈sub〉CAB〈/sub〉) extracted from the images are shown to play a major role in these performances. GF〈sub〉GREENPIX〈/sub〉 is the GF estimate obtained by thresholding the Visible Atmospherically Resistant Index (〈em〉VARI〈/em〉) image, and is shown to be an accurate and robust (e.g., against variable illumination conditions) proxy of the structure of sugar beet canopies, i.e., GF and GAI. VI〈sub〉CAB〈/sub〉 is the 〈em〉mND〈/em〉〈sub〉〈em〉blue〈/em〉〈/sub〉 index value averaged over the darkest green pixels, and provides critical information on C〈sub〉ab〈/sub〉. When exploited within uni- or multivariate empirical models, these two variables improve the GF, GAI, C〈sub〉ab〈/sub〉, CCC and CNC estimates obtained with standard approaches, with gains in estimation accuracy of 24, 8, 26, 37 and 8%, respectively. For example, the best CCC estimates (〈em〉R〈/em〉〈sup〉2〈/sup〉 = 0.90) are obtained by multiplying C〈sub〉ab〈/sub〉 and GAI estimates respectively derived from VI〈sub〉CAB〈/sub〉 and a log-transformed version of GF〈sub〉GREENPIX〈/sub〉, log(1-GF〈sub〉GREENPIX〈/sub〉).〈/p〉 〈p〉The GF〈sub〉GREENPIX〈/sub〉 and VI〈sub〉CAB〈/sub〉 variables, which are only accessible from centimeter-scale imagery, contributes to a better identification of the effects of canopy structure and leaf biochemistry, whose influences may be confounded when considering coarser resolution observations. Such results emphasize the strong benefits of centimeter-scale UAV imagery over satellite or airborne remote sensing, and demonstrate the relevance of low-cost multispectral cameras to retrieve a number of plant traits, e.g., for agricultural applications.〈/p〉 〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Huan Li, Hongjie Xie, Stefan Kern, Wei Wan, Burcu Ozsoy, Stephan Ackley, Yang Hong〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We use total (sea ice plus snow) freeboard as estimated from Ice, Cloud and land Elevation Satellite (ICESat) Geophysical Laser Altimeter System (GLAS) observations to compute Antarctic sea-ice thickness and volume. In order to overcome assumptions made about the relationship between snow depth and total freeboard or biases in snow depth products from satellite microwave radiometry, we implement a new algorithm. We treat the sea ice-snow system as one layer with reduced density, which we approximate by means of a priori information about the snow depth to sea-ice thickness ratio. We derive this a priori information directly from ICESat total freeboard data using empirical equations relating in-situ measurements of total freeboard to snow depth or sea-ice thickness. We apply our new algorithm (one-layer method or OLM), which uses the buoyancy equation approach without the need for auxiliary snow depth data, to compute sea-ice thickness for every ICESat GLAS footprint from a valid total freeboard. An improved method for sea-ice volume retrieval is also used to derive ice volume at 6.25 km scale. Spatio-temporal variations of sea-ice thickness and volume are then analyzed in the circumpolar Antarctic as well as its six sea sectors: Pacific Ocean, Indian Ocean, Weddell East, Weddell West, Bell-Amund Sea, and Ross Sea, under both interannual and seasonal scales. Because the OLM algorithm relies on only one parameter, the total freeboard, and is independent of auxiliary snow depth information, it is believed to become a viable alternative sea-ice thickness retrieval method for satellite altimetry.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: Available online 19 June 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment〈/p〉 〈p〉Author(s): Xinjie Liu, Luis Guanter, Liangyun Liu, Alexander Damm, Zbyněk Malenovský, Uwe Rascher, Dailiang Peng, Shanshan Du, Jean-Philippe Gastellu-Etchegorry〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Solar-induced chlorophyll fluorescence (SIF), an electromagnetic signal that can potentially indicate vegetation photosynthetic activity, can be retrieved from ground-based, airborne and satellite measurements. However, due to the scattering and re-absorption effects inside the leaves and canopy, SIF measured at the canopy level is only a small part of the total SIF emission at the photosystem level. Therefore, a downscaling mechanism of SIF from the canopy level to the photosystem level is important for better understanding the relationship between SIF and the vegetation gross primary production (GPP). In this study, firstly, we analyzed the canopy scattering effects using a simple parameterization model based on the spectral invariant theory. The probability for SIF photons to escape from the canopy was found to be related to the anisotropic spectral reflectance, canopy interception of the upward solar radiation, and leaf absorption. An empirical approach based on a Random Forest (RF) regression algorithm was applied to downscale SIF constrained by the red, red-edge and far-red anisotropic reflectance. The RF was trained using simulations conducted with the Soil Canopy Observation, Photochemistry and Energy fluxes (SCOPE) model. The performance of the SIF downscaling method was evaluated with SCOPE and Discrete Anisotropic Radiative Transfer (DART) model simulations, ground measurements and airborne data. Results show that estimated SIF at the photosystem level matches well with simulated reference data, and the relationship between SIF and photosynthetically active radiation absorbed by chlorophyll is improved by SIF downscaling. This finding in combination with other evaluation criteria suggests the downscaling of canopy SIF as an efficient strategy to normalize species dependent effects of canopy structure and varying solar-view geometries. Based on our results for the SIF-APAR relationship, we expect that such normalization approaches can be helpful to improve estimates of photosynthesis using remote sensing measurements of SIF.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Nan Chen, Wei Li, Charles Gatebe, Tomonori Tanikawa, Masahiro Hori, Rigen Shimada, Teruo Aoki, Knut Stamnes〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Cloud detection and screening constitute critically important first steps required to derive many satellite data products. Traditional threshold-based cloud mask algorithms require a complicated design process and fine tuning for each sensor, and they have difficulties over areas partially covered with snow/ice. Exploiting advances in machine learning techniques and radiative transfer modeling of coupled environmental systems, we have developed a new, threshold-free cloud mask algorithm based on a neural network classifier driven by extensive radiative transfer simulations. Statistical validation results obtained by using collocated CALIOP and MODIS data show that its performance is consistent over different ecosystems and significantly better than the MODIS Cloud Mask (MOD35 C6) during the winter seasons over snow-covered areas in the mid-latitudes. Simulations using a reduced number of satellite channels also show satisfactory results, indicating its flexibility to be configured for different sensors. Compared to threshold-based methods and previous machine-learning approaches, this new cloud mask (i) does not rely on thresholds, (ii) needs fewer satellite channels, (iii) has superior performance during winter seasons in mid-latitude areas, and (iv) can easily be applied to different sensors.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Xiaoping Wang, Fei Zhang, Hsiang-te Kung, Verner Carl Johnson〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉This study aimed to improve the potential of Analytical Spectral Devices (ASD) hyperspectral and Landsat Operational Land Imager (OLI) data in predicting soil organic matter content (SOMC) in the bare topsoil of the Ebinur Lake Wetland National Nature Reserve (ELWNNR) in northwest China. The results indicated that the correlation of coefficients (R) between SOMCs and hyperspectral data processed by fractional derivative were significant at the 0.01 level; the number of wave bands increased initially and then decreased when the order increased. The correlation of coefficient peak appeared at the 1.2 order with a value of 0.52. The correlation of coefficients (R) between SOMCs and the optimal remote sensing indexes (the ratio index, RI; difference index, DI; and the normalized difference index, NDI) of peaked at the 1.2 order, with correlation of coefficients (R) values of 0.81, 0.86 and 0.82, respectively. Six SOMC estimation models were created by means of a single band and optimal remote sensing indexes using Gray Relational Analysis-BP Neural Network (GRA-BPNN). This study found that the optimal model was a 1.2 order derivative model, where the lowest root mean square error (RMSE) was 3.26 g/kg, the highest was 0.92, and the residual prediction deviation (RPD) was 2.26. To complete the high accuracy retrieval of SOMCs, based on Landsat OLI operational land images data, more ‘hidden’ information from the Landsat OLI images were obtained by employing the subsection of spectral band method and the fractional derivative algorithm. Accuracy of the SOMC map was attained by the optimal model of the ground hyperspectral data and the Landsat OLI data, which had low RMSE values of 4.21 g/kg and 4.16 g/kg, respectively. Therefore, we conclude that the SOMC can be estimated and retrieved using a fractional derivative algorithm, the subsection of spectral band method, and the optimal remote sensing index.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Liguang Jiang, Ole Baltazar Andersen, Karina Nielsen, Guoqing Zhang, Peter Bauer-Gottwein〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Water surface elevation (WSE) is an essential quantity for water resource monitoring and hydrodynamic modeling. Satellite altimetry has provided data for inland water bodies. The height that is derived from altimetry measurement is ellipsoidal height. In order to convert the ellipsoidal height to orthometric height, which has physical meaning, accurate estimates of the geoid are needed. This paper evaluates the suitability of geodetic altimetric measurements for improvement of global geoid models over a large lake in the Tibetan Plateau. CryoSat-2 and SARAL/AltiKa are used to derive the high-frequency geoid correction. A validation of the local geoid correction is performed with data from in-situ observations, a laser altimetry satellite (ICESat), a Ka-band radar altimetry satellite (SARAL) and a SAR radar altimetry satellite (Sentinel-3). Results indicate that the geodetic altimetric dataset can capture the high-resolution geoid information. By applying local geoid correction, the precision of ICESat, SARAL and Sentinel-3 retrievals are significantly improved. We conclude that using geodetic altimetry to correct for local geoid residual over large lakes significantly decreases the uncertainty of WSE estimates. These results also indicate the potential of geodetic altimetry missions to determine local geoid residual with centimeter-level accuracy, which can be used to improve global and regional geopotential models.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Radoslaw Guzinski, Héctor Nieto〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The Sentinel satellite missions are designed to provide remote-sensing observational capability to many diverse operational applications, including in the field of agriculture and food security. They do this by acquiring frequent observations from a combination of optical, thermal and microwave sensors at various spatial resolutions. However, one currently missing capability, that would enable monitoring of evapotranspiration, crop water stress and water use at field scale, is the lack of high-resolution (tens of meters) thermal sensor. In this study we evaluate a methodology for bridging this data gap by employing a machine learning algorithm to sharpen low-resolution thermal observations from the Sentinel-3 satellites using images acquired by high-resolution optical sensors on the Sentinel-2 satellites. The resulting dataset is then used as input to land-surface energy balance model to estimate evapotranspiration. The methodology is tested using Terra and Landsat satellite observations, due to lack of sufficiently long time-series of Sentinel observations, and benchmarked against fluxes derived with high-resolution thermal observations acquired by the Landsat satellites. We then apply the methodology to Sentinel-2 and Sentinel-3 images to confirm its applicability to this type of data. The results show that the fluxes derived with sharpened thermal data are of acceptable accuracy (relative error lower than 20%) and provide more information at flux-tower footprint scale than the corresponding low-resolution fluxes. They also replicate the spatial and temporal patterns of fluxes derived with high-resolution thermal observations. However, the increase in error of the modelled fluxes compared to using high-resolution thermal observations and the inherent limitations of the sharpening approach point to the need to add high-resolution thermal mission to the Sentinels' constellation.〈/p〉〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Wenlu Qi, Seung-Kuk Lee, Steven Hancock, Scott Luthcke, Hao Tang, John Armston, Ralph Dubayah〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Interferometric Synthetic Aperture Radar (InSAR) and lidar are increasingly used active remote sensing techniques for forest structure observation. The TanDEM-X (TDX) InSAR mission of German Aerospace Center (DLR) and the upcoming Global Ecosystem Dynamics Investigation (GEDI) of National Aeronautics and Space Administration (NASA) together may provide more accurate estimates of global forest structure and biomass via their synergic use. In this paper, we explored the efficacy of simulated GEDI data in improving height estimates from TDX InSAR data. Our study sites span three major forest types: a temperate forest, a mountainous conifer forest, and a tropical rainforest. The GEDI lidar coverage was simulated for the full nominal two-year mission duration, under both cloud-free and 50%-cloud conditions. We then used these GEDI data to parameterize the Random Volume over Ground (RVoG) model driven by TDX imagery. In particular, we explored the following three strategies for forest structure estimation: 1) TDX data alone; 2) TDX + GEDI-derived digital terrain model (DTM); and 3) TDX + GEDI DTM + GEDI canopy height. We then validated the retrieved forest heights against wall-to-wall airborne lidar measurements. We found relatively large biases at 90 [m] spatial resolution, from 4.2–11.9 [m], and root mean square errors (RMSEs), from 7.9–12.7 [m] when using TDX data alone under constrained RVoG assumptions of a fixed extinction coefficient (〈em〉σ〈/em〉) and a zero ground-to-volume amplitude ratio (〈em〉μ〈/em〉 〈em〉=〈/em〉 〈em〉0〈/em〉). Results improved significantly with the aid of a DTM derived from GEDI data which enabled estimation of spatially-varying 〈em〉σ〈/em〉 values (vs. fixed extinction) under a 〈em〉μ〈/em〉 = 0 assumption, with biases reduced to 1.7–4.2 [m] and RMSEs to 4.9–8.6 [m] across cloudy and cloud-free cases. The best agreement was achieved in the third strategy by also incorporating information of GEDI-derived canopy height to further enhance the RVoG parameters. The improved model, when still assuming 〈em〉μ〈/em〉 = 0, reduced biases to less than or close to 1 m and further reduced RMSEs to 4.0–6.7 [m]. Finally, we used GEDI data to estimate spatially-varying 〈em〉μ〈/em〉 in the RVoG model. We found biases of between −0.7–0.9 [m] and RMSEs in the range from 2.6–7.1 [m] over the three sites. Our results suggest that use of GEDI data improves height inversion from TDX, providing heights at more accuracy than can be achieved by TDX alone, and enabling wall-to-wall height estimation at much finer spatial resolution than can be achieved by GEDI alone.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Liang Chang, Ruya Xiao, Abhnil Amtesh Prasad, Guoping Gao, Guiping Feng, Yu Zhang〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Water vapor is the primary greenhouse gas of the Earth-atmosphere system and plays a vital role in understanding climate change, if correctly measured from satellites. The Moderate Resolution Imaging Spectroradiometer (MODIS) can monitor water vapor retrievals at near-infrared (nIR) bands in the daytime as well as at infrared (IR) bands in both daytime and night time. However, the accuracy of IR retrievals under confident clear conditions (〉99% probability) is much poorer than that of nIR retrievals. Additionally, IR retrievals under unconfident clear conditions (〉95%, 〉66% and ≤66% probabilities) are usually discarded because the possible presence of clouds would further reduce their accuracy. In this study, we develop a cloud mask-related differential linear adjustment model (CDLAM) to adjust IR retrievals under all confident clear conditions. The CDLAM-adjusted IR retrievals are evaluated with the linear least square (LS) adjusted nIR retrievals under confident clear condition and Global Positioning System (GPS) observations under different probabilities of clear conditions. Both case studies in the USA and global (65° S~65° N) evaluation reveal that the CDLAM can significantly reduce uncertainties in IR retrievals at all clear-sky confidence levels. Moreover, the accuracy of the CDLAM-adjusted IR retrievals under unconfident clear conditions is much better than IR retrievals without adjustment under confident clear conditions, highlighting the effectiveness of the CDLAM in enhancing the accuracy of IR retrievals at all clear-sky confidence levels as well as the data availability improvement of IR retrievals after adjustment with the CDLAM (14% during the analyzed time periods). The most likely reason for the efficiency of the CDLAM may be that the deviation of the differential water vapor information derived by the differential process is significantly shrunken after the linear regression analysis in the presented model. Therefore, the CDLAM is a promising tool for effectively adjusting IR retrievals under all probabilities of clear conditions and can improve our knowledge of the water vapor distribution and variation.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Eduardo Eiji Maeda, Filipe Lisboa, Laura Kaikkonen, Kari Kallio, Sampsa Koponen, Vanda Brotas, Sakari Kuikka〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Monitoring temporal changes in phytoplankton dynamics in high latitude lakes is particularly timely for understanding the impacts of warming on aquatic ecosystems. In this study, we analyzed 33-years of high resolution (30 m) Landsat (LT) data for reconstructing seasonal patterns of chlorophyll 〈em〉a〈/em〉 (chl 〈em〉a〈/em〉) concentration in four lakes across Finland, between 60°N and 64°N. Chl 〈em〉a〈/em〉 models based on LT spectral bands were calibrated using 17-years (2000–2016) of field measurements collected across the four lakes. These models were then applied for estimating chl 〈em〉a〈/em〉 using the entire LT-5 and 7 archives. Approximately 630 images, from 1984 to 2017, were analyzed for each lake. The chl 〈em〉a〈/em〉 seasonal patterns were characterized using phenology metrics, and the time-series of LT-based chl 〈em〉a〈/em〉 estimates were used for identifying temporal shifts in the seasonal patterns of chl 〈em〉a〈/em〉 concentration. Our results showed an increase in the length of phytoplankton growth season in three of the lakes. The highest increase was observed in Lake Köyliönjärvi, where the length of growth season has increased by 28 days from the baseline period of 1984–1994 to 2007–2017. The increase in the length of season was mainly attributed to an earlier start of phytoplankton blooms. We further analyzed surface temperature (T〈sub〉s〈/sub〉) and precipitation data to verify if climatic factors could explain the shifts in the seasonal patterns of chl 〈em〉a〈/em〉. We found no direct relationship between T〈sub〉s〈/sub〉 and chl 〈em〉a〈/em〉 seasonal patterns. Similarly, the phenological metrics of Ts, in particular length of season, did not show significant temporal trends. On the other hand, we identify potential links between changes in precipitation patterns and the increase in the phytoplankton season length. We verified a significant increase in the rainfall contribution to the total precipitation during the autumn and winter, accompanied by a decline in snowfall volumes. This could indicate an increasing runoff volume during the beginning of spring, contributing to an earlier onset of the phytoplankton blooms, although further assessments are needed to analyze historical streamflow values and nearby land cover data. Likewise, additional studies are needed to better understand why chl 〈em〉a〈/em〉 patterns in some lakes seem to be more resilient than in others.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Wei Zhao, Si-Bo Duan, Ainong Li, Gaofei Yin〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Land surface temperature (LST) plays a key role in connecting land surface energy and water exchanges with near-surface atmosphere. However, the spatial distribution of LST over mountainous areas is not only strongly affected by the differences in surface thermal properties but also by the differences in thermal or radiative environment induced by the topographic variations, presenting significant terrain effect. This effect greatly hinders researches on surface energy fluxes and soil moisture estimation in these regions. To normalize the terrain effect, a random forest (RF)-based normalization approach was developed in this study and applied to Moderate Resolution Imaging Spectroradiometer (MODIS) data over a typical mountain region in the eastern part of the Tibetan Plateau, China. An LST linking model was first constructed to express the complicated interrelationship between LST and other surface variables, including normalized difference vegetation index (NDVI), enhanced vegetation index (EVI), leaf area index (LAI), surface albedo (ALB), cumulative incident solar radiation (CSR), normalized difference water index (NDWI), surface elevation (ELV), and surface slope (SLP). The estimation results well indicated the good accuracy of the model with the coefficient of determination (R〈sup〉2〈/sup〉) above 0.92 for 90 selected days in 2015. Based on the good LST linking model, the LST normalization was achieved by replacing the topography affected factors (CSR, ELV, and SLP) with reference values for each pixel when keeping the rest status factors as their original values. The normalization result was indirectly validated by the normalization results with the traditional method based on the linear correction upon the regression between surface elevation and LST. The cross-validation clearly indicated the proposed method had an obvious advantage in reducing the topography-induced LST difference based on the average decrease in the temperature range (9.51 K) and the standard deviation (1.09 K) for the images on the 90 selected days over the study area. In contrast, the traditional method was hard to capture the terrain effect only based on the relationship between LST and surface elevation due to the complex interaction between LST and other factors. Overall, the proposed method shows a good application potential for normalizing the terrain effect on LST, and the results will be helpful for the LST-based estimation of surface energy fluxes or soil moisture over mountainous areas.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): J. Gómez-Enri, C.J. González, M. Passaro, S. Vignudelli, O. Álvarez, P. Cipollini, R. Mañanes, M. Bruno, M.P. López-Carmona, A. Izquierdo〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Coastal altimetry products are available and are being extensively validated. Their accuracy has been assessed in many coastal zones around the world and they are ready for exploitation near the shore. This opens a variety of applications of the sea level data obtained from the specific reprocessing of radar altimeter signals in the coastal strip. In this work, we retracked altimeter waveforms of the European Space Agency satellites: ERS-2 RA and Envisat RA-2 from descending track (#0360) over the eastern side of the Strait of Gibraltar using the Adaptive Leading Edge Sub-waveform (ALES) retracker. We estimated along-track Sea Level Anomaly (AT_SLA) profiles (RA-2) at high posting rate (18 Hz) using improved range and geophysical corrections. Tides were removed with a global model (DTU10) that displays a good performance in the study area: the mean root square sum (〈em〉RSS〈/em〉) of the main constituents obtained with 〈em〉DTU10〈/em〉 and 11 tide gauge stations was 4.3 cm in agreement with the 〈em〉RSS〈/em〉 using a high-resolution local hydrodynamic model (〈em〉UCA2.5D〈/em〉) (4.2 cm). We also estimated a local mean sea surface by reprocessing ERS-2/Envisat waveforms (track #0360) with ALES. The use of this local model gave more realistic AT_SLA than the values obtained with the global model DTU15MSS. Finally, the along-track Absolute Dynamic Topography (AT_ADT) was estimated using a local Mean Dynamic Topography obtained with the local hydrodynamic model UCA2.5D. We analysed the cross-strait variability of the sea level difference between the African/Spanish coasts along the selected track segment. This was compared to the sea level cross-strait difference from the records of two tide gauges located in the African (Ceuta) and Spanish (Tarifa) coasts. The sea level differences from altimetry and tide gauges were linked to the zonal component of the wind. We found a positive and significant (〉95% c.l.) correlation between easterlies/westerlies and positive/negative cross-strait sea level differences between the southern and northern coasts of the Strait in both datasets (altimetry: r = 0.54 and in-situ: r = 0.82).〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Zhanmang Liao, Binbin He, Xingwen Quan, Albert I.J.M. van Dijk, Shi Qiu, Changming Yin〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉With the upcoming BIOMASS mission, P-band PolInSAR is expected to provide new perspectives on global forest aboveground biomass (AGB). However, its performance has not yet been fully evaluated for dense tropical forests with complex structure and very high biomass. Based on the TropiSAR campaign in French Guiana, we explored the challenges of the three most commonly used PolInSAR measures to capture AGB in tropical forests; coherence magnitude, interferometric phase, and backscatter. An improved AGB estimation approach was developed by integrating multiple information derived from single-baseline PolInSAR data. The approach involves ground-volume backscatter decomposition and combines volume backscatter with the retrieved forest height. Volume backscatter from the forest canopy was the best predictor of AGB for tropical forests, whereas the ground backscatter contribution was affected by the complex underlying surface and terrain slope. Both LiDAR- and PolInSAR-derived forest heights showed limited correlation with high AGB due to the varying forest basal area. The linear combination of PolInSAR-derived forest height and volume backscatter complemented each other and produced improved AGB estimates. Comparing three different PolInSAR data pairs, the proposed method produced an AGB map with an average R〈sup〉2〈/sup〉 of 0.7 and RMSE of 34 tons/ha (relative RMSE of 9.4%) at a spatial resolution of 125 × 125 m〈sup〉2〈/sup〉 for biomass between 250–500 tons/ha.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Mark J. Lara, Melissa L. Chipman, Feng Sheng Hu〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Anthropogenic climate change has been linked to the degradation of permafrost across northern ecosystems, with notable implications for regional to global carbon dynamics. However, our understanding of the spatial distribution, temporal trends, and seasonal timing of episodic landscape deformation events triggered by permafrost degradation is hampered by the limited spatial and temporal coverage of high-resolution optical, RADAR, LIDAR, and hyperspectral remote sensing products. Here we present an automated approach for detecting permafrost degradation (thermoerosion), using meso-scale high-frequency remote sensing products (i.e., Landsat image archive). This approach was developed, tested, and applied in the ice-rich lowlands of the Noatak National Preserve (NOAT; 12,369 km〈sup〉2〈/sup〉) in northwestern Alaska. We identified thermoerosion (TE) by capturing the spectral signal associated with episodic sediment plumes in adjacent water bodies following TE. We characterized and extracted this episodic turbidity signal within lakes during the snow-free period (June 15–October 1) for 1986–2016 (continuous data limited to 1999–2016), using the cloud-based geospatial parallel processing platform, Google Earth Engine™. Thermoerosional detection accuracy was calculated using seven consecutive years of sub-meter high-resolution imagery (2009–2015) covering 798 (~33%) of the 2456 lakes in the NOAT lowlands. Our automated TE detection algorithm had an overall accuracy and kappa coefficient of 86% and 0.47 ± 0.043, indicating that episodic sediment pulses had a “moderate agreement” with landscape deformation associated with permafrost degradation. We estimate that lake shoreline erosion, thaw slumps, catastrophic lake drainage, and gully formation accounted for 62, 23, 13, and 2%, respectively, of active TE across the NOAT lowlands. TE was identified in ~5% of all lakes annually in the lowlands between 1999 and 2016, with a wide range of inter-annual variation (ranging from 0.2% in 2001 to 22% in 2004). Inter-annual variability in TE occurrence and spatial patterns of TE probability were correlated with annual snow cover duration and snow persistence, respectively, suggesting that earlier snowmelt accelerates permafrost degradation (e.g. TE) in this region. This work improves our ability to detect and attribute change in permafrost degradation across space and time.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Barbara Barzycka, Małgorzata Błaszczyk, Mariusz Grabiec, Jacek Jania〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The warming climate of the Arctic affects the mass budget of glaciers, and changes in the distribution of glacier facies are indicative of their response to climate change. The glacial mass budget over large land ice masses can be estimated by remote sensing techniques, but selecting an efficient remote sensing method for recognizing and mapping glacier facies in the Arctic remains a challenge. In this study, we compared several methods of distinguishing the facies of the Vestfonna ice cap, Svalbard, based upon Synthetic Aperture Radar (SAR) images and terrestrial high frequency Ground Penetrating Radar (GPR) measurements. Glacier zones as determined using the backscattering coefficient (sigma0) of SAR images were compared against GPR data, and an alternative application of Internal Reflection Energy (IRE) calculated from terrestrial GPR data was also used for differentiating the extent of glacier facies. The IRE coefficient was found to offer a suitable method for distinguishing glacier zones and for validating SAR analysis. Furthermore, results of analysis of fully polarimetric Phased Array type L-band Synthetic Aperture Radar (ALOS PALSAR) and European Remote Sensing Synthetic Aperture Radar (ERS-2 SAR) images were compared with the IRE coefficient classification. Especially promising method is H-α segmentation, where the glacier zone boundaries corresponded very well with both GPR visual interpretation and IRE classification results. The IRE coefficient's simplicity of calculation makes it a good alternative to the subjective GPR visual interpretation method, where results strongly depend on the operator's level of experience. We therefore recommend for GPR profiles to be used for additional validation of SAR image analysis in studies of glacier facies on the High Arctic ice masses.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Donghwan Kim, Hanwen Yu, Hyongki Lee, Edward Beighley, Michael Durand, Douglas E. Alsdorf, Euiho Hwang〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Hydraulic variables obtained from remotely sensed data have been successfully used to estimate river discharge (〈em〉Q〈/em〉). However, most studies have used a rating curve based on a single hydraulic variable or the Manning equation (multiplicative method). In this study, we developed a mathematically different approach to estimating 〈em〉Q〈/em〉 by applying the ensemble learning regression method (here termed ELQ), which is one of the machine learning techniques that linearly combine several functions to reduce errors, over the Congo mainstem as a test-bed. Using the training dataset (November 2002 – November 2006) of water levels (〈em〉H〈/em〉) derived from different Envisat altimetry observations, the ELQ-estimated 〈em〉Q〈/em〉 at the Brazzaville in-situ station showed reduced root-mean-square error (RMSE) of 823 m〈sup〉3〈/sup〉 s〈sup〉−1〈/sup〉 (relative RMSE (RMSE normalized by the average in-situ 〈em〉Q〈/em〉, RRMSE) of 2.08%) compared to the 〈em〉Q〈/em〉 obtained using a single rating curve. ELQ also showed improved performance for the validation dataset (December 2006 – September 2010). Based on the error analysis, we found the correlation coefficients between input variables affect the performance of ELQ. Thus, we introduced an index, termed the Degree of compensation (〈em〉I〈/em〉〈sub〉〈em〉DoC〈/em〉〈/sub〉), which describes how ELQ performs compared to the classic hydraulic relation (e.g., 〈em〉H〈/em〉-〈em〉Q〈/em〉 rating curve). The performance of ELQ improves when 〈em〉I〈/em〉〈sub〉〈em〉DoC〈/em〉〈/sub〉 increases because the additional information could be added in the ELQ process. Since ELQ can combine several variables obtained over different locations, it would be advantageous, particularly if there exist few virtual stations along a river reach. It is expected that ELQ can be also applied to the products of the Surface Water Ocean Topography (SWOT) mission, which will provide direct measurements of surface water slope (〈em〉S〈/em〉), effective river width (〈em〉W〈/em〉〈sub〉〈em〉e〈/em〉〈/sub〉), and 〈em〉H〈/em〉, to be launched in 2021.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 March 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 222〈/p〉 〈p〉Author(s): Emma K. Fiedler, Alison McLaren, Viva Banzon, Bruce Brasnett, Shiro Ishizaki, John Kennedy, Nick Rayner, Jonah Roberts-Jones, Gary Corlett, Christopher J. Merchant, Craig Donlon〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Six global, gridded, gap-free, daily sea surface temperature (SST) analyses covering a period of at least 20 years have been intercompared: ESA SST CCI analysis long-term product v1.0, MyOcean OSTIA reanalysis v1.0, CMC 0.2 degree, AVHRR_ONLY Daily 1/4 degree OISST v2.0, HadISST2.1.0.0 and MGDSST. A seventh SST product of the ensemble median of all six has also been produced using the GMPE (Group for High Resolution SST Multi-Product Ensemble) system. Validation against independent near-surface Argo data, a long timeseries of moored buoy data from the tropics and anomalies to the GMPE median have been used to examine the temporal and spatial homogeneity of the analyses. A comparison of the feature resolution of the analyses has also been undertaken. A summary of relative strengths and weaknesses of the SST datasets is presented, intended to help users to make an informed choice of which analysis is most suitable for their proposed application.〈/p〉〈/div〉

Beschreibung: 〈p〉Publication date: Available online 21 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment〈/p〉 〈p〉Author(s): Yun Yang, Martha Anderson, Feng Gao, Christopher Hain, Asko Noormets, Ge Sun, Randolph Wynne, Valerie Thomas, Liang Sun〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Forest ecosystem services such as clean water, wildlife habitat, and timber supplies are increasingly threatened by drought and disturbances (e.g., harvesting, fires and conversion to other uses), which can have great impacts on stand development and water balance. Improved understanding of the hydrologic response of forested systems to drought and disturbance at spatiotemporal resolutions commensurate with these impacts is important for effective forest management. Evapotranspiration (ET) is a key hydrologic variable in assessing forest functioning and health, but it remains a challenge to accurately quantify ET at landscape scales with the spatial and temporal detail required for effective decision-making. In this study, we apply a multi-sensor satellite data fusion approach to study the response of forest ET to drought and disturbance over a 7-year period. This approach combines Landsat and Moderate Resolution Imaging Spectroradiometer (MODIS) ET product time series retrieved using a surface energy balance model to generate a multi-year ET datacube at 30-m resolution and daily timesteps. The study area (~900 km〈sup〉2〈/sup〉) contains natural and managed forest as well as croplands in the humid lower coastal plains in North Carolina, USA, and the simulation period from 2006 to 2012 includes both normal and severe drought conditions. The model results were evaluated at two AmeriFlux sites (US-NC2 and US-NC1) dominated by a mature and a recently clearcut pine plantation, respectively, and showed good agreement with observed fluxes, with 8–13% relative errors at monthly timesteps. Changes in water use patterns in response to drought and disturbance as well as forest stand aging were assessed using the remotely sensed time series describing total evapotranspiration, the transpiration (T) component of ET, and a moisture stress metric given by the actual-to-reference ET ratio (〈em〉f〈/em〉〈sub〉〈em〉RET〈/em〉〈/sub〉). Analyses demonstrate differential response to drought by land cover type and stand age, with larger impacts on total ET observed in young pine stands than in mature stands which have substantially deeper rooting systems. Transpiration flux shows a clear ascending trend with the growth of young pine plantations, while stand thinning within the plantation leads to decreases in both remotely sensed leaf area index and T, as expected. Time series maps of 〈em〉f〈/em〉〈sub〉〈em〉RET〈/em〉〈/sub〉 anomalies at 30-m resolution capture signals of drought, disturbance and the subsequent recovery after clearcut at the stand scale and may be an effective indicator for water use change detection and monitoring in forested landscapes.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Abolfazl Irani Rahaghi, Ulrich Lemmin, Daniel Sage, David Andrew Barry〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉A two-platform measurement system for realizing airborne thermography of the Lake Surface Water Temperature (LSWT) with ~0.8 m pixel resolution (sub-pixel satellite scale) is presented. It consists of a tethered Balloon Launched Imaging and Monitoring Platform (BLIMP) that records LSWT images and an autonomously operating catamaran (called ZiviCat) that measures in situ surface/near surface temperatures within the image area, thus permitting simultaneous ground-truthing of the BLIMP data. The BLIMP was equipped with an uncooled InfraRed (IR) camera. The ZiviCat was designed to measure along predefined trajectories on a lake. Since LSWT spatial variability in each image is expected to be low, a poor estimation of the common spatial and temporal noise of the IR camera (nonuniformity and shutter-based drift, respectively) leads to errors in the thermal maps obtained. Nonuniformity was corrected by applying a pixelwise two-point linear correction method based on laboratory experiments. A Probability Density Function (PDF) matching in regions of overlap between sequential images was used for the drift correction. A feature matching-based algorithm, combining blob and region detectors, was implemented to create composite thermal images, and a mean value of the overlapped images at each location was considered as a representative value of that pixel in the final map. The results indicate that a high overlapping field of view (~95%) is essential for image fusion and noise reduction over such low-contrast scenes. The in situ temperatures measured by the ZiviCat were then used for the radiometric calibration. This resulted in the generation of LSWT maps at sub-pixel satellite scale resolution that revealed spatial LSWT variability, organized in narrow streaks hundreds of meters long and coherent patches of different size, with unprecedented detail.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Paheding Sidike, Vasit Sagan, Maitiniyazi Maimaitijiang, Matthew Maimaitiyiming, Nadia Shakoor, Joel Burken, Todd Mockler, Felix B. Fritschi〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Accurately mapping heterogeneous agricultural landscape is an important prerequisite for agricultural field management (e.g., weed control), plant phenotyping and yield prediction, as well as ecological characterization. Compared to traditional mapping practices that require intensive field surveys, remote sensing technologies offer efficient and cost-effective means for crop type mapping from regional to global scales. However, mapping heterogeneous agricultural landscape is a challenge because of diverse and complex spectral profiles of crops. We propose a novel deep learning method, namely deep progressively expanded network (dPEN), for mapping nineteen different objects including crop types, weeds and crop residues, in a heterogeneous agricultural field using WorldView-3 (WV-3) imagery. To assess the mapping accuracy of dPEN, we created a calibrated WV-3 dataset with the corresponding ground truth. In addition, the suitability of visible/near-infrared (VNIR, 400–1040 nm) and short-wave infrared (SWIR, 1195 nm–2365 nm) bands of WV-3 to classification accuracy were examined and discussed in detail. To the best of our knowledge, this is the first effort to explore the significance of all SWIR bands in WV-3 for classification accuracy in a heterogeneous agricultural landscape. The results demonstrated that: (1) The proposed dPEN allows for building a deeper neural network from multispectral data which was the limitation of many convolutional neural networks; (2) dPEN was able to extract more discriminative features from VNIR and SWIR bands by producing the highest overall accuracy (OA: 86.06%) over competing methods such as support vector machine and random forest; (3) The inclusion of WV-3 SWIR bands greatly improved the classification accuracy; (4) SWIR bands were particularly beneficial to improve the classification accuracy of some individual classes such as weeds, crop residues, and corn and soybean during late developmental stages; (5) The red-edge band (705–745 nm) was identified as the most important band affecting the classification accuracy nearly 10%, whereas the coastal band (400–450 nm) provided the lowest contribution; and (6) SWIR-5 band (2145–2185 nm) contributed most to OA by enhancing it approximately 4% when combined with VNIR bands, while SWIR-1 (1195–1225 nm) yielded the lowest improvement (1.55%) for OA. These research outcomes provide useful information for efficiently mapping agricultural landscape, and indicate the potential practices of dPEN and contributions of spectral bands in WV-3 for plant phenotyping, weed control, and crop residue retention.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Vaughn Smith, Carlos Portillo-Quintero, Arturo Sanchez-Azofeifa, Jose L. Hernandez-Stefanoni〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Tracking the occurrence of deforestation events is an essential task in tropical dry forest (TDF) conservation efforts. Ideally, deforestation monitoring systems would identify a TDF clearing with near real time precision and high spatial detail, and alert park managers and environmental practitioners of illegal forest clearings occurring anywhere in a region of interest. Over the past several years there have been significant advances in the design and application of continuous land cover change mapping algorithms with these capabilities, but no studies have implemented such methods over human dominated TDF environments where small–scale deforestation (〈5 ha) is widespread and hard to detect with moderate resolution sensors. The general objective for this research was to evaluate the overall accuracy of the BFASTSpatial R Package for detecting and monitoring small-scale deforestation in four sites located in tropical dry forest landscapes of Mexico and Costa Rica using greenness and moisture spectral indices derived from Landsat time series. Results show a high degree of spatial agreement (90%–94%) between the distribution of TDF clearings occurred during the 2013–2016 period (as indicated by VHR imagery interpretation) and BFASTSpatial outputs. NDMI and NBR2 had the best performance than other indices and this is evidenced by the combined overall, user's and producer's accuracies. In particular, NBR2 were the most accurate predictor of deforestation with an overall accuracy of 94.5%. Our results also imply that monitoring sites at an annual basis is feasible using BFASTSpatial and LTS, but that lower confidence should be given to sub-annual products given significant systematic temporal differences between the BFASTSpatial monthly product and reference data. The possibility of including more clear observations at the spatial resolution of Landsat (30-m) or higher will greatly increase the spatial and temporal accuracies of the method. Given its performance, BFASTSpatial can help monitor hotspots of small-scale TDF loss across Central and North America at little or no cost. Users of the method should have a strong knowledge of the local land use and land cover dynamics and the ecophysiology of vegetation types present in the landscape. This local expertise is necessary for interpreting and validating results as well as communicating its output to decision-makers and stakeholders.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Jianbo Qi, Donghui Xie, Tiangang Yin, Guangjian Yan, Jean-Philippe Gastellu-Etchegorry, Linyuan Li, Wuming Zhang, Xihan Mu, Leslie K. Norford〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Three-dimensional (3D) radiative transfer modeling of the transport and interaction of radiation through earth surfaces is challenging due to the complexity of the landscapes as well as the intensive computational cost of 3D radiative transfer simulations. To reduce computation time, current models work with schematic landscapes or with small-scale realistic scenes. The computer graphics community provides the most accurate and efficient models (known as renderers) but they were not designed specifically for performing scientific radiative transfer simulations. In this study, we propose LESS, a new 3D radiative transfer modeling framework. LESS employs a weighted forward photon tracing method to simulate multispectral bidirectional reflectance factor (BRF) or flux-related data (e.g., downwelling radiation) and a backward path tracing method to generate sensor images (e.g., fisheye images) or large-scale (e.g. 1 km〈sup〉2〈/sup〉) spectral images. The backward path tracing also has been extended to simulate thermal infrared radiation by using an on-the-fly computation of the sunlit and shaded scene components. This framework is achieved through the development of a user-friendly graphic user interface (GUI) and a set of tools to help construct the landscape and set parameters. The accuracy of LESS is evaluated with other models as well as field measurements in terms of directional BRFs and pixel-wise simulated image comparisons, which shows very good agreement. LESS has the potential in simulating datasets of realistically reconstructed landscapes. Such simulated datasets can be used as benchmarks for various applications in remote sensing, forestry investigation and photogrammetry.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Lian Feng, Xuejiao Hou, Yi Zheng〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Various aspects of the water qualities of the large lakes in the Yangtze Plain (YP) have been studied, while the efforts have been spent on one of the most important parameters -water transparency were only limited to several large lakes. Using in situ remote sensing reflectance and Secchi-disk depth (Z〈sub〉sd〈/sub〉) datasets, we assessed the performance of a semi-analytical model recently proposed by Lee et al. (2015) for remote sensing of the Z〈sub〉sd〈/sub〉 (Z〈sub〉sd, Lee〈/sub〉). The results show that a linear scaling correction over Z〈sub〉sd, Lee〈/sub〉 (Z′〈sub〉sd, Lee〈/sub〉) could lead to improved agreement between remote sensing estimates and field measurements (root mean square error 〈 35%) in the study region. The Z′〈sub〉sd, Lee〈/sub〉 scheme was then applied to MODIS/Aqua observations between 2003 and 2016 to obtain the spatial and temporal dynamics of water transparency in 50 large lakes in the YP. The long-term mean Z〈sub〉sd〈/sub〉 of the entire region was 0.39 ± 1.17 m during the observation period, with high and low values occurring in warm and cold seasons, respectively. Of the 50 examined lakes, half demonstrated decreasing or increasing Z〈sub〉sd〈/sub〉 trends, and the number of lakes exhibiting significantly decreasing trends also comparable to the number exhibiting increasing trends. The relative contributions of the seven potential driving factors (from both human activities and natural processes) to the interannual changes in water clarity were quantified for each lake using a multiple general linear model regression analysis. The responses of Z〈sub〉sd〈/sub〉 to these drivers showed considerable differences among lakes, and human activities demonstrated significant roles in more lakes than those affected by natural variability, accounting for 50% (25/50) and 20% (10/50) of the lakes, respectively. This study provides the first comprehensive basin-scale estimate of the water transparency in the lakes in the YP, and the Z〈sub〉sd〈/sub〉 results and analysis of driving forces can provide important information for local water quality conservation and restoration.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 March 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 222〈/p〉 〈p〉Author(s): Tihomir S. Kostadinov, Rina Schumer, Mark Hausner, Kat J. Bormann, Rowan Gaffney, Kenneth McGwire, Thomas H. Painter, Scott Tyler, Adrian A. Harpold〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The distribution of snow cover is critical for predicting ecohydrological processes and underpins mountain water supplies in ranges like the Sierra Nevada in the Western United States. Many key water supply areas are covered by montane forests, which have substantial effects on the amount and timing of snowmelt. In-situ observations of snow-forest interactions have limited spatial coverage and remote sensing using optical sensors (e.g. MODIS) cannot observe snow cover below the canopy. In this study, we developed and verified a lidar-based method to detect snow cover under canopy, investigated how fractional snow covered area (fSCA) varies with topography in open versus under canopy areas and developed a correction factor that could be used to improve satellite-derived fSCA products. We developed our new method using three snow-on lidar overflights and verified it with in-situ distributed temperature sensor (DTS) observations at Sagehen Creek watershed in the Sierra Nevada, California, USA. DTS validation of lidar classifications showed excellent agreement at 85–96%, including high agreement and large number of returns in under canopy locations. The lidar-derived fSCA observations generally showed earlier snow disappearance under the canopy than in open positions, which is consistent with relatively warm temperatures and greater longwave radiation. However, in contrast to expectations, areas with high solar exposure (i.e. high southwestness) exhibited higher fSCA under the canopy. Results indicated that the 〈em〉k〈/em〉 factor (the ratio of under canopy fSCA to open fSCA) varied systematically with southwestness and elevation. Using this factor to correct the study domain fSCA indicated that the typical assumption that 〈em〉k〈/em〉 = 1 could lead to an up to ~0.05 bias (in fSCA units) towards overestimation. However, within 10 and 100-m individual pixels the fSCA overprediction bias can be 25–30% for higher fSCA values. Although uncertainty would be reduced using higher snow-on lidar point densities, our method shows promise to improve the typical assumption that snow disappearance is identical in under the canopy and in the open (〈em〉k〈/em〉 = 1). Future applications of our lidar-based method at different sites with varying climate, topography and vegetation structure has the dual potential to expand understanding of snow-forest interactions in complex terrain and improve operational fSCA products.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 March 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 222〈/p〉 〈p〉Author(s): E. Roteta, A. Bastarrika, M. Padilla, T. Storm, E. Chuvieco〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉A locally-adapted multitemporal two-phase burned area (BA) algorithm has been developed using as inputs Sentinel-2 MSI reflectance measurements in the short and near infrared wavebands plus the active fires detected by Terra and Aqua MODIS sensors. An initial burned area map is created in the first step, from which tile dependent statistics are extracted for the second step. The whole Sub-Saharan Africa (around 25 M km〈sup〉2〈/sup〉) was processed with this algorithm at a spatial resolution of 20 m, from January to December 2016. This period covers two half fire seasons on the Northern Hemisphere and an entire fire season in the South. The area was selected as existing BA products account it to include around 70% of global BA. Validation of this product was based on a two-stage stratified random sampling of Landsat multitemporal images. Higher accuracy values than existing global BA products were observed, with Dice coefficient of 77% and omission and commission errors of 26.5% and 19.3% respectively. The standard NASA BA product (MCD64A1 c6) showed a similar commission error (20.4%), but much higher omission errors (59.6%), with a lower Dice coefficient (53.6%). The BA algorithm was processed over 〉11,000 Sentinel-2 images to create a database that would also include small fires (〈100 ha). This is the first time a continental BA product is generated from medium resolution sensors (spatial resolution = 20 m), showing their operational potential for improving our current understanding of global fire impacts. Total BA estimated from our product was 4.9 M km〈sup〉2〈/sup〉, around 80% larger area than what the NASA BA product (MCD64A1 c6) detected in the same period (2.7 M km〈sup〉2〈/sup〉). The main differences between the two products were found in regions where small fires (〈100 ha) account for a significant proportion of total BA, as global products based on coarse pixel sizes (500 m for MCD64A1) unlikely detect them. On the negative side, Sentinel-2 based products have lower temporal resolution and consequently are more affected by cloud/cloud shadows and have less temporal reporting accuracy than global BA products. The product derived from S2 imagery would greatly contribute to better understanding the impacts of small fires in global fire regimes, particularly in tropical regions, where such fires are frequent. This product is named FireCCISFD11 and it is publicly available at: 〈a href="https://www.esa-fire-cci.org/node/262" target="_blank"〉https://www.esa-fire-cci.org/node/262〈/a〉, last accessed on November 2018.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Wenyu Gong, Margaret M. Darrow, Franz J. Meyer, Ronald P. Daanen〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Frozen debris lobes (FDLs) are slow-moving landslides along permafrost-affected slopes, and consist of soil, rock, organic debris, and areas of massive infiltration ice. Based on their proximity to the adjacent infrastructure, their size, and their flow dynamics, FDLs represent potential geohazards. Eight FDLs (FDL-A, -B, -C, -D, -4, -5, -7, -11) within the Dalton Highway corridor in the Brooks Range, Alaska, USA are the subject of this paper. We examined temporal and spatial variation of FDL displacement using medium-resolution Synthetic Aperture Radar (SAR) images and differential SAR Interferometry (dInSAR) techniques. European Remote Sensing satellite 1/2 (ERS 1/2) and Phased Array type L-band Synthetic Aperture Radar (PALSAR) acquisitions were used to generate coherent interferograms to study the displacement history of FDLs for 1995–1996 and 2006–2010. We conducted an initial assessment of the capability of satellite InSAR to monitor FDL movement depending on data resolution, season, and vegetation coverage, which also helped us to select useful interferograms. With multi-temporal interferometric displacement maps, we found that seven investigated FDLs (FDL-7 was excluded from this sub-experiment due to limited data coverage) demonstrated strong spatial and seasonal variations in their movement patterns, with maximum displacement rates typically occurring in October and minimum displacement rates during February or March, which is consistent with previously published field study results. Overall, through this study we: (1) delineated the active FDLs during the winter period using a wrapped PALSAR interferogram; (2) analyzed the spatial variation of the deformation field within each FDL body; (3) modeled the seasonal changes of FDL deformation rates through the analysis of multi-temporal ERS tandem interferograms; and (4) integrated InSAR-derived deformation rates with those obtained through historical imagery analysis to determine long-term deformation rates. Results from this study fill the gaps left in the historical imagery analysis and provide important seasonal and spatial deformation data, which are essential in the development of a mitigation plan as these features approach infrastructure. We also summarize our research experience studying these moving features using satellite radar interferometry and believe this can be useful for future studies of similar features.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Zifeng Hu, Yiquan Qi, Xianqiang He, Yu-Huai Wang, Dong-Ping Wang, Xuhua Cheng, Xiaohui Liu, Tao Wang〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The Taiwan Strait is the only channel connecting the South China Sea (SCS) and the East China Sea (ECS), and plays a key role in the material exchanges between SCS and ECS. The region however is poorly sampled during the northeast (NE) monsoon when severe weather hinders shipboard measurements. In this study, the surface circulations in the Taiwan Strait and the adjacent East China Sea are derived from the Geostationary Ocean Color Imager (GOCI) with a high spatiotemporal resolution. The satellite observations are limited by cloud over. Nevertheless, they provide the only surface flow observations useful in hypothesis testing. It is demonstrated that the transient surface currents can be classified into three regimes: a southwestward flow for strong winds, a northeastward flow for weak winds, and a cross-strait flow for moderate winds. The satellite observations are also valuable in validating model currents. Two high-resolution global ocean reanalysis products, the Hybrid Coordinate Ocean Model (HYCOM) and the Copernicus Marine Environment Monitoring Service (CMEMS), are examined. The reanalysis products are generally consistent with the satellite observations. The CMEMS moreover shows a superior performance. This study illustrates the unique capability of the geostationary ocean color satellite in providing spatially extensive surface flow observations in broad shelf regions where current meter data are sparse or non-existent.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: Available online 17 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment〈/p〉 〈p〉Author(s): Lan H. Nguyen, Deepak R. Joshi, David E. Clay, Geoffrey M. Henebry〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Over the last 20 years, substantial amounts of grassland have been converted to other land uses in the Northern Great Plains. Most of land cover/land use (LCLU) assessments in this region have been based on the U.S. Department of Agriculture - Cropland Data Layer (USDA - CDL), which may be inconsistent. Here, we demonstrate an approach to map land cover utilizing multi-temporal Earth Observation data from Landsat and MODIS. We first built an annual time series of accumulated growing degree-days (AGDD) from MODIS 8-day composites of land surface temperatures. Using the Enhanced Vegetation Index (EVI) derived from Landsat Collection 1's surface reflectance, we then fit at each pixel a downward convex quadratic model to each year's progression of AGDD (〈em〉i.e.〈/em〉, EVI = α + β × AGDD − γ × AGDD〈sup〉2〈/sup〉). Phenological metrics derived from fitted model and the goodness of fit then are submitted to a random forest classifier (RFC) to characterize LCLU for four sample counties in South Dakota in three years (2006, 2012, 2014) when reference point datasets are available for training and validation. To examine the sensitivity of the RFC to sample size and design, we performed classifications under different sample selection scenarios. The results indicate that our proposed method accurately mapped major crops in the study area but showed limited accuracy for non-vegetated land covers. Although all RFC models exhibit high accuracy, estimated land cover areas from alternative models could vary widely, suggesting the need for a careful examination of model stability in any future land cover supervised classification study. Among all sampling designs, the “same distribution” models (proportional distribution of the sample is like proportional distribution of the population) tend to yield best land cover prediction. RFC used only the most eight important variables (〈em〉e.g.〈/em〉, three fitted parameter coefficients [α, β, and γ]; maximum modeled EVI; AGDD at maximum modeled EVI; the number of observations used to fit CxQ model; and the number of valid observations) have slightly higher accuracy compared to those using all variables. By summarizing annual image time series through land surface phenology modeling, LCLU classification can embrace both seasonality and interannual variability, thereby increasing the accuracy of LCLU change detection.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Dirk Pflugmacher, Andreas Rabe, Mathias Peters, Patrick Hostert〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉This study analyzed, for the first time, the potential of combining the large European-wide land survey LUCAS (Land Use/Cover Area frame Survey) and Landsat-8 data for mapping pan-European land cover and land use. We used annual and seasonal spectral-temporal metrics and environmental features to map 12 land cover and land use classes across Europe. The spectral-temporal metrics provided an efficient means to capture seasonal variations of land surface spectra and to reduce the impact of clouds and cloud-shadows by relaxing the otherwise strong cloud cover limitations imposed by image-based classification methods. The best classification model was based on Landsat-8 data from three years (2014–2016) and achieved an accuracy of 75.1%, nearly 2 percentage points higher than the classification model based on a single year of Landsat data (2015). Our results indicate that annual pan-European land cover maps are feasible, but that temporally dynamic classes like artificial land, cropland, and grassland still benefit from more frequent satellite observations. The produced pan-European land cover map compared favorably to the existing CORINE (Coordination of Information on the Environment) 2012 land cover dataset. The mapped country-wide area proportions strongly correlated with LUCAS-estimated area proportions (〈em〉r〈/em〉 = 0.98). Differences between mapped and LUCAS sample-based area estimates were highest for broadleaved forest (map area was 9% higher). Grassland and seasonal cropland areas were 7% higher than the LUCAS estimate, respectively. In comparison, the correlation between LUCAS and CORINE area proportions was weaker (〈em〉r〈/em〉 = 0.84) and varied strongly by country. CORINE substantially overestimated seasonal croplands by 63% and underestimated grassland proportions by 37%. Our study shows that combining current state-of-the-art remote sensing methods with the large LUCAS database improves pan-European land cover mapping. Although this study focuses on European land cover, the unique combination of large survey data and machine learning of spectral-temporal metrics, may also serve as a reference case for other regions. The pan-European land cover map for 2015 developed in this study is available under 〈a href="https://doi.pangaea.de/10.1594/PANGAEA.896282" target="_blank"〉https://doi.pangaea.de/10.1594/PANGAEA.896282〈/a〉.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): M.G. Hethcoat, D.P. Edwards, J.M.B. Carreiras, R.G. Bryant, F.M. França, S. Quegan〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Hundreds of millions of hectares of tropical forest have been selectively logged, either legally or illegally. Methods for detecting and monitoring tropical selective logging using satellite data are at an early stage, with current methods only able to detect more intensive timber harvest (〉20 m〈sup〉3〈/sup〉 ha〈sup〉−1〈/sup〉). The spatial resolution of widely available datasets, like Landsat, have previously been considered too coarse to measure the subtle changes in forests associated with less intensive selective logging, yet most present-day logging is at low intensity. We utilized a detailed selective logging dataset from over 11,000 ha of forest in Rondônia, southern Brazilian Amazon, to develop a Random Forest machine-learning algorithm for detecting low-intensity selective logging (〈15 m〈sup〉3〈/sup〉 ha〈sup〉−1〈/sup〉). We show that Landsat imagery acquired before the cessation of logging activities (i.e. the final cloud-free image of the dry season during logging) was better at detecting selective logging than imagery acquired at the start of the following dry season (i.e. the first cloud-free image of the next dry season). Within our study area the detection rate of logged pixels was approximately 90% (with roughly 20% commission and 8% omission error rates) and approximately 40% of the area inside low-intensity selective logging tracts were labelled as logged. Application of the algorithm to 6152 ha of selectively logged forest at a second site in Pará, northeast Brazilian Amazon, resulted in the detection of 2316 ha (38%) of selective logging (with 20% commission and 7% omission error rates). This suggests that our method can detect low-intensity selective logging across large areas of the Amazon. It is thus an important step forward in developing systems for detecting selective logging pan-tropically with freely available data sets, and has key implications for monitoring logging and implementing carbon-based payments for ecosystem service schemes.〈/p〉〈/div〉 〈/div〉 〈div xml:lang="en"〉 〈h5〉Graphical abstract〈/h5〉 〈div〉〈p〉〈figure〉〈img src="https://ars.els-cdn.com/content/image/1-s2.0-S0034425718305534-ga1.jpg" width="301" alt="Unlabelled Image" title="Unlabelled Image"〉〈/figure〉〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: Available online 5 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment〈/p〉 〈p〉Author(s): Xinxin Wang, Xiangming Xiao, Zhenhua Zou, Bangqian Chen, Jun Ma, Jinwei Dong, Russell B. Doughty, Qiaoyan Zhong, Yuanwei Qin, Shengqi Dai, Xiangping Li, Bin Zhao, Bo Li〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Tidal flats (non-vegetated area), along with coastal vegetation area, constitute the coastal wetlands (intertidal zone) between high and low water lines, and play an important role in wildlife, biodiversity and biogeochemical cycles. However, accurate annual maps of coastal tidal flats over the last few decades are unavailable and their spatio-temporal changes in China are unknown. In this study, we analyzed all the available Landsat TM/ETM+/OLI imagery (~44,528 images) using the Google Earth Engine (GEE) cloud computing platform and a robust decision tree algorithm to generate annual frequency maps of open surface water body and vegetation to produce annual maps of coastal tidal flats in eastern China from 1986 to 2016 at 30-m spatial resolution. The resulting map of coastal tidal flats in 2016 was evaluated using very high-resolution images available in Google Earth. The total area of coastal tidal flats in China in 2016 was about 731,170 ha, mostly distributed in the provinces around Yellow River Delta and Pearl River Delta. The interannual dynamics of coastal tidal flats area in China over the last three decades can be divided into three periods: a stable period during 1986–1992, an increasing period during 1993–2001 and a decreasing period during 2002–2016. The resulting annual coastal tidal flats maps could be used to support sustainable coastal zone management policies that preserve coastal ecosystem services and biodiversity in China.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Tongren Xu, Xinlei He, Sayed M. Bateni, Thomas Auligne, Shaomin Liu, Ziwei Xu, Ji Zhou, Kebiao Mao〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Estimation of turbulent heat fluxes by assimilating sequences of land surface temperature (LST) measurements into variational data assimilation (VDA) frameworks has been the subject of several studies. The VDA approaches estimate turbulent heat fluxes by minimizing the difference between LST observations and estimations from the heat diffusion equation. The VDA methods have been tested only with high temporal resolution LST observations (e.g., from geostationary satellites) when applied at regional scales. Geostationary satellites can capture the diurnal cycle of LST, but they have a relatively low spatial resolution and mainly focus on low latitudes. To overcome these shortcomings, this study assimilates high spatial resolution LST data from polar orbiting satellites (e.g., Moderate Resolution Imaging Spectroradiometer, MODIS) into the combined-source (CS) and dual-source (DS) VDA schemes. An expression is developed to obtain an a priori evaporative fraction (EF) estimate from leaf area index (LAI) or apparent thermal inertia (ATI). The a priori EF estimate is used as an initial guess in the VDA approach. The results indicate that the VDA method is able to find the optimal value of EF by assimilating the low-temporal resolution MODIS LST data. The predicted turbulent heat fluxes from VDA are compared with the measurements from the large-aperture scintillometer at three sites (Arou, Daman, and Sidaoqiao) in the Heihe River Basin (located in northwest China). The findings indicate that the CS and DS VDA models perform well in various hydrological and vegetative conditions. The three-site-average root mean square errors (RMSEs) of sensible and latent heat fluxes estimates from the CS scheme are 37.44 W m〈sup〉−2〈/sup〉 and 94.30 W m〈sup〉−2〈/sup〉, respectively. The DS model reduces the abovementioned RMSEs by 19.82% and 21.37%, respectively. Overall, the results show that using the a priori EF estimate from the proposed expression in the VDA approach eliminates the need for the high resolution LST data from geostationary satellites, and allows the VDA method to estimate turbulent heat fluxes by assimilating LST data from polar orbiting satellites. Finally, several numerical tests are conducted to assess the effect of LST temporal sampling on the turbulent heat fluxes estimates. The results show that the LST measurement at 1400 Local Time (LT) has the most amount of information for partitioning the available energy into sensible and latent heat fluxes.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Liheng Zhong, Lina Hu, Hang Zhou〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉This study aims to develop a deep learning based classification framework for remotely sensed time series. The experiment was carried out in Yolo County, California, which has a very diverse irrigated agricultural system dominated by economic crops. For the challenging task of classifying summer crops using Landsat Enhanced Vegetation Index (EVI) time series, two types of deep learning models were designed: one is based on Long Short-Term Memory (LSTM), and the other is based on one-dimensional convolutional (Conv1D) layers. Three widely-used classifiers were also tested for comparison, including a gradient boosting machine called XGBoost, Random Forest, and Support Vector Machine. Although LSTM is widely used for sequential data representation, in this study its accuracy (82.41%) and F1 score (0.67) were the lowest among all the classifiers. Among non-deep-learning classifiers, XGBoost achieved the best result with 84.17% accuracy and an F1 score of 0.69. The highest accuracy (85.54%) and F1 score (0.73) were achieved by the Conv1D-based model, which mainly consists of a stack of Conv1D layers and an inception module. The behavior of the Conv1D-based model was inspected by visualizing the activation on different layers. The model employs EVI time series by examining shapes at various scales in a hierarchical manner. Lower Conv1D layers of the optimized model capture small scale temporal variations, while upper layers focus on overall seasonal patterns. Conv1D layers were used as an embedded multi-level feature extractor in the classification model which automatically extracts features from input time series during training. The automated feature extraction reduces the dependency on manual feature engineering and pre-defined equations of crop growing cycles. This study shows that the Conv1D-based deep learning framework provides an effective and efficient method of time series representation in multi-temporal classification tasks.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): George Azzari, Patricio Grassini, Juan Ignacio Rattalino Edreira, Shawn Conley, Spyridon Mourtzinis, David B. Lobell〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Low-intensity tillage has become more popular among farmers in the United States and many other regions. However, accurate data on when and where low-intensity tillage methods are being used remain scarce, and this scarcity impedes understanding of the factors affecting the adoption and the agronomic or environmental impacts of these practices. In this study, we used composites of satellite imagery from Landsat 5, 7, and 8, and Sentinel-1 in combination with producer data from about 5900 georeferenced fields to train a random forest classifier and generate annual large-scale maps of tillage intensity from 2005 to 2016. We tested different combinations of hyper-parameters using cross-validation, splitting the training and testing data alternatively by field, year, and state to assess the influence of clustering on validation results and evaluate the generalizability of the classification model. We found that the best model was able to map tillage practices across the entire North Central US region at 30 m-resolution with accuracies spanning between 75% and 79%, depending on the validation approach. We also found that although Sentinel-1 provides an independent measure that should be sensitive to surface moisture and roughness, it currently adds relatively little to classification performance beyond what is possible with Landsat. When aggregated to the state level, the satellite estimates of percentage low- and high-intensity tillage agreed well with a USDA survey on tillage practices in 2006 (R〈sup〉2〈/sup〉 = 0.55). The satellite data also revealed clear increases in low-intensity tillage area for most counties in the past decade. Overall, the ability to accurately map spatial and temporal patterns in tillage should facilitate further study of this important practice in the United States, as well as other regions with fewer survey-based estimates.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Jiwei Li, Qian Yu, Yong Q. Tian, Brian L. Becker, Paul Siqueira, Nathan Torbick〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Bottom reflectance is often the main cause of high uncertainty in Colored Dissolved Organic Matter (CDOM) estimation for optically shallow waters. This study presents a Landsat-8 based Shallow Water Bio-optical Properties (SBOP) algorithm to overcome bottom effects so as to successfully observe spatial and temporal CDOM dynamics in inland waters. We evaluated the algorithm via 58 images and a large set of field measurements collected across seasons of multiple years in the Saginaw Bay, Lake Huron. Results showed that the SBOP algorithm reduced estimation errors by as much as 4 times (RMSE = 0.17 and 〈em〉R〈/em〉〈sup〉〈em〉2〈/em〉〈/sup〉 = 0.87 in the Saginaw Bay) when compared to the QAA-CDOM algorithm that did not take into account bottom reflectance. These improvements in CDOM estimation are consistent and robust across broad range CDOM absorption. Our analysis revealed: 1) the proposed remote sensing algorithm resulted in significant improvements in tracing spatial-temporal CDOM inputs from terrestrial environments to lakes, 2) CDOM distribution captured with high resolution land-viewing satellite is useful in revealing the impacts of terrestrial ecosystems on the aquatic environment, and 3) Landsat-8 OLI, with its 16 days revisit time, provides valuable time series data for studying CDOM seasonal variations at land-water interface and has the potential to reveal its relationship to adjacent terrestrial biogeography and hydrology. The study presents a shallow water algorithm for studying freshwater or coastal ecology, as well as carbon cycling science.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Fulvia Baratelli, Nicolas Flipo, Agnès Rivière, Sylvain Biancamaria〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The quantification of aquifer contribution to river discharge is of primary importance to evaluate the impact of climatic and anthropogenic stresses on the availability of water resources. Several baseflow estimation methods require river discharge measurements, which can be difficult to obtain at high spatio-temporal resolution for large basins. The future Surface Water and Ocean Topography (SWOT) satellite mission will provide discharge estimations for large rivers (〉50–100 m wide) even in ungauged basins. The frequency of these estimations depends mainly on latitude and ranges from zero to more than ten values in the 21-day satellite cycle. This work aims at answering the following question: can baseflow be estimated from SWOT observations during the mission lifetime? An algorithm based on hydrograph separation by Chapman's filter was developed to automatically estimate the baseflow in a river network at regional scale (〉10 000 km〈sup〉2〈/sup〉). The algorithm was applied to the Seine river basin (75 000 km〈sup〉2〈/sup〉, France) using the discharge time series simulated at daily time step by a coupled hydrological-hydrogeological model to obtain the reference baseflow estimations. The same algorithm is then forced with discharge time series sampled at SWOT observation frequency. The average baseflow is estimated with good accuracy for all the reaches which are observed at least once per cycle (relative bias less than 8%). The time evolution of baseflow is also rather well retrieved, with a Nash-Sutcliffe coefficient above 0.7 for 96% of the network length. An analysis of the effect of SWOT discharge uncertainties on baseflow estimation shows that bias is the component of discharge error that most contributes to the error on baseflow. Anyway, when the combined effect of SWOT discharge sampling and SWOT discharge uncertainties is considered, the error on baseflow estimates is slightly smaller than that on discharge. This work provides new potential for the SWOT mission in terms of global hydrological analysis and water cycle closure.〈/p〉〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Sarah J. Graves, T. Trevor Caughlin, Gregory P. Asner, Stephanie A. Bohlman〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Agricultural land now exceeds forests as the dominant global biome. Because of their global dominance, and potential expansion or loss, methods to estimate biomass and carbon in agricultural areas are necessary for monitoring global terrestrial carbon stocks and predicting carbon dynamics. Agricultural areas in the tropics have substantial tree cover and associated above ground biomass (AGB) and carbon. Active remote sensing data, such as airborne LiDAR (light detection and ranging), can provide accurate estimates of biomass stocks, but common plot-based methods may not be suitable for agricultural areas with dispersed and heterogeneous tree cover. The objectives of this research are to quantify AGB of a tropical agricultural landscape using a tree-based method that directly incorporates the size of individual trees, and to understand how landscape estimates of AGB from a tree-based method compare to estimates from a plot-based method. We use high-resolution (1.12 m) airborne LiDAR data collected on a 9280-ha region of the Azuero Peninsula of Panama. We model individual tree AGB with canopy dimensions from the LiDAR data. We apply the model to individual tree crown polygons and aggregate AGB estimates to compare with previously developed plot-based estimates. We find that agricultural trees are a distinct and dominant part of our study site. The tree-based approach estimates greater AGB in pixels with low forest cover than the plot-based approach, resulting a 2-fold difference in landscape AGB estimates between the methods for non-forested areas. Additionally, one third of the total landscape AGB exists in areas having 〈10% cover, based on a global tree cover product. Our study supports the continued use and development of allometric models to predict individual tree biomass from LiDAR-derived canopy dimensions and demonstrates the potential for spatial information from high-resolution data, such as relative isolation of canopies, to improve allometric models.〈/p〉〈/div〉 〈/div〉 〈div xml:lang="en"〉 〈h5〉Graphical abstract〈/h5〉 〈div〉〈p〉〈figure〉〈img src="https://ars.els-cdn.com/content/image/1-s2.0-S0034425718304206-ga1.jpg" width="500" alt="Unlabelled Image" title="Unlabelled Image"〉〈/figure〉〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Chaoying Zhao, Chuanjin Liu, Qin Zhang, Zhong Lu, Chengsheng Yang〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The Linfen-Yuncheng Basin (LYB) in China is a region possessing severe geo-hazards, including active tectonic fault movement, land subsidence and ground fissures among others. Interferometric Synthetic Aperture Radar (InSAR) technique is applied to map surface deformation associated with various geo-hazards in this basin. The poly-interferogram rate and time-series estimator algorithm (Π-RATE) is used over forty-nine scenes of SAR data to generate the deformation maps over the entire LYB. The precision of InSAR results is around 3 mm/yr. Some active faults and ground fissures are successfully detected. The spatiotemporal characteristics of tableland uplift, faults displacement and basin subsidence are quantitatively monitored with InSAR technique ranging from 2 mm/yr to 142 mm/yr. Finally, the mechanisms of surface deformation regarding large scale Zhongtiaoshan fault, middle scale basin land subsidence and small scale ground fissures are discussed in terms of interseismic movement, underground water level changes and hydrostratigraphic heterogeneity.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Richard Massey, Temuulen T. Sankey, Kamini Yadav, Russell G. Congalton, James C. Tilton〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Accurate information on cropland spatial distribution is required for global-scale assessments and agricultural land use policies. Cloud computing platforms such as Google Earth Engine (GEE) provide unprecedented opportunities for large-scale classifications of Landsat data. We developed a novel method to fuse pixel-based random forest classification of continental-scale Landsat data on GEE and an object-based segmentation approach known as recursive hierarchical segmentation (RHSeg). Using our fusion method, we produced a continental-scale cropland extent map for North America at 30 m spatial resolution for the nominal year 2010. The total cropland area for North America was estimated at 275.18 million hectares (Mha). The overall accuracies of the map are 〉90% across the continent. This map also compares well with the United States Department of Agriculture (USDA) cropland data layer (CDL), Agriculture and Agri-food Canada (AAFC) annual crop inventory (ACI), and the Mexican government agency Servicio de Información Agroalimentaria y Pesquera (SIAP)'s agricultural boundaries. Furthermore, our map compared well with sub-country statistics including state-wise and county-wise cropland statistics in regression models resulting in 〈em〉R〈/em〉〈sup〉〈em〉2〈/em〉〈/sup〉 〉 0.84. This key contribution paves the way for more detailed products such as crop intensity, crop type, and crop irrigation, and provides a method for creating high-resolution cropland extent maps for other countries where spatial information about croplands are not as prevalent.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): S.M. Punalekar, A. Verhoef, T.L. Quaife, D. Humphries, L. Bermingham, C.K. Reynolds〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉A large proportion of the global land surface is covered by pasture. The advent of the Sentinel satellites program provides free datasets with good spatiotemporal resolution that can be a valuable source of information for monitoring pasture resources. We combined optical remote sensing data (proximal hyperspectral and Sentinel 2A) with a radiative transfer model (PROSAIL) to estimate leaf Area Index (〈em〉LAI〈/em〉), and biomass, in a dairy farming context. Three sites in Southern England were used: two pasture farms that differed in pasture type and management, and a set of small agronomy trial plots with different mixtures of grasses, legumes and herbs, as well as pure perennial ryegrass. The proximal and satellite spectral data were used to retrieve 〈em〉LAI via〈/em〉 PROSAIL model inversion, which were compared against field observations of 〈em〉LAI〈/em〉. The potential of bands of Sentinel 2A that corresponded with a 10 m resolution was studied by convolving narrow spectral bands (from a handheld hyperspectral sensor) into Sentinel 2A bands (10 m). Retrieved 〈em〉LAI〈/em〉, using these spectrally resampled S2A data, compared well with measured 〈em〉LAI〈/em〉, for all sites, even for those with mixed species cover (although retrieved 〈em〉LAI〈/em〉 was somewhat overestimated for pasture mixtures with high 〈em〉LAI〈/em〉). This proved the suitability of 10 m Sentinel 2A spectral bands for capturing 〈em〉LAI〈/em〉 dynamics for different types of pastures. We also found that inclusion of 20 m bands in the inversion scheme did not lead to any further improvement in retrieved 〈em〉LAI〈/em〉. Sentinel 2A image based retrieval yielded good agreement with 〈em〉LAI〈/em〉 measurements obtained for a typical perennial ryegrass based pasture farm. 〈em〉LAI〈/em〉 retrieved in this way was used to create biomass maps (that correspond to indirect biomass measurements by Rising Plate Meter (RPM)), for mixed-species paddocks for a farm for which limited field data were available. These maps compared moderately well with farmer-collected RPM measurements for this farm. We propose that estimates of paddock-averaged and within-paddock variability of biomass are more reliably obtained from a combined Sentinel 2A-PROSAIL approach, rather than by manual RPM measurements. The physically based radiative transfer model inversion approach outperformed the Normalized Difference Vegetation Index based retrieval method, and does not require site specific calibrations of the inversion scheme.〈/p〉〈/div〉 〈/div〉 〈div xml:lang="en"〉 〈h5〉Graphical abstract〈/h5〉 〈div〉〈p〉〈figure〉〈img src="https://ars.els-cdn.com/content/image/1-s2.0-S0034425718304486-ga1.jpg" width="448" alt="Unlabelled Image" title="Unlabelled Image"〉〈/figure〉〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Ryan J. Fisher, Ben Sawa, Beatriz Prieto〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Native grassland in North America is considered one of the most imperiled and altered ecosystems. Unfortunately, an assessment of how much native grassland remains in North America is difficult because all nation-wide landcover mapping products do not reliably distinguish native grasslands from grasslands that have been deliberately planted with tame grasses and forbs (i.e., tame grasslands). We established a 218.5 km〈sup〉2〈/sup〉 study area in southwestern Saskatchewan, Canada to evaluate the use of high-resolution Light Detection and Ranging (LiDAR) for classification of native (i.e., fields with native-dominant species mixes or fields that were formerly native species dominated but have been invaded by exotic species) and tame grasslands (i.e., grasslands dominated by exotic grass and forb species), and compared these classifications to the best-available landcover mapping product that is currently available for this area. We used the presence of tractor furrows, identified from the LiDAR digital terrain hillshade product, in tame-dominated grasslands to distinguish them from native-dominated grasslands that had an absence of tractor furrows. The LiDAR method achieved substantially better classification success (Cohen's Kappa = 0.57) at distinguishing native-dominated (〈em〉N〈/em〉 = 82) from tame-dominated grasslands (〈em〉N〈/em〉 = 45), than the currently available landcover product (Cohen's Kappa = 0.13) over our 218.5 km〈sup〉2〈/sup〉 study area. Misclassification by LiDAR of fields that had been planted with tame grasses and forbs, but were starting to be re-established by native plants appeared to be one weakness of the method in the study area. Our research highlights a novel and time-efficient method for classifying LiDAR imagery using easily available image analysis features in ArcGIS.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Aparna R. Phalke, Mutlu Özdoğan〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Accurate and up-to-date cropland maps play an important role in the study of food security. Traditional mapping of croplands using medium resolution (10–100 m) remote sensing imagery involving a “one-time, one-place” approach requires significant computing and labor resources. Although high mapping accuracies can be achieved using this approach, it is tedious and expensive to collect reference information to train the classifiers at each location and to apply over large areas, such as a continent. Moreover, large area cropland mapping presents additional challenges including a wide range of agricultural management practices, climatic conditions, and crop types. To overcome these challenges, here we report on a generalized image classifier to map cropland extent, which builds a classification model using training data from one location and time period, applied to other times and locations without the need for additional training data. The study was demonstrated across eight agro-ecological zones (AEZs) in Europe, the Middle East and North Africa using Landsat data acquired between 2009 and 2011. To reduce between-scene variability associated with image availability and cloud cover, input data were reduced to salient temporal statistics derived from enhanced vegetation index (EVI) combined with topographic variables. The generalized classifier was then tested across three levels of generalization: 1. 〈u〉〈em〉individual〈/em〉〈/u〉 - where training data were extracted from and applied to the same Landsat footprint; 2. 〈u〉〈em〉AEZ〈/em〉〈/u〉 where training data were extracted from a set of Landsat footprints within an AEZ and applied to any other Landsat footprint in the same AEZ; and 3. 〈u〉〈em〉regional〈/em〉〈/u〉 where training data were extracted from a set of Landsat footprints in the whole study area and applied to any other Landsat footprint inside the study area. Results showed that the generalized classifier is successful in identifying and mapping croplands with comparable success across all three levels of generalization with minimal cost: average loss in accuracy (as measured by overall accuracy) from the 〈em〉individual〈/em〉 level (average overall accuracy of 80 ± 5%) to 〈em〉regional〈/em〉 level (average overall accuracy of 74 ± 10%) is between 2 and 10% depending on the location. Results also show that generalization is not sensitive to the choice of the classification algorithm – the Linear Discriminant Analysis (LDA) model performs equally well compared to many popular machine learning algorithms found in the literature. This work suggests the generalization/signature extension framework has a great potential for rapid identification and mapping of croplands with reasonable accuracies over large areas using only easily computed vegetation indices with very little user input and ground information requirement.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Xiao Xiang Zhu, Gerald Baier, Marie Lachaise, Yilei Shi, Fathalrahman Adam, Richard Bamler〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The primary objective of the German TanDEM-X mission is the generation of a globally available, highly accurate and detailed digital elevation model (DEM), with the final product having 12 m posting, 2 m relative and 10 m absolute vertical accuracy. The first version of this global DEM has been finalized by the German Aerospace Center (DLR), in September 2016. Our experience with the experimental application of non-local means filters to TanDEM-X data suggests that TanDEM-X has the potential of producing DEMs of even higher resolution and accuracy. The goal of this investigation is to explore the possibility of employing non-local InSAR filters to achieve an effective resolution of 6 m, with an equivalent posting, and a relative height error below 0.8 m, i.e. an increase of quality by a factor of 2 × 2 in resolution and a factor of 2 m/0.8 m = 2.5 in height accuracy — all in all one order of magnitude.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Qi Huang, Di Long, Mingda Du, Chao Zeng, Gang Qiao, Xingdong Li, Aizhong Hou, Yang Hong〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉River discharge is an important variable in the water cycle that is related to water supply, irrigation, and flood forecasting. However, gauging stations are extremely limited across most high-mountain regions such as the Tibetan Plateau (TP), known as the Asia's water towers. Remote sensing, in combination with partial in situ discharge measurements, bridges the gap in monitoring river discharge over ungauged and poorly gauged basins. Of great importance for the successful retrieval of river discharge using remote sensing are river width (water surface area) and water level (water surface elevation), but it is challenging to retrieve accurate discharge values for high-mountain regions because of narrow river channels, complex terrain, and limited observations from a single satellite platform. Here, we used 1237 high-spatial-resolution images (Landsat series and Sentinel-1/2) to derive water surface areas with the Google Earth Engine (GEE), and satellite altimetry (Jason-2/3 and Satellite with Argos and AltiKa (SARAL/Altika)) to derive water levels for the Upper Brahmaputra River (UBR, the Yarlung Zangbo River in China) in the TP where the river width is typically less than 400 m. Using three power function equations, discharge was estimated for cross-sections around the four gauging stations in the UBR with triangular cross-sections outperforming their trapezoidal counterparts. It was also found that the equation combining both river width and water level produced the best discharge estimates whereas the other two equations (requiring either river width or water level as the input data) were complementary and could be used to extend the time series of discharge estimates. The Nash–Sutcliffe efficiency coefficient values for the discharge estimates range from 0.68 to 0.98 during the study period 2000–2017. The proposed method is feasible to estimate discharge in the UBR and potentially other high-mountain rivers globally.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Marcello Passaro, Zulfikar Adlan Nadzir, Graham D. Quartly〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉The sea state bias (SSB) is a large source of uncertainty in the estimation of sea level from satellite altimetry. It is still unclear to what extent it depends on errors in parameter estimations (numerical source) or to the wave physics (physical source).〈/p〉 〈p〉By improving the application of this correction we compute 20-Hz sea level anomalies that are about 30% more precise (i.e. less noisy) than the current standards. The improvement is two-fold: first we prove that the SSB correction should be applied directly to the 20-Hz data (12 to 19% noise decrease); secondly, we show that by recomputing a regional SSB model (based on the 20-Hz estimations) even a simple parametric relation is sufficient to further improve the correction (further 15 to 19% noise decrease).〈/p〉 〈p〉We test our methodology using range, wave height and wind speed estimated with two retrackers applied to Jason-1 waveform data: the MLE4 retracked-data available in the Sensor Geophysical Data Records of the mission and the ALES retracked-data available in the OpenADB repository (〈a href="https://openadb.dgfi.tum.de/" target="_blank"〉https://openadb.dgfi.tum.de/〈/a〉). The regional SSB models are computed parametrically by means of a crossover analysis in the Mediterranean Sea and North Sea.〈/p〉 〈p〉Correcting the high-rate data for the SSB reduces the correlation between retracked parameters. Regional variations in the proposed models might be due to differences in wave climate and remaining sea-state dependent residual errors. The variations in the empirical model with respect to the retracker used recall the need for a specific SSB correction for any retracker.〈/p〉 〈p〉This study, while providing a significantly more precise solution to exploit high-rate sea level data, calls for a re-thinking of the SSB correction in both its physical and numerical component, gives robustness to previous theories and provides an immediate improvement for the application of satellite altimetry in the regions of study.〈/p〉 〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Agnieszka Kamińska, Maciej Lisiewicz, Krzysztof Stereńczak, Bartłomiej Kraszewski, Rafał Sadkowski〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉The assessment of the health conditions of trees in forests is extremely important for biodiversity, forest management, global environment monitoring, and carbon dynamics. There is a vast amount of research using remote sensing (RS) techniques for the assessment of the current condition of a forest, but only a small number of these are concerned with detection and classification of dead trees. Among the available RS techniques, only the airborne laser scanner (ALS) enables dead tree detection at the single tree level with high accuracy.〈/p〉 〈p〉The main objective of the study was to identify spruce, pine and deciduous trees by alive or dead classifications. Three RS data sets including ALS (leaf-on and leaf-off) and color-infrared (CIR) imagery (leaf-on) were used for the study. We used intensity and structural variables from the ALS data and spectral information derived from aerial imagery for the classification procedure. Additionally, we tested the differences in the classification accuracy of all variants contained in the data integration. In the study, the random forest (RF) classifier was used. The study was carried out in the Polish part of the Białowieża Forest (BF).〈/p〉 〈p〉In general, we can state that all classifications, with different combinations of ALS features and CIR, resulted in high overall accuracy (OA ≥ 90%) and Kappa (κ 〉 0.86). For the best variant (CIR_ALS〈sub〉WSn-FH〈/sub〉), the mean values of overall accuracy and Kappa were equal to 94.3% and 0.93, respectively. The leaf-on point cloud features alone produced the lowest accuracies (OA = 75–81% and κ = 0.68–0.76). Improvements of 0-0.04 in the Kappa coefficient and 0–3.1% in the overall classification accuracy were found after the point cloud normalization for all variants. Full-height point cloud features (F) produced lower accuracies than the results based on features calculated for half of the tree height point clouds (H) and combined FH.〈/p〉 〈p〉The importance of each of the predictors for different data sets for tree species classification provided by the RF algorithm was investigated. The lists of top features were the same, independent of intensity normalization. For the classification based on both of the point clouds (leaf–on and leaf-off), three structural features (a proportion of first returns for both half-height and full-height variants and the canopy relief ratio of points) and two intensity features from first returns and half-height variant (the coefficient of variation and skewness) were rated as the most important. In the classification based on the point cloud with CIR features, two image features were among the most important (the NDVI and mean value of reflectance in the green band).〈/p〉 〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Jaroslav Čepl, Jan Stejskal, Zuzana Lhotáková, Dana Holá, Jiří Korecký, Milan Lstibůrek, Ivana Tomášková, Marie Kočová, Olga Rothová, Markéta Palovská, Jakub Hejtmánek, Anna Krejzková, Salvador Gezan, Ross Whetten, Jana Albrechtová〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉Foliar reflectance is readily used in evaluating physiological status of agricultural crops and forest stands. However, in the case of forest trees, underlying genetics of foliar spectral reflectance and pigment content have rarely been investigated. We studied a structured population of Scots pine, replicated on two sites, with the selected trees´ pedigree reconstructed via DNA markers. This allowed us to decompose phenotypic variance of pigment and reflectance traits into its causal genetic components, and to estimate narrow-sense heritability (〈em〉h〈/em〉〈sup〉〈em〉2〈/em〉〈/sup〉).〈/p〉 〈p〉We found statistically significant 〈em〉h〈/em〉〈sup〉〈em〉2〈/em〉〈/sup〉 ranging from 0.07 to 0.22 for most of the established reflectance indices. Additionally, we investigated the profile of heritable variation along the reflectance curve in 1 nm wavelength (WL) bands. We show that the maximum 〈em〉h〈/em〉〈sup〉〈em〉2〈/em〉〈/sup〉 value (0.39; SE 0.13) across the 400 to 2500 nm spectral range corresponds to the red edge inflection point, in this case to 722 nm WL band. Resultant 〈em〉h〈/em〉〈sup〉〈em〉2〈/em〉〈/sup〉 distribution indicates that additive gene effects fluctuate along the reflectance curve.〈/p〉 〈p〉Furthermore, 〈em〉h〈/em〉〈sup〉〈em〉2〈/em〉〈/sup〉 of the most widely used formats of reflectance indices, i.e. the simple ratio and the normalized difference, was estimated for all WL bands combined along the observed reflectance spectrum. The highest 〈em〉h〈/em〉〈sup〉〈em〉2〈/em〉〈/sup〉 estimates for both formats were obtained by combining WL bands of the red edge spectrum.〈/p〉 〈p〉These new genetically driven pigment- and spectral reflectance- based markers (proxies of adaptive traits) may facilitate selection of stress resistant plant genotypes. Recent development of high-resolution spectral sensors carried by airborne and spaceborn devices make foliage spectral traits a viable technology for mass phenotyping in forest trees.〈/p〉 〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Asa Gholizadeh, Daniel Žižala, Mohammadmehdi Saberioon, Luboš Borůvka〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Soil Organic Carbon (SOC) is a useful representative of soil fertility and an essential parameter in controlling the dynamics of various agrochemicals in soil. Soil texture is also used to calculate soil's ability to retain water for plant growth. SOC and soil texture are thus important parameters of agricultural soils and need to be regularly monitored. Optical satellite remote sensing offers the potential for frequent surveys over large areas. In addition, the recently-operated Sentinel-2 missions provide free imagery. This study compared the capabilities of Sentinel-2 for monitoring and mapping of SOC and soil texture (clay, silt and sand content) with those obtained from airborne hyperspectral (CASI/SASI sensors) and lab ASD FieldSpec spectroradiometer measurements at four agricultural sites in the Czech Republic. Combination of 10 extracted bands of the Sentinel-2 and 18 spectral indices, as independent variables, were used to train prediction models and then produce spatial distribution maps of the selected attributes. Results showed that the prediction accuracy based on lab spectroscopy, airborne and Sentinel-2 in the majority of the sites was adequate for SOC and fair for clay; however, Sentinel-2 imagery could not be used to detect and map variations in silt and sand. The SOC and clay maps derived from the airborne and spaceborne datasets showed similar trend, with both performing better where SOC levels were relatively high, though at the highest levels Sentinel-2 was able to create the SOC map more precisely than the airborne sensors. Taken across all SOC levels measured in the reference data, Sentinel-2 results were marginally lower than lab spectroscopy and airborne imagery, but this reduction in precision may be offset by the extensive geographical coverage and more frequent revisit characteristic of satellite observation. The increased temporal revisit and area are expected to be positive enhancements to the acquisition of high-quality information on variations in SOC and clay content of bare soils.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Jing Ge, Baoping Meng, Tiangang Liang, Qisheng Feng, Jinlong Gao, Shuxia Yang, Xiaodong Huang, Hongjie Xie〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Monitoring changes in grassland cover is essential in assessment of grassland health as well as the effects of anthropogenic interventions and global climate change on grassland ecosystems. Remote sensing is an effective approach for providing rapid and dynamic monitoring of vegetation cover over large grassland areas. In this study, four types of remote sensing retrieval models (i.e., pixel dichotomy models, univariate vegetation index (VI) regression models, multivariate regression models, and a support vector machine (SVM) model) are built to derive grassland cover based on moderate resolution imaging spectroradiometer (MODIS) data and the measured grassland cover data collected by unmanned aerial vehicle during the grassland peak growing season from 2014 to 2016. The optimal model is then used to map the spatial distribution of grassland cover and its dynamic change in the headwater region of the Huanghe River (Yellow River) (HRHR) of the northeastern Tibetan Plateau over the 16 years period (2001 to 2016). The results show that (1) the pixel dichotomy models based on MODIS VI data are inappropriate for estimating grassland cover in the HRHR when their endmembers (VI〈sub〉soil〈/sub〉 and VI〈sub〉veg〈/sub〉) are determined based only on the MODIS data; (2) the multivariate regression models present better performance than the univariate VI (normalized difference vegetation index (NDVI) or enhanced vegetation index (EVI)) models; (3) MODIS NDVI outperforms MODIS EVI for modeling grassland cover in the study area; (4) the SVM model based on nine factors is the optimal model (R〈sup〉2〈/sup〉: 0. 75 and RMSE: 6.85%) for monitoring alpine grassland cover in the study area; and (5) majority of the grassland area (59.9%) of the HRHR showed increase in yearly maximum grassland cover from 2001 to 2016, while the average yearly maximum grassland cover for the 16 years exhibited a generally increasing trend from west to east and from north to south. This study provides a more suitable remote sensing inversion model to greatly improve the accuracy of modeling alpine grassland cover in the HRHR, and to better assess grassland health status and the impacts of warming climate to grasslands in regions of remote and harsh environments.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): William D. Barnhart, William L. Yeck, Daniel E. McNamara〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Oklahoma experienced three earthquakes of M〈sub〉w〈/sub〉5.0 or greater in 2016: the 13-Feb. Fairview earthquake (M〈sub〉w〈/sub〉5.1), the 03-Sep. Pawnee earthquake (M〈sub〉w〈/sub〉5.8), and the 07-Nov. Cushing earthquake (M〈sub〉w〈/sub〉5.0). These events are the first earthquakes in the state exceeding M〈sub〉w〈/sub〉5.0 since the 2011 M〈sub〉w〈/sub〉5.7 Prague earthquake and likely result from wide-scale deep fluid-injection. We use interferometric synthetic aperture radar (InSAR) observations to quantify the magnitude and location of surface deformation associated with these three events, determine the depth ranges of fault slip, and assess the spatial relationship between fault slip and well-calibrated mainshock and aftershock locations. We also include newly reported, calibrated event locations for the Cushing earthquake. We find that the Pawnee earthquake ruptured within the crystalline basement with the shallowest slip occurring at depths of 3.1–4.3 km. We find a similar, though shallower, crystalline basement source for the Cushing earthquake with a minimum depth to slip of 1.6–2.3 km. Despite the smaller magnitude of the Cushing earthquake, it generated anomalously high ground motions and damage compared to the larger Pawnee and Fairview earthquakes. We postulate that the shallow source of the Cushing earthquakes provides one explanation for the higher than expected ground motions. The Fairview earthquake generated no detectable co-seismic displacements, which is consistent with a relatively deep earthquake source (~8.5 km). We do, however, identify a 16 km stretch of floodplain where widespread liquefaction occurred in response to the Fairview earthquake, and where 30 gas production wells were exposed to surface displacements exceeding 5 cm. Consequently, the depth to crystalline basement, which limits the depth of injection-induced earthquakes in Oklahoma, and the potential for liquefaction are important factors in assessing shaking risk in the central United States.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): 〈/p〉

Beschreibung: 〈p〉Publication date: Available online 2 July 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment〈/p〉 〈p〉Author(s): D.P. Roy, L. Yan〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Reliable satellite monitoring of agriculture is often difficult because surface variations occur rapidly compared to the cloud-free satellite observation frequency. Harmonic time series models, i.e., superimposed sequences of sines and cosines, have an established provenance for fitting satellite vegetation index time series to coarse resolution satellite data, but their application to medium resolution Landsat data for crop monitoring has been limited. Non-linear harmonic models have been shown to perform well over agricultural sites using single-year Moderate Resolution Imaging Spectroradiometer (MODIS) time series, but have not been explored with Landsat data. The 2017 availability of Landsat Analysis Ready Data (ARD) over the United States provides the opportunity to investigate the utility of temporally rich Landsat data for 30 m pixel-level crop monitoring. In this paper, the capability of 5- and 7-parameter linear harmonic models and a 5-parameter non-linear harmonic model applied to a year of Landsat 5 Thematic Mapper (TM) and Landsat 7 Enhanced Thematic Mapper Plus (ETM+) ARD is investigated. The analysis is undertaken over six sites, each defined by a 5000 × 5000 30 m pixel ARD tile, that together include the major conterminous United States (CONUS) crops identified by inspection of the United States Department of Agriculture (USDA) Cropland Data Layer (CDL). The model fits are evaluated as the root mean square difference (RMSD) between the fitted and the observed Landsat data. Considering locations with at least 21 annual Landsat observations, the 7-parameter linear harmonic model (tile mean crop NDVI RMSD values ranging from 0.052 to 0.072) and the 5-parameter non-linear harmonic model (tile mean crop NDVI RMSD values ranging from 0.054 to 0.074) are shown to be able to fit annual Landsat NDVI time series for most CONUS crops, whereas the 5-parameter linear harmonic model cannot (tile mean crop NDVI RMSD values ranging from 0.072 to 0.099). If there are between 15 and 20 annual Landsat observations, the 5-parameter non-linear harmonic model is recommended for fitting annual NDVI crop time series, and if there are ≥21 observations, then either the 5-parameter non-linear or the 7-parameter linear model can be used. The 7-parameter model had marginally smaller mean NDVI RMSD values but larger standard deviations than the 5-parameter non-linear model, likely due to the relative robustness of the non-linear model to over-fitting and oscillations. None of the models could reliably fit crops with multiple stages, such as alfalfa, that are insufficiently sampled using combined Landsat 5 TM and Landsat 7 ETM+ time series. Given the utility of the growing season peak NDVI for crop yield applications, the date and magnitude of the model fitted peak NDVI are compared to quantify model reporting differences. The differences between the 7-parameter linear and the 5-parameter non-linear harmonic models are not large. For each ARD tile, the mean absolute differences in the estimated peak NDVI days varied from 〈2 days in the northern ARD tiles, which had short growing seasons and similar crops, to less than a week for the other tiles except for nearly 10 days for the California tile that had longer growing seasons and more diverse crops including crops with multiple stages. The paper concludes with a discussion and recommendations for future research.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Martin Claverie, Junchang Ju, Jeffrey G. Masek, Jennifer L. Dungan, Eric F. Vermote, Jean-Claude Roger, Sergii V. Skakun, Christopher Justice〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉The Harmonized Landsat and Sentinel-2 (HLS) project is a NASA initiative aiming to produce a Virtual Constellation (VC) of surface reflectance (SR) data acquired by the Operational Land Imager (OLI) and Multi-Spectral Instrument (MSI) aboard Landsat 8 and Sentinel-2 remote sensing satellites, respectively. The HLS products are based on a set of algorithms to obtain seamless products from both sensors (OLI and MSI): atmospheric correction, cloud and cloud-shadow masking, spatial co-registration and common gridding, bidirectional reflectance distribution function normalization and spectral bandpass adjustment. Three products are derived from the HLS processing chain: (i) S10: full resolution MSI SR at 10 m, 20 m and 60 m spatial resolutions; (ii) S30: a 30 m MSI Nadir BRDF (Bidirectional Reflectance Distribution Function)-Adjusted Reflectance (NBAR); (iii) L30: a 30 m OLI NBAR. All three products are processed for every Level-1 input products from Landsat 8/OLI (L1T) and Sentinel-2/MSI (L1C). As of version 1.3, the HLS data set covers 10.35 million km〈sup〉2〈/sup〉 and spans from first Landsat 8 data (2013); Sentinel-2 data spans from October 2015.〈/p〉 〈p〉The L30 and S30 show a good consistency with coarse spatial resolution products, in particular MODIS Collection 6 MCD09CMG products (overall deviations do not exceed 11%) that are used as a reference for quality assurance. The spatial co-registration of the HLS is improved compared to original Landsat 8 L1T and Sentinel-2A L1C products, for which misregistration issues between multi-temporal data are known. In particular, the resulting computed circular errors at 90% for the HLS product are 6.2 m and 18.8 m, for S10 and L30 products, respectively. The main known issue of the current data set remains the Sentinel-2 cloud mask with many cloud detection omissions. The cross-comparison with MODIS was used to flag products with most evident non-detected clouds. A time series outlier filtering approach is suggested to detect remaining clouds. Finally, several time series are presented to highlight the high potential of the HLS data set for crop monitoring.〈/p〉 〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Francisco Zambrano, Anton Vrieling, Andy Nelson, Michele Meroni, Tsegaye Tadesse〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Global food security is negatively affected by drought. Climate projections show that drought frequency and intensity may increase in different parts of the globe. These increases are particularly hazardous for developing countries. Early season forecasts on drought occurrence and severity could help to better mitigate the negative consequences of drought. The objective of this study was to assess if interannual variability in agricultural productivity in Chile can be accurately predicted from freely-available, near real-time data sources. As the response variable, we used the standard score of seasonal cumulative NDVI (zcNDVI), based on 2000–2017 data from Moderate Resolution Imaging Spectroradiometer (MODIS), as a proxy for anomalies of seasonal primary productivity. The predictions were performed with forecast lead times from one- to six-month before the end of the growing season, which varied between census units in Chile. Predictor variables included the zcNDVI obtained by cumulating NDVI from season start up to prediction time; standardised precipitation indices derived from satellite rainfall estimates, for time-scales of 1, 3, 6, 12 and 24 months; the Pacific Decadal Oscillation and the Multivariate ENSO oscillation indices; the length of the growing season, and latitude and longitude. For each of the 758 census units considered, the time series of the response and the predictor variables were averaged for agricultural areas resulting in a 17-season time series per unit for each variable. We used two prediction approaches: (i) optimal linear regression (OLR) whereby for each census unit the single predictor was selected that best explained the interannual zcNDVI variability, and (ii) a multi-layer feedforward neural network architecture, often called deep learning (DL), where all predictors for all units were combined in a single spatio-temporal model. Both approaches were evaluated with a leave-one-year-out cross-validation procedure. Both methods showed good prediction accuracies for small lead times and similar values for all lead times. The mean 〈em〉R〈/em〉〈sub〉〈em〉cv〈/em〉〈/sub〉〈sup〉2〈/sup〉 values for OLR were 0.95, 0.83, 0.68, 0.56, 0.46 and 0.37, against 0.96, 0.84, 0.65, 0.54, 0.46 and 0.38 for DL, for one, two, three, four, five, and six months lead time, respectively. Given the wide range of climates and vegetation types covered within the study area, we expect that the presented models can contribute to an improved early warning system for agricultural drought in different geographical settings around the globe.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Yidi Xu, Le Yu, Feng R. Zhao, Xueliang Cai, Jiyao Zhao, Hui Lu, Peng Gong〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Ensuring food security has been the top priority of many regions, particularly in developing countries in Africa. In recent decades, increasing population, together with growing food demands, have put great pressure on the world's food production. Long-term, up-to-date, annual cropland mapping at high resolution (i.e., at tens-of-metre levels) is in urgent demand for tracking spatial and temporal patterns of cropland change. However, because of the difficulty of capturing seasonality and flexible cropping systems, few studies have focused on understanding the dynamics of cropland using Landsat data in Africa. Here, we propose a new method of updating annual cropland mapping using a change-detection approach and post-classification to improve on traditional bi-temporal change vector analysis. Three Landsat footprints in Africa were selected (Egypt, Ethiopia and South Africa) as our study areas based on their different cropping systems and field sizes. The potential annual change areas were detected by employing multiple indices and thresholds in reference and long-term annual composite Landsat images. Next, map updates were conducted in the potential change pixels using random forest-based classification. Different training sample metrics were used (seasonal and annual samples) and compared in the classification step. The long-term cropland mapping accuracies for these three sites ranged from 88.04% to 94.30% (Egypt), 76.28% to 82.88% (Ethiopia) and 56.52% to 67.53% (South Africa). The results showed improvements in the accuracy and consistency of updating the annual cropland information using change-detection approaches, accounting for accuracy increases of 2.40%, 10.62% and 0.55% compared with a yearly cropland mapping approach in our previous research. The best results using annual samples extracted from the same season with the classified images supported the use of annual and growing samples in long-term annual mapping. Overall, a common trend of cropland expansion in all three sites was revealed, with an increase rate of 10.06, 3.73 and 1.35 kha/year in Egypt, Ethiopia and South Africa, respectively. The results indicated a rapid increasing pattern from bare land (desert) to irrigated systems (Egyptian site) but smaller and stable cropland changes in smallholder and farming-pastoral ecotones (Ethiopian and South African site), where limited land was still available for an expansion of agricultural area. This study highlights the potential application of time-series Landsat data in documenting and contributing missing cropland distribution information required for assessing and solving food security in Africa.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Arjan J.H. Meddens, Lee A. Vierling, Jan U.H. Eitel, Jyoti S. Jennewein, Joanne C. White, Michael A. Wulder〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Digital terrain models (DTMs) and vegetation canopy height models (CHMs) are used in a wide range of earth and environmental sciences. An increasing number of CHM products are available from active, passive, and photogrammetric remotely sensed data; however, high-resolution (≤5 m), wall-to-wall CHMs for the arctic and northern boreal domains that are suitable for detailed spatial analysis are lacking. Recently, a 5-m spatial resolution pan-arctic digital surface model – the ArcticDEM – was created using automated stereopair analysis of high-resolution satellite data. The ArcticDEM is unprecedented in extent and spatial resolution, yet the product generally follows the uppermost surface elevation (i.e., representing a digital surface model, DSM) without regard to whether the surface is comprised of vegetation or bare-earth terrain. To address this limitation, we developed and tested an approach to map vegetation canopy height at a 5-m spatial resolution (hereafter called the local ArcticCHM), and then subtracted these estimated canopy heights from the ArcticDEM in order to create a 5-m resolution DTM (local ArcticDTM). We selected three pilot study areas (total 58 km〈sup〉2〈/sup〉) across a north-south gradient in Alaska, representing a range of vegetation types and topographic conditions. We estimated and mapped canopy height using randomForest and imputation modeling approaches, with the ArcticDEM and high spatial resolution multispectral satellite data (WorldView-2) used as predictors. Airborne laser scanning (ALS) data was used for model calibration and independent validation. Canopy height was reliably predicted across the three study areas, with the best models ranging from root mean square errors (RMSE) 2.2 to 2.6 m and R〈sup〉2〈/sup〉 ranging from 0.59 to 0.76 relative to ALS-based CHM reference data. Similarly, the RMSE between the new local ArcticDTM product and ALS-based DTM reference data was 45–68% less than similar comparisons with the ArcticDEM. Our results offer a means to extend these local ArcticDTM and CHM products to establish high-resolution products elsewhere in Alaska of high value for a wide range of earth and environmental sciences research investigations.〈/p〉〈/div〉 〈/div〉 〈div xml:lang="en"〉 〈h5〉Graphical abstract〈/h5〉 〈div〉〈p〉〈figure〉〈img src="https://ars.els-cdn.com/content/image/1-s2.0-S0034425718304218-ga1.jpg" width="301" alt="Unlabelled Image" title="Unlabelled Image"〉〈/figure〉〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: Available online 13 February 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment〈/p〉 〈p〉Author(s): Á. González-Zamora, N. Sánchez, M. Pablos, J. Martínez-Fernández〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉In this research, the active, passive and combined Climate Change Initiative (CCI) Soil Moisture (SM) products were evaluated in comparison with 〈em〉in situ〈/em〉 SM measurements from five networks in Spain that have different spatial and temporal scales, densities and environmental conditions. Three of these networks, namely Rinconada, Morille and the Soil Moisture Measurement Stations Network of the University of Salamanca (REMEDHUS), are small- to medium-scale networks and have high station densities, whereas the other two (Inforiego and FluxNet) are sparse and large-scale networks.〈/p〉 〈p〉The results of the comparisons with the former v02.2 version (before the inclusion of the SM retrieved by the Soil Moisture and Ocean Salinity mission, SMOS, in the CCI dataset) showed that the combined CCI performed better than the active or passive, affording correlation coefficients (R) above 0.8 and errors between 0.03 and 0.08 m〈sup〉3〈/sup〉 m〈sup〉−3〈/sup〉 for the area-average, with biases close to zero. Regarding the land uses and environmental conditions, the stations that were located in the agricultural areas and some forested areas showed the best results, and those that were located in pasture and certain specific agricultural locations showed the poorest results.〈/p〉 〈p〉To test the opportunity of including SMOS in CCI, both datasets were compared over the same areas and coincident periods. After the results, the combined CCI and SMOS SM products matched very well (〈em〉R〈/em〉 = 0.83 on average), although the SMOS and CCI under- and overestimate the ground soil moisture measurements, respectively.〈/p〉 〈p〉Finally, the new version of the combined CCI (v03.2, after including SMOS) showed similar correlations to the previous one, but it significantly reduced the bias, leading to slightly lower errors (RMSD and cRMSD). Hence, it was shown that including SMOS in the CCI database enhanced its performance.〈/p〉 〈p〉The results in this work may improve knowledge of the CCI SM and its potential applications.〈/p〉 〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Gourav Misra, Allan Buras, Marco Heurich, Sarah Asam, Annette Menzel〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉In the past, studies have successfully identified climatic controls on the temporal variability of the land surface phenology (LSP). Yet we lack a deeper understanding of the spatial variability observed in LSP within a land cover type and the factors that control it. Here we make use of a high resolution LiDAR based dataset to study the effect of subpixel forest stand characteristics on the spatial variability of LSP metrics based on MODIS NDVI. Multiple linear regression techniques (MLR) were applied on forest stand information and topography derived from LiDAR as well as land cover information (i.e. CORINE and proprietary habitat maps for the year 2012) to predict average LSP metrics of the mountainous Bavarian Forest National Park, Germany. Six different LSP metrics, i.e. start of season (SOS), end of season (EOS), length of season (LOS), NDVI integrated over the growing season (NDVIsum), maximum NDVI value (NDVImax) and day of maximum NDVI (maxDOY) were modelled in this study. It was found that irrespective of the land cover, the mean SOS, LOS and NDVIsum were largely driven by elevation. However, inclusion of detailed forest stand information improved the models considerably. The EOS however was more complex to model, and the subpixel percentage of broadleaf forests and the slope of the terrain were found to be more strongly linked to EOS. The explained variance of the NDVImax improved from 0.45 to 0.71 when additionally considering land cover information, which further improved to 0.84 when including LiDAR based subpixelforest stand characteristics. Since completely homogenous pixels are rare in nature, our results suggest that incorporation of subpixel forest stand information along with land cover type leads to an improved performance of topography based LSP models. The novelty of this study lies in the use of topography, land cover and subpixel vegetation characteristics derived from LiDAR in a stepwise manner with increasing level of complexity, which demonstrates the importance of forest stand information on LSP at the pixel level.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Vitor S. Martins, João V. Soares, Evlyn M.L.M. Novo, Claudio C.F. Barbosa, Cibele T. Pinto, Jeferson S. Arcanjo, Amy Kaleita〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉A practical atmospheric correction algorithm, called Coupled Moderate Products for Atmospheric Correction (CMPAC), was developed and implemented for the Multispectral Camera (MUX) on-board the China-Brazil Earth Resources Satellite (CBERS-4). This algorithm uses a scene-based processing and sliding window technique to derive MUX surface reflectance (SR) at continental scale. Unlike other optical sensors, MUX instrument imposes constraints for atmospheric correction due to the absence of spectral bands for aerosol estimation from imagery itself. To overcome this limitation, the proposed algorithm performs a further processing of atmospheric products from Moderate Resolution Imaging Spectroradiometer (MODIS) and Visible Infrared Imaging Radiometer Suite (VIIRS) sensors as input parameters for radiative transfer calculations. The success of CMPAC algorithm was fully assessed and confirmed by comparison of MUX SR data with the Landsat-8 OLI Level-2 and Aerosol Robotic Network (AERONET)-derived SR products. The spectral adjustment was performed to compensate for the differences of relative spectral response between MUX and OLI sensors. The results show that MUX SR values are fairly similar to operational Landsat-8 SR products (mean difference 〈 0.0062, expressed in reflectance). There is a slight underestimation of MUX SR compared to OLI product (except the NIR band), but the error metrics are typically low and scattered points are around the line 1:1. These results suggest the potential of combining these datasets (MUX and OLI) for quantitative studies. Further, the robust agreement of MUX and AERONET-derived SR values emphasizes the quality of moderate atmospheric products as input parameters in this application, with root-mean-square deviation lower than 0.0047. These findings confirm that (i) CMPAC is a suitable tool for estimating surface reflectance of CBERS MUX data, and (ii) ancillary products support the application of atmospheric correction by filling the gap of atmospheric information. The uncertainties of atmospheric products, negligence of the bidirectional effects, and two aerosol models were also identified as a limitation. Finally, this study presents a framework basis for atmospheric correction of CBERS-4 MUX images. The utility of CBERS data comes from its use, and this new product enables the quantitative remote sensing for land monitoring and environmental assessment at 20 m spatial resolution.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Jean-François Côté, Richard A. Fournier, Joan E. Luther, Olivier R. van Lier〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Improving the quality of information that can be obtained from forest inventories can enhance planning for the best use of forest resources. In this study, we demonstrate the capability to improve the characterization of forest inventory attributes using terrestrial laser scanner (TLS) data, a fine-scale architectural model (〈em〉L-Architect〈/em〉), and airborne laser scanner (ALS) data. Terrestrial laser scanning provides detailed and accurate three-dimensional data and has the potential to characterize forest plots with comprehensive structural information. We use TLS data and in situ measurements as input to 〈em〉L-Architect〈/em〉 to create reference plots. The use of 〈em〉L-Architect〈/em〉 for modeling was validated by comparing selected attributes of the reference plots with validation plots produced using simulated TLS data, with normalized root-mean square error (NMRSE) values below 17%. Surrogate plots were then created using a library of tree models where individual trees were selected according to three attributes—tree height, diameter at breast height, and crown projected area—either measured from in situ plots or derived from ALS data. The accuracy of the surrogate plots was assessed by comparing several key forest attributes from the reference plots, including branching structure (e.g., number of whorls, knot surface), crown shape and size (e.g., base height, asymmetry), heterogeneity (e.g., lacunarity, fractal dimension), tree volume, and the spatial distribution of material (e.g., Weibull fit, leaf area index). Overall, the surrogate plots reproduced the attributes of the reference plots with NRMSE mean value of 17% (〈em〉R〈/em〉〈sup〉2〈/sup〉 = 0.68) using in situ ground measurements and 24% (〈em〉R〈/em〉〈sup〉2〈/sup〉 = 0.51) using inputs estimated with ALS. Some attributes, such as leaf area index, knot surface, and fractal dimension, were well predicted (〈em〉R〈/em〉〈sup〉2〈/sup〉 〉 0.80), whereas others, like crown asymmetry and lacunarity, had weak correspondence (〈em〉R〈/em〉〈sup〉2〈/sup〉 〈 0.16). The ability to create surrogate forest plots with 〈em〉L-Architect〈/em〉 makes it possible to estimate detailed structural attributes that are difficult to measure with conventional forest mensuration techniques and that can be used for model calibration with above-canopy remote-sensing data sets.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Álvaro Moreno-Martínez, Gustau Camps-Valls, Jens Kattge, Nathaniel Robinson, Markus Reichstein, Peter van Bodegom, Koen Kramer, J. Hans C. Cornelissen, Peter Reich, Michael Bahn, Ülo Niinemets, Josep Peñuelas, Joseph M. Craine, Bruno E.L. Cerabolini, Vanessa Minden, Daniel C. Laughlin, Lawren Sack, Brady Allred, Christopher Baraloto, Chaeho Byun〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉This paper introduces a modular processing chain to derive global high-resolution maps of leaf traits. In particular, we present global maps at 500 m resolution of specific leaf area, leaf dry matter content, leaf nitrogen and phosphorus content per dry mass, and leaf nitrogen/phosphorus ratio. The processing chain exploits machine learning techniques along with optical remote sensing data (MODIS/Landsat) and climate data for gap filling and up-scaling of in-situ measured leaf traits. The chain first uses random forests regression with surrogates to fill gaps in the database (〉 45〈em〉%〈/em〉 of missing entries) and maximizes the global representativeness of the trait dataset. Plant species are then aggregated to Plant Functional Types (PFTs). Next, the spatial abundance of PFTs at MODIS resolution (500 m) is calculated using Landsat data (30 m). Based on these PFT abundances, representative trait values are calculated for MODIS pixels with nearby trait data. Finally, different regression algorithms are applied to globally predict trait estimates from these MODIS pixels using remote sensing and climate data. The methods were compared in terms of precision, robustness and efficiency. The best model (random forests regression) shows good precision (normalized RMSE≤ 20〈em〉%〈/em〉) and goodness of fit (averaged Pearson's correlation R = 0.78) in any considered trait. Along with the estimated global maps of leaf traits, we provide associated uncertainty estimates derived from the regression models. The process chain is modular, and can easily accommodate new traits, data streams (traits databases and remote sensing data), and methods. The machine learning techniques applied allow attribution of information gain to data input and thus provide the opportunity to understand trait-environment relationships at the plant and ecosystem scales. The new data products – the gap-filled trait matrix, a global map of PFT abundance per MODIS gridcells and the high-resolution global leaf trait maps – are complementary to existing large-scale observations of the land surface and we therefore anticipate substantial contributions to advances in quantifying, understanding and prediction of the Earth system.〈/p〉〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Lisheng Song, Shaomin Liu, William P. Kustas, Hector Nieto, Liang Sun, Ziwei Xu, Todd H. Skaggs, Yang Yang, Minguo Ma, Tongren Xu, Xuguang Tang, Qiuping Li〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Operational estimation of spatio-temporal continuously daily evapotranspiration (〈em〉ET〈/em〉), and the components evaporation (〈em〉E〈/em〉) and transpiration (〈em〉T〈/em〉), at river basin scale is very useful for developing sustainable water resource strategies, particularly in regions of limited water supplies. In this study, multi-year all-weather daily 〈em〉ET〈/em〉, 〈em〉E〈/em〉 and 〈em〉T〈/em〉 were estimated using MODIS-based (Dual Temperature Difference) DTD model under different land covers in the Heihe river basin in China, with a total area of approximately 143 × 10〈sup〉3〈/sup〉 km〈sup〉2〈/sup〉. The remotely sensed ET was validated using ground measurements from large aperture scintillometer systems, with a source area of several kilometers, over grassland, cropland and riparian shrub-forest land cover. The results showed that the remotely sensed ET produced mean absolute percent differences (MAPD) of around 20% with the ground measurements during the growing season under clear sky conditions, but the model performance deteriorated for cloudy days. However, the daily ET product gave reasonable estimates for croplands with an MAPD value of about 20% and the estimates of T/ET and E/ET in good agreement with ground measurements. The DTD model also significantly outperformed other remote sensing-based models being applied globally. Based on these results the DTD model is considered reliable for monitoring crop water use and stress and to develop efficient irrigation strategies.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Milad Niroumand-Jadidi, Alfonso Vitti, David R. Lyzenga〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Remote mapping of bathymetry can play a key role in gaining spatial and temporal insight into fluvial processes, ranging from hydraulics and morphodynamics to habitat conditions. This research introduces Multiple Optimal Depth Predictors Analysis (MODPA), which combines previously developed depth predictors along with additional predictors derived from the intensity component of the HSI color space transformation. MODPA empirically selects a set of optimal predictors among all candidates utilizing partial least squares (PLS), stepwise, or principal component (PC) regression models. The primary focus of this study was on shallow (〈1 m deep) and clearly flowing streams where substrate variability could have a pronounced effect on depth retrieval. Spectroscopic experiments were performed under controlled conditions in a hydraulic laboratory to examine the robustness of bathymetry models with respect to changes in bottom type. Further, simulations from radiative transfer modeling were used to extend the analysis by isolating the effect of inherent optical properties (IOPs) and by investigating the performance of bathymetry models in optically complex and deeper streams. The bathymetry of the Sarca River, a shallow river in the Italian Alps, was mapped using a WorldView-2 (WV-2) image, for which we evaluated the atmospheric compensation (AComp) product. Results indicated the greater robustness of multiple-predictor models particularly MODPA rather than single-predictor models, such as Optimal Band Ratio Analysis (OBRA), with respect to heterogeneity of bottom types, IOPs, and atmospheric effects. The HSI intensity component enhanced the accuracy of depth retrieval, particularly in optically-complex waters and also for low spectral resolution imagery (e.g., GeoEye). Further, the enhanced spectral resolution of WV-2 imagery improved bathymetry retrieval compared to 4-band GeoEye data.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 1 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 218〈/p〉 〈p〉Author(s): Rami Piiroinen, Fabian Ewald Fassnacht, Janne Heiskanen, Eduardo Maeda, Benjamin Mack, Petri Pellikka〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉〈em〉Eucalyptus〈/em〉 spp. and 〈em〉Acacia mearnsii〈/em〉 are common exotic tree species in eastern Africa that have shown (strong) invasive behavior in some regions. 〈em〉Acacia mearnsii〈/em〉 is considered a highly invasive species that is replacing native species and 〈em〉Eucalyptus〈/em〉 spp. are known to consume high amounts of groundwater with suspected effects on native flora. Mapping the occurrence of these species in the Taita Hills, Kenya (part of the Eastern Arc Mountains Biodiversity Hotspot) is important as there is lack of knowledge on their occurrence and ecological impact in the area. Mapping methods that require a lot of fieldwork are impractical in areas like the Taita Hills, where the terrain is rugged and the infrastructure is poor. Our aim was hence to map the occurrence of these tree species in a 100 km〈sup〉2〈/sup〉 area using airborne imaging spectroscopy and laser scanning. We used a one class biased support vector machine (BSVM) classifier as it needs labeled training data only for the positive classes (〈em〉A〈/em〉. 〈em〉mearnsii〈/em〉 and 〈em〉Eucalyptus〈/em〉 spp.), which potentially reduces the amount of required fieldwork. We also introduce a new approach for parameterizing and setting the threshold level simultaneously for the BSVM classifier. The second aim was to link the occurrence of these species to selected environmental variables. The results showed that the BSVM classifier is suitable for mapping 〈em〉Acacia mearnsii〈/em〉 and 〈em〉Eucalyptus〈/em〉 spp., holding the potential to improve the efficiency of field data collection. The introduced parametrization/threshold selection method performed better than other commonly used approaches. The crown level F1-score was 0.76 for 〈em〉Eucalyptus〈/em〉 spp. and 0.78 for 〈em〉A〈/em〉. 〈em〉mearnsii〈/em〉. We show that 〈em〉Eucalyptus〈/em〉 spp. and 〈em〉A. mearnsii〈/em〉 trees cover 0.8% and 1.6% of the study area, respectively. Both species are particularly located on steeper slopes and higher altitudes. Both species have significant occurrences in areas close to the biggest remaining native forest patch (Ngangao) in the study area. Nonetheless, follow-up studies are needed to evaluate their impact on the native flora and fauna, as well as their impact on the water resources. The maps created in this study in combination with such follow-up studies could serve as base data to generate guidelines that authorities can use to take action in handling the problems these species are causing.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: Available online 4 July 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment〈/p〉 〈p〉Author(s): Tom W. Bell, James G. Allen, Kyle C. Cavanaugh, David A. Siegel〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉Global, repeat satellite imagery serves as an essential tool to understand ecological patterns and processes over diverse temporal and spatial scales. Recently, the use of spaceborne imagery has become indispensable for monitoring giant kelp, a globally distributed foundation species that displays variable seasonal and interannual dynamics. In order to develop and maintain a continuous and spatially expansive time series, we describe a fully automated protocol to classify giant kelp canopy biomass across three Landsat sensors. This required correcting kelp canopy estimates to account for changes in the spectral response functions between the three sensors by simulating data using hyperspectral imagery. Combining multiple sensors also necessitated the use of an extended (15 year) time series of diver estimated kelp biomass to validate each sensor. We also describe a novel gap filling technique using known spatial scales of kelp biomass synchrony to correct for missing data due to the Enhanced Thematic Mapper Plus scan line corrector failure. These developments have led to a publicly available 34-year, seasonal time series of kelp canopy biomass across ~1500 km of California coastline.〈/p〉 〈p〉We then use this time series to examine the role of temporal and spatial scale on the detection of long-term biomass trends. We found that kelp canopy biomass trends are associated with trends in low frequency marine climate oscillations, like the North Pacific Gyre Oscillation. Long-term (~20 year) increases in the state of the North Pacific Gyre Oscillation have led to a cooler, nutrient-rich environment that benefits the growth of giant kelp across a large portion of the kelp biomass time series, however recent warming events have led to weak, or nonexistent trends over the length of the time series. The cyclical nature of these low frequency marine climate oscillations complicates the detection of trends that may be associated with anthropogenic climate change. A longer and continuous time series is needed to analyze long-term canopy biomass trends outside of natural low frequency climate variability for giant kelp ecosystems along the coast of California, if we are to make accurate assessments of the impacts of climate change.〈/p〉 〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: January 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 220〈/p〉 〈p〉Author(s): Nima Pahlevan, Sandeep K. Chittimalli, Sundarabalan V. Balasubramanian, Vincenzo Vellucci〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Sentinel-2 and Landsat data products when combined open opportunities for capturing the dynamics of nearshore coastal and inland waters at rates that have never been possible before. Recognizing the differences in their spectral and spatial sampling, to generate a seamless data record for global water quality monitoring, it is critical to quantify how well the derived data products agree under various atmospheric and aquatic conditions. This study provides an extensive quantitative assessment of how Landsat-8 and Sentinel-2A/B equivalent data products compare and discusses implications on differences in downstream products generated via the SeaWiFS Data Analysis System (SeaDAS). These products include the top-of-atmosphere (TOA) reflectance (〈em〉ρ〈/em〉〈sub〉〈em〉t〈/em〉〈/sub〉), the remote-sensing reflectance (〈em〉R〈/em〉〈sub〉〈em〉rs〈/em〉〈/sub〉), as well as biogeochemical properties, such as the total suspended solids (TSS). The analyses are conducted a) for Landsat-8 and Sentinel-2A/B near-simultaneous nadir overpasses (n-SNO) and b) over several highly turbid/eutrophic inland/nearshore waters. Following the implementation of vicarious gains for Sentinel-2A, the n-SNO analyses indicated that Landsat-8 and Sentinel-2A agree within ±1% in 〈em〉ρ〈/em〉〈sub〉〈em〉t〈/em〉〈/sub〉 and ±5% in 〈em〉R〈/em〉〈sub〉〈em〉rs〈/em〉〈/sub〉 products across the visible and near-infrared (NIR) bands. Similar evaluations with preliminary vicarious gains for Sentinel-2B showed ±2% in 〈em〉ρ〈/em〉〈sub〉〈em〉t〈/em〉〈/sub〉 and ±7% in 〈em〉R〈/em〉〈sub〉〈em〉rs〈/em〉〈/sub〉 products. Considering Landsat-8-derived 〈em〉R〈/em〉〈sub〉〈em〉rs〈/em〉〈/sub〉 products as a reference, we found 〈5% difference in Sentinel-2A and -2B 〈em〉R〈/em〉〈sub〉〈em〉rs〈/em〉〈/sub〉 products. Analyses of combined TSS and 〈em〉R〈/em〉〈sub〉〈em〉rs〈/em〉〈/sub〉 time-series products over several aquatic systems further corroborated these results and demonstrated the remarkable value of combined products. Occasional negative retrievals of 〈em〉R〈/em〉〈sub〉〈em〉rs〈/em〉〈/sub〉 products over hyper-eutrophic and highly turbid waters suggest the need for improvements in the atmospheric correction procedure to empower science/application community to fully explore Landsat-Sentinel-2 products. With very similar absolute radiometric observations and products, the science community should consider developments of suitable biogeochemical algorithms to maximize the utility of merged Landsat-Sentinel-2 products.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): Andrew F. Feldman, Ruzbeh Akbar, Dara Entekhabi〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Vegetation cover absorbs and scatters L-band microwave emission measured by SMOS and SMAP satellites. Misrepresentation of this phenomena results in uncertainties when inferring, for instance, surface soil moisture in retrieval algorithms that commonly utilize the tau-omega model which is most applicable for a weakly scattering medium. In this study, we investigate the degree to which multiple-scattering is prevalent over a range of land cover classifications (from lightly vegetated grasslands to dense forests) at the satellite scale by explicitly accounting for multiple-scattering in a first-order radiative transfer model, developed here. Even though the tau-omega model with effective parameters can possibly capture higher-order scattering contributions, deliberately partitioning scattering into different components is required to estimate multiple-scattering properties. Specifically, we aim to determine how one can partition between zeroth and first-order radiative transfer terms within a retrieval algorithm without ancillary information, determine whether this method can detect first-order scattering at the SMAP measurement scale without ancillary information, and quantify the magnitude of detected scattering. A simplified first-order radiative transfer model which characterizes single interactions of microwaves with a scattering medium is developed for implementation within retrieval algorithms. This new emission model is implemented within a recently developed retrieval algorithm, the multi-temporal dual channel algorithm (MT-DCA), which does not require ancillary land use information. Scattering parameters as well as 〈em〉SM〈/em〉 and vegetation optical depth (〈em〉τ〈/em〉) are retrieved simultaneously in Africa and South America using the first year of SMAP brightness temperature measurements on a 36 km grid. Specifically, an introduced time invariant first-order scattering coefficient (〈em〉ω〈/em〉〈sub〉1〈/sub〉) is retrieved representing microwave emission interaction with the canopy. We find that 〈em〉ω〈/em〉〈sub〉1〈/sub〉 is typically zero in lightly vegetated biomes and non-zero (~0.06) in 74% of the forest pixels. In forest-dominated pixels, the median first-order emissivity is 0.04, or about 4.3% of a given SMAP radiometer brightness temperature measurement. Additionally, explicitly accounting for first-order scattering terms in the radiative transfer model tends to increase 〈em〉SM〈/em〉 and 〈em〉τ〈/em〉 retrievals by a median of 0.02 m〈sup〉3〈/sup〉/m〈sup〉3〈/sup〉 and 0.1, respectively, only in forested regions. This study demonstrates the first attempt to explicitly partition higher-order scattering terms in a retrieval algorithm at a satellite scale and ultimately provides a fundamental understanding and quantification of multiple-scattering from grasslands to forests.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: 15 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 219〈/p〉 〈p〉Author(s): N-E. Tsendbazar, M. Herold, S. de Bruin, M. Lesiv, S. Fritz, R. Van De Kerchove, M. Buchhorn, M. Duerauer, Z. Szantoi, J.-F. Pekel〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The production of global land cover products has accelerated significantly over the past decade thanks to the availability of higher spatial and temporal resolution satellite data and increased computation capabilities. The quality of these products should be assessed according to internationally promoted requirements e.g., by the Committee on Earth Observation Systems-Working Group on Calibration and Validation (CEOS-WGCV) and updated accuracy should be provided with new releases (Stage-4 validation). Providing updated accuracies for the yearly maps would require considerable effort for collecting validation datasets. To save time and effort on data collection, validation datasets should be designed to suit multiple map assessments and should be easily adjustable for a timely validation of new releases of land cover products. This study introduces a validation dataset aimed to facilitate multi-purpose assessments and its applicability is demonstrated in three different assessments focusing on validating discrete and fractional land cover maps, map comparison and user-oriented map assessments. The validation dataset is generated primarily to validate the newly released 100 m spatial resolution land cover product from the Copernicus Global Land Service (CGLS-LC100). The validation dataset includes 3617 sample sites in Africa based on stratified sampling. Each site corresponds to an area of 100 m × 100 m. Within site, reference land cover information was collected at 100 subpixels of 10 m × 10 m allowing the land cover information to be suitable for different resolution and legends. Firstly, using this dataset, we validated both the discrete and fractional land cover layers of the CGLS-LC100 product. The CGLS-LC100 discrete map was found to have an overall accuracy of 74.6 ± 2.1% (at 95% confidence level) for the African continent. Fraction cover products were found to have mean absolute errors of 9.3, 8.8, 16.2, and 6.5% for trees, shrubs, herbaceous vegetation and bare ground, respectively. Secondly, for user-oriented map assessment, we assessed the accuracy of the CGLS-LC100 map from four user groups' perspectives (forest monitoring, crop monitoring, biodiversity and climate modelling). Overall accuracies for these perspectives vary between 73.7% ± 2.1% and 93.5% ± 0.9%, depending on the land cover classes of interest. Thirdly, for map comparison, we assessed the accuracy of the Globeland30-2010 map at 30 m spatial resolution. Using the subpixel level validation data, we derived 15,252 sample pixels at 30 m spatial resolution. Based on these sample pixels, the overall accuracy of the Globeland30-2010 map was found to be 66.6 ± 2.4% for Africa. The three assessments exemplify the applicability of multi-purpose validation datasets which are recommended to increase map validation efficiency and consistency. Assessments of subsequent yearly maps can be conducted by augmenting or updating the dataset with sample sites in identified change areas.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): P.M. Salgado-Hernanz, M.-F. Racault, J.S. Font-Muñoz, G. Basterretxea〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Seventeen years (1998–2014) of satellite-derived chlorophyll concentration (Chl) are used to analyse the seasonal and non-seasonal patterns of Chl variability and the long-term trends in phytoplankton phenology in the Mediterranean Sea. With marked regional variations, we observe that seasonality dominates variability representing up to 80% of total Chl variance in oceanic areas, whereas in shelf-sea regions high frequency variations may be dominant representing up to 49% of total Chl variance. Seasonal variations are typically characterized by a phytoplankton growing period occurring in spring and spanning on average 170 days in the western basin and 150 days in the eastern basin. The variations in peak Chl concentrations are higher in the western basin (0.88 ± 1.01 mg m〈sup〉−3〈/sup〉) compared to the eastern basin (0.35 ± 1.36 mg m〈sup〉−3〈/sup〉). Differences in the seasonal cycle of Chl are also observed between open ocean and coastal waters where more than one phytoplankton growing period are frequent (〉0.8 probability). During the study period, on average in the western Mediterranean basin (based on significant trends observed over ~95% of the basin), we show a positive trend in Chl of +0.015 ± 0.016 mg m〈sup〉−3〈/sup〉 decade〈sup〉−1〈/sup〉, and an increase in the amplitude and duration of the phytoplankton growing period by +0.27 ± 0.29 mg m〈sup〉−3〈/sup〉 decade〈sup〉−1〈/sup〉 and +11 ± 7 days decade〈sup〉−1〈/sup〉 respectively. Changes in Chl concentration in the eastern (and more oligotrophic) basin are generally low, with a trend of −0.004 ± 0.024 mg m〈sup〉−3〈/sup〉 decade〈sup〉−1〈/sup〉 on average (based on observed significant trends over ~70% of the basin). In this basin, the Chl peak has declined by −0.03 ± 0.08 mg m〈sup〉−3〈/sup〉 decade〈sup〉−1〈/sup〉 and the growing period duration has decreased by −12 ± 7 days decade〈sup〉−1〈/sup〉. The trends in phytoplankton Chl and phenology, estimated in this study over the period 1998–2014, do not reveal significant overall decline/increase in Chl concentration or earlier/delayed timings of the seasonal peak on average over the entire Mediterranean Sea basin. However, we observed large regional variations, suggesting that the response of phytoplankton to environmental and climate forcing may be complex and regionally driven.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Gang Shao, Scott C. Stark, Danilo R.A. de Almeida, Marielle N. Smith〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstracts〈/h5〉 〈div〉〈p〉Leaf area dynamics offer information about changes in forest biomass and canopy function critical to understanding the role of forests in the climate system and carbon cycle. Airborne small footprint lidar is a potential major source for the detection of variation in leaf area density (LAD), LAD vertical profiles, and total leaf area (leaf area index, LAI), from sites to regional scales. However, the sensitivities of lidar-based LAD and LAI estimation are not yet well known, particularly in dense forests, over landscape heterogeneity, sensor system, and survey differences, and through time. To address these questions, we compared 16 pairs of multitemporal airborne lidar surveys with four different laser sensors across six Amazon forest sites with resurvey intervals ranging from one to nine years. We tested whether the different laser sensors, and the pulse return density of laser sampling (variable between and within each survey) introduce systematic biases. Laser sensors created consistent biases that accounted for up to 18.20% of LAD differences between surveys, but biases could be corrected with a simple regression approach. Lidar pulse return density had little appreciable bias impact when above 20 returns per m〈sup〉2〈/sup〉. After correction, repeated mean and site maximum LAI estimates became significantly correlated (R〈sup〉2〈/sup〉 ~0.8), while LAD profiles revealed site differences. Heterogeneity and change in LAD structure were detectable at the ecologically relevant 1/4 ha forest neighborhood grid scale, as evidenced by the high correlation of profile variation between surveys, with the strength of correlation (R〈sup〉2〈/sup〉 value) significantly decreasing with increasing survey interval (0.74 to 0.16 from one to nine years), consistent with accumulating effects of forest dynamics. Sensor-induced biases trended towards correlation with lidar footprint (beam width). The LAD estimation and bias correction approach developed in this study provides the standardization critical for heterogeneous lidar networks that offer high throughput functional ecological monitoring of climatically important forests like the Amazon.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: January 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 220〈/p〉 〈p〉Author(s): David P.S. Bekaert, Cathleen E. Jones, Karen An, Mong-Han Huang〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The Sacramento-San Joaquin Delta, like other heavily engineered deltas throughout the world, has undergone substantial subsidence since the late 1800s when the natural estuary was leveed to form islands. Today, islands within the Delta have subsided to the point where most lie below mean sea level. Long-term sustainability of the Delta requires reversal of subsidence, but spatially comprehensive maps of subsidence have not been available to inform and monitor remediation. Reported here is the first spatially dense map of recent subsidence rates across the Delta, based on synthetic aperture radar interferometry and constrained by GNSS observations. The analysis uses a temporally dense and spatially overlapping set of data acquired by the Uninhabited Aerial Vehicle SAR (UAVSAR) sensor in 2009–2015. On average, the Delta is subsiding by 9.2 ± 4.4 mm/yr with high variability even at the sub-island scale, including peak subsidence of 160 ± 4.4 mm/yr occurring just inland of the levee of the westernmost island and average subsidence of 11 mm/yr along the planned path of water conveyance tunnels through the Delta. The results are compared to local sea level rise (SLR), measured at a tidal gauge in San Francisco Bay, to show that subsidence is currently the dominant contributor to local relative sea level rise (RSLR) by nearly a factor of five. Based on recently measured and modeled acceleration of SLR, subsidence in the Delta is likely to dominate RSLR until the middle of the 21th century.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: January 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 220〈/p〉 〈p〉Author(s): Brian B. Barnes, Jennifer P. Cannizzaro, David C. English, Chuanmin Hu〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Satellite ocean color datasets have vast potentials for assessing and monitoring of marine environments. However, with the MODIS sensor aging and the VIIRS sensor reaching maturity, it is important to continuously evaluate the quality of reflectance data from both instruments. Here, we critically assess the statistical performance of both MODIS and VIIRS, including analysis of two separate (and commonly used) VIIRS processing routines. In addition, we note variability in the literature as to the methods used to identify and remove low-quality data during similar validation exercises. Although most studies use some implementation of satellite quality flags (L2 flags) and many exclude data based on spatial heterogeneity or large temporal gap from satellite overpasses, critical assessment of these methods indicates variable performance. Indeed, we found little improvement in validation statistics after implementation of these data culling techniques, with substantial variability in effectiveness between wavebands and sensors. Overall, these findings highlight the need to critically assess the impact (on both data quantity and quality) of exclusion criteria, towards more effective techniques to ensure quality and consistency of satellite ocean color datasets.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Ce Zhang, Isabel Sargent, Xin Pan, Huapeng Li, Andy Gardiner, Jonathon Hare, Peter M. Atkinson〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Land cover (LC) and land use (LU) have commonly been classified separately from remotely sensed imagery, without considering the intrinsically hierarchical and nested relationships between them. In this paper, for the first time, a highly novel Joint Deep Learning framework is proposed and demonstrated for LC and LU classification. The proposed Joint Deep Learning (JDL) model incorporates a multilayer perceptron (MLP) and convolutional neural network (CNN), and is implemented via a Markov process involving iterative updating. In the JDL, LU classification conducted by the CNN is made conditional upon the LC probabilities predicted by the MLP. In turn, those LU probabilities together with the original imagery are re-used as inputs to the MLP to strengthen the spatial and spectral feature representations. This process of updating the MLP and CNN forms a joint distribution, where both LC and LU are classified simultaneously through iteration. The proposed JDL method provides a general framework within which the pixel-based MLP and the patch-based CNN provide mutually complementary information to each other, such that both are refined in the classification process through iteration. Given the well-known complexities associated with the classification of very fine spatial resolution (VFSR) imagery, the effectiveness of the proposed JDL was tested on aerial photography of two large urban and suburban areas in Great Britain (Southampton and Manchester). The JDL consistently demonstrated greatly increased accuracies with increasing iteration, not only for the LU classification, but for both the LC 〈em〉and〈/em〉 LU classifications, achieving by far the greatest accuracies for each at around 10 iterations. The average overall classification accuracies were 90.18% for LC and 87.92% for LU for the two study sites, far higher than the initial accuracies and consistently outperforming benchmark comparators (three each for LC and LU classification). This research, thus, represents the first attempt to unify the remote sensing classification of LC (state; what is there?) and LU (function; what is going on there?), where previously each had been considered separately only. It, thus, has the potential to transform the way that LC and LU classification is undertaken in future. Moreover, it paves the way to address effectively the complex tasks of classifying LC and LU from VFSR remotely sensed imagery via joint reinforcement, and in an automatic manner.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Alice César Fassoni-Andrade, Rodrigo Cauduro Dias de Paiva〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Water composition in floodplains plays a key role in ecological processes and is affected by both local water sources and flooding from the main river. Despite local studies, still lacks a complete understanding on the relationship between hydrological processes and sediment distribution in the river-floodplain system of the Amazon basin. This paper presents the first mapping of the dynamics of suspended sediments in rivers and lakes (〉0.25 km〈sup〉2〈/sup〉) of central Amazon, considering different water types. Previous studies have considered only white-water rivers with high sediment concentration. This study also describes some river-lake systems in terms of their spatial-temporal pattern of sediments and water flows. Time series between 2003 and 2017 of red and infrared reflectance of Moderate Resolution Imaging Spectroradiometer (MODIS) images were temporally filtered to retrieve incomplete and low-quality data. These images were assessed as a proxy of the surface suspended sediments concentration; and maps of the spatial-temporal variation of sediments were created, such as the class frequency map. This map represents a 15-year frequency at which each pixel remains in one of the surface suspended sediments concentration classes: high, moderate, and low, with an overall accuracy of 71%. Our findings allowed to observe the variation of sediment concentration along the Solimões-Amazonas River, such as, for instance, its increase from the confluence with the Tapajós River to the mouth. Some hydrological processes were also observed in lakes of the middle reach, such as overbank flow and resuspension of sediments in depression lakes. In some ria lakes, the main water source comes from local basin with the backwater promoting sediment input in these lakes during the low-water period. The maps produced by this study are available and useful for supporting biogeochemical studies.〈/p〉〈/div〉 〈/div〉 〈div xml:lang="en"〉 〈h5〉Graphical abstract〈/h5〉 〈div〉〈p〉〈figure〉〈img src="https://ars.els-cdn.com/content/image/1-s2.0-S0034425718305005-ga1.jpg" width="500" alt="Unlabelled Image" title="Unlabelled Image"〉〈/figure〉〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Jeffrey Pickering, Stephen V. Stehman, Alexandra Tyukavina, Peter Potapov, Pete Watt, Samuel M. Jantz, Pradeepa Bholanath, Matthew C. Hansen〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Guyana is a high forest cover, low deforestation country. Since 2011–2014 the Guyana Forestry Commission (GFC) has used visual interpretation of 5 m resolution RapidEye imagery to map forest loss and nearby degradation for the entire country. According to the GFC produced national map, 58% of all forest loss events cover less than 1 ha, so forest loss is very rare and occurs at a scale that can be difficult to detect and monitor. Nearly all (~97%) of the forest loss and nearby degradation is due to small-scale mining and its associated infrastructure. For any country wanting to accurately map and monitor forest loss, sample-based area estimation can be considerably less expensive than wall-to-wall mapping. To quantify the tradeoff between precision and cost, we evaluated the standard errors of area estimators for several sample-based strategies using Guyana as a case study. We partitioned Guyana into 374 blocks, with each block 24 km × 24 km to align with the image tiles provided by RapidEye's 3A product. The area of forest loss and area of degradation for each of the 374 blocks were determined from the GFC map to create the population data used to evaluate the precision of different sample-based strategies. We compared the standard errors of estimators of area of forest loss and area of degradation obtained from simple random, stratified random and systematic sampling as applied to these population data. To construct strata for the stratified design, we evaluated two forest loss maps produced from Landsat data, a 30 m global forest loss product and a 30 m national forest loss map produced specifically for Guyana. For both of these maps, several options for defining stratum boundaries and allocating sample size to strata were evaluated. All stratified design options reduced the standard error of the area estimators relative to simple random sampling. The Dalenius-Hodges and Jenks methods for choosing stratum boundaries yielded greater improvements in precision than the Equal Area method. For this Guyana case study, optimal and equal allocation of sample size to strata led to substantially better precision than proportional allocation. Only small improvements in precision were attained by increasing the number of strata from three to five. For simple random sampling, incorporating the global and national forest loss map information in a regression estimator reduced standard errors of estimated area of forest loss and area of degradation relative to the standard error of the Horvitz-Thompson estimator that does not incorporate the forest loss map information. However, the best performing stratified sampling design options had better precision than the regression estimator. RapidEye data cost 800 USD per ‘3A’ tile so the annual imagery cost for mapping Guyana would be 274,000 USD. For an investment of 75,200 USD (the cost of 94 RapidEye tiles, the largest sample size evaluated), the best performing stratified random sample achieved a relative standard error (100% times the standard error divided by area) of 7% for estimating area of forest loss and 6% for estimating area of degradation. This study demonstrates the utility of Landsat-based forest loss maps to provide effective strata to improve precision of area estimators produced from a sample of RapidEye imagery in a country for which forest loss and degradation are very rare features of the landscape.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Menghua Li, Lu Zhang, Xuguo Shi, Mingsheng Liao, Mengshi Yang〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Since 2009, the Guobu landslide has been very active, posing a safety threat to the Laxiwa Hydropower Station, located only several hundred meters downstream along the upper Yellow River in China. To investigate the current state of this landslide, we analyzed two stacks of X-band TerraSAR-X (TSX) High-resolution Spotlight (HS) mode images acquired from September 2015 to April 2017 in descending orbits with different look angles. A new time-series point-like target offset tracking (TS-PTOT) method is proposed to retrieve time-series surface displacements at point-like targets (PTs) from SAR image pairs properly combined with large temporal baselines and small spatial baselines. According to an evaluation of the standard deviation of time-series displacements at stable points, the TS-PTOT method increased the measurement precision of offset tracking by about 25% over the results using a single master. Our TSX-HS TS-PTOT results manifested a spatial pattern and magnitude of displacements highly similar to DInSAR result produced with an ALOS-2 PALSAR-2 image pair. The maximum displacement rate at the upper part of the slope during the study period was around 80 cm/yr in the line-of-sight (LOS) direction, which is much lower than the displacement rate measured in 2010. Furthermore, three-dimensional (3D) displacements at those identified homologous PTs were estimated by combining two-dimensional (2D) displacements measured by TS-PTOT from the two SAR data stacks. The 3D deformation pattern of the Guobu landslide properly verified its toppling-sliding deformation mechanism.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Yinan He, Gang Chen, Angela De Santis, Dar A. Roberts, Yuyu Zhou, Ross K. Meentemeyer〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Forest ecosystems are subject to recurring fires as one of their most significant disturbances. Accurate mapping of burn severity is crucial for post-fire land management and vegetation regeneration monitoring. Remote-sensing-based monitoring of burn severity faces new challenges when forests experience both fire and non-fire disturbances, which may change the biophysical and biochemical properties of trees in similar ways. In this study, we develop a Disturbance Weighting Analysis Model (DWAM) for accurately mapping burn severity in a forest landscape that is jointly affected by wildfire and an emerging infectious disease – sudden oak death. Our approach treats burn severity in each basic mapping unit (e.g., 30 m grid from a post-fire Landsat image) as a linear combination of burn severity of trees affected (diseased) and not affected by the disease (healthy), weighted by their areal fractions in the unit. DWAM is calibrated using two types of inputs: i) look-up tables (LUTs) linking burn severity and post-fire spectra for diseased and healthy trees, derived from field observations, hyperspectral sensors [e.g., Airborne Visible InfraRed Imaging Spectrometer (AVIRIS)], and radiative transfer models; and ii) pre-fire fractional maps of diseased and healthy trees, derived by decomposing a pre-fire Landsat image using Multiple Endmember Spectral Mixture Analysis (MESMA). Considering the presence of tree disease in DWAM improved the overall map accuracy by 42%. The superior performance is consistent across all three stages of disease progression. Our approach demonstrates the potential for improved mapping of forest burn severity by reducing the confounding effects of other biotic disturbances.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: Available online 15 November 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment〈/p〉 〈p〉Author(s): J.-B. Féret, G. le Maire, S. Jay, D. Berveiller, R. Bendoula, G. Hmimina, A. Cheraiet, J.C. Oliveira, F.J. Ponzoni, T. Solanki, F. de Boissieu, J. Chave, Y. Nouvellon, A. Porcar-Castell, C. Proisy, K. Soudani, J.-P. Gastellu-Etchegorry, M.-J. Lefèvre-Fonollosa〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉 〈p〉Leaf mass per area (〈em〉LMA〈/em〉) and leaf equivalent water thickness (〈em〉EWT〈/em〉) are key leaf functional traits providing information for many applications including ecosystem functioning modeling and fire risk management. In this paper, we investigate two common conclusions generally made for 〈em〉LMA〈/em〉 and 〈em〉EWT〈/em〉 estimation based on leaf optical properties in the near-infrared (NIR) and shortwave infrared (SWIR) domains: (1) physically-based approaches estimate 〈em〉EWT〈/em〉 accurately and 〈em〉LMA〈/em〉 poorly, while (2) statistically-based and machine learning (ML) methods provide accurate estimates of both 〈em〉LMA〈/em〉 and 〈em〉EWT〈/em〉.〈/p〉 〈p〉Using six experimental datasets including broadleaf species samples of 〉150 species collected over tropical, temperate and boreal ecosystems, we compared the performances of a physically-based method (PROSPECT model inversion) and a ML algorithm (support vector machine regression, SVM) to infer 〈em〉EWT〈/em〉 and 〈em〉LMA〈/em〉 based on leaf reflectance and transmittance. We assessed several merit functions to invert PROSPECT based on iterative optimization and investigated the spectral domain to be used for optimal estimation of 〈em〉LMA〈/em〉 and 〈em〉EWT〈/em〉. We also tested several strategies to select the training samples used by the SVM, in order to investigate the generalization ability of the derived regression models.〈/p〉 〈p〉We evidenced that using spectral information from 1700 to 2400 nm leads to strong improvement in the estimation of 〈em〉EWT〈/em〉 and 〈em〉LMA〈/em〉 when performing a PROSPECT inversion, decreasing the 〈em〉LMA〈/em〉 and 〈em〉EWT〈/em〉 estimation errors by 55% and 33%, respectively.〈/p〉 〈p〉The comparison of various sampling strategies for the training set used with SVM suggests that regression models show limited generalization ability, particularly when the regression model is applied on data fully independent from the training set. Finally, our results demonstrate that, when using an appropriate spectral domain, the PROSPECT inversion outperforms SVM trained with experimental data for the estimation of 〈em〉EWT〈/em〉 and 〈em〉LMA〈/em〉. Thus we recommend that estimation of 〈em〉LMA〈/em〉 and 〈em〉EWT〈/em〉 based on leaf optical properties should be physically-based using inversion of reflectance and transmittance measurements on the 1700 to 2400 nm spectral range.〈/p〉 〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Guglielmo Lacorata, Raffaele Corrado, Federico Falcini, Rosalia Santoleri〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We present an independent validation of the satellite surface velocity fields released by the ESA Data User Element GlobCurrent Project. The validation methodology is based on the analysis of Lagrangian numerical trajectories integrated from satellite-based sea surface currents. Two case studies were considered: Mediterranean Sea and North Atlantic Ocean.The Finite-Scale Lyapunov Exponents provide a rigorous, quantitative tool to evaluate Lagrangian simulations with respect to real drifter trajectories. An accuracy threshold scale can be identified as the scale above which the error size, propagating along a numerical trajectory, grows no faster than the relative separation between real drifters. Below this threshold, the error growth rate tends to diverge linearly as the error decreases. The mean error growth speed, at early stage, is found to be related to the kinetic energy of the missing scales of motion, not resolved by the GlobCurrent products. Established kinematic Lagrangian models are, also, exploited to compensate the energy gap between real and numerical trajectories in the unresolved scale range. Ultimately, GlobCurrent surface velocity fields are shown to have overall good ‘Lagrangian skills' for large-scale transport and dispersion numerical simulations.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Mahmud Haghshenas Haghighi, Mahdi Motagh〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Most of the developed groundwater basins in Iran are subject to land subsidence hazards resulting from the over-extraction of groundwater. Several areas in Tehran, the capital city and a provincial center in north-central Iran, have been reported to be subsiding at different rates. In this study, we present the results of an Interferometric Synthetic Aperture Radar (InSAR) time series analysis of Tehran using different SAR data between 2003 and 2017. By constructing more than 400 interferograms derived from 39 Envisat ASAR (C-band), 10 ALOS PALSAR (L-band), 48 TerraSAR-X (X-band), and 64 Sentinel-1 (C-band) SAR datasets, we compile displacement time series from interferometric observations using the Small Baseline (SB) technique. Our analysis identifies 3 distinct subsidence features in Tehran with rates exceeding 25 cm/yr in the western Tehran Plain, approximately 5 cm/yr in the immediate vicinity of Tehran international airport, and 22 cm/yr in the Varamin Plain to the southeast of Tehran city. The temporal pattern of land subsidence, which is dominated by a decreasing trend, generally follows the regional decline in groundwater level, which suggests that anthropogenic processes caused by excessive groundwater extraction are the primary cause of land subsidence. Integrating a decadal time series of subsidence constructed from multi-sensor InSAR with in-situ observations suggests that inelastic and permanent compaction dominates the main deformation regime of the Tehran aquifer, and the ratio between elastic and inelastic deformation is approximately 0.4. A geological analysis indicates that the shape of the subsidence bowl in the western Tehran Plain does not follow the trend of major mapped faults in the region. In contrast, the subsidence bowl in Varamin is controlled by the Pishva Fault, which suggests that either this fault acts as a hydrologic barrier to the groundwater flow in this region or that the differences in sediment thickness causes the discontinuity in land subsidence.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Alexander Osadchiev, Roman Sedakov〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉We use near simultaneous ocean color satellite imagery from NASA's Landsat 8 and ESA's Sentinel-2 missions to reconstruct surface currents along the northeastern shore of the Black Sea and study the spread of a small river plume formed in this area. Several times a year Landsat 8 and Sentinel-2 satellites both pass over the study area within a time interval of 〈10 min. Analysis of near simultaneous ocean color composites obtained during these periods provides an opportunity for accurate reconstruction of surface currents. It is especially efficient for detecting motion of frontal zones associated with river plumes which are visible in optical satellite imagery. Using an optical flow algorithm we reconstructed surface currents associated with the motion of a small plume formed by the Mzymta River in response to different wind forcing conditions. We show that the dynamics of the Mzymta plume are significantly different to those of large plumes. First, a near-field jet of the small plume is abruptly decelerated by friction with the subjacent ocean and does not form a recirculating bulge even under low wind forcing conditions. As a result, freshwater discharge does not accumulate near the river mouth and is transported to the far-field part of the plume. Second, under certain external forcing conditions the angle between wind direction and Ekman transport within the plume takes anomalously large values of 60–80°. As a result, onshore winds cause upstream accumulation of the river plume, while offshore/downwelling and upwelling winds result in downstream and offshore transport of freshwater discharge, respectively.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Pierre Defourny, Sophie Bontemps, Nicolas Bellemans, Cosmin Cara, Gérard Dedieu, Eric Guzzonato, Olivier Hagolle, Jordi Inglada, Laurentiu Nicola, Thierry Rabaute, Mickael Savinaud, Cosmin Udroiu, Silvia Valero, Agnès Bégué, Jean-François Dejoux, Abderrazak El Harti, Jamal Ezzahar, Nataliia Kussul, Kamal Labbassi, Valentine Lebourgeois〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The convergence of new EO data flows, new methodological developments and cloud computing infrastructure calls for a paradigm shift in operational agriculture monitoring. The Copernicus Sentinel-2 mission providing a systematic 5-day revisit cycle and free data access opens a completely new avenue for near real-time crop specific monitoring at parcel level over large countries. This research investigated the feasibility to propose methods and to develop an open source system able to generate, at national scale, cloud-free composites, dynamic cropland masks, crop type maps and vegetation status indicators suitable for most cropping systems. The so-called Sen2-Agri system automatically ingests and processes Sentinel-2 and Landsat 8 time series in a seamless way to derive these four products, thanks to streamlined processes based on machine learning algorithms and quality controlled 〈em〉in situ〈/em〉 data. It embeds a set of key principles proposed to address the new challenges arising from countrywide 10 m resolution agriculture monitoring. The full-scale demonstration of this system for three entire countries (Ukraine, Mali, South Africa) and five local sites distributed across the world was a major challenge met successfully despite the availability of only one Sentinel-2 satellite in orbit. 〈em〉In situ〈/em〉 data were collected for calibration and validation in a timely manner allowing the production of the four Sen2-Agri products over all the demonstration sites. The independent validation of the monthly cropland masks provided for most sites overall accuracy values higher than 90%, and already higher than 80% as early as the mid-season. The crop type maps depicting the 5 main crops for the considered study sites were also successfully validated: overall accuracy values higher than 80% and F1 Scores of the different crop type classes were most often higher than 0.65. These respective results pave the way for countrywide crop specific monitoring system at parcel level bridging the gap between parcel visits and national scale assessment. These full-scale demonstration results clearly highlight the operational agriculture monitoring capacity of the Sen2-Agri system to exploit in near real-time the observation acquired by the Sentinel-2 mission over very large areas. Scaling this open source system on cloud computing infrastructure becomes instrumental to support market transparency while building national monitoring capacity as requested by the AMIS and GEOGLAM G-20 initiatives.〈/p〉〈/div〉 〈/div〉 〈div xml:lang="en"〉 〈h5〉Graphical abstract〈/h5〉 〈div〉〈p〉〈figure〉〈img src="https://ars.els-cdn.com/content/image/1-s2.0-S0034425718305145-ga1.jpg" width="343" alt="Unlabelled Image" title="Unlabelled Image"〉〈/figure〉〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Yunxia Wang, Guy Ziv, Marcos Adami, Edward Mitchard, Sarah A. Batterman, Wolfgang Buermann, Beatriz Schwantes Marimon, Ben Hur Marimon Junior, Simone Matias Reis, Domingos Rodrigues, David Galbraith〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Tropical disturbed forests play an important role in global carbon sequestration due to their rapid post-disturbance biomass accumulation rates. However, the accurate estimation of the carbon sequestration capacity of disturbed forests is still challenging due to large uncertainties in their spatial distribution. Using Google Earth Engine (GEE), we developed a novel approach to map cumulative disturbed forest areas based on the 27-year time-series of Landsat surface reflectance imagery. This approach integrates single date features with temporal characteristics from six time-series trajectories (two Landsat shortwave infrared bands and four vegetation indices) using a random forest machine learning classification algorithm. We demonstrated the feasibility of this method to map disturbed forests in three different forest ecoregions (seasonal, moist and dry forest) in Mato Grosso, Brazil, and found that the overall mapping accuracy was high, ranging from 81.3% for moist forest to 86.1% for seasonal forest. According to our classification, dry forest ecoregion experienced the most severe disturbances with 41% of forests being disturbed by 2010, followed by seasonal forest and moist forest ecoregions. We further separated disturbed forests into degraded old-growth forests and post-deforestation regrowth forests based on an existing post-deforestation land use map (TerraClass) and found that the area of degraded old-growth forests was up to 62% larger than the extent of post-deforestation regrowth forests, with 18% of old-growth forests actually being degraded. Application of this new classification approach to other tropical areas will provide a better constraint on the spatial extent of disturbed forest areas in Tropics and ultimately towards a better understanding of their importance in the global carbon cycle.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: Available online 6 December 2018〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment〈/p〉 〈p〉Author(s): Peiqi Yang, Christiaan van der Tol, Wout Verhoef, Alexander Damm, Anke Schickling, Thorsten Kraska, Onno Muller, Uwe Rascher〈/p〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The growing availability of global measurements of sun-induced chlorophyll fluorescence (SIF) can help in improving crop monitoring, especially the monitoring of photosynthetic activity. However, variations in top-of-canopy (TOC) SIF cannot be directly interpreted as physiological changes because of the confounding effects of vegetation biochemistry (i.e. pigments, dry matter and water) and structure. In this study, we propose an approach of using radiative transfer models (RTMs) and TOC reflectance to estimate the biochemical and structural effects on TOC SIF, as a necessary step in retrieving physiological information from TOC SIF. The approach was assessed by using airborne (〈em〉HyPlant〈/em〉) reflectance and SIF data acquired over an agricultural experimental farm in Germany on two days, before and during a heat event in summer 2015 with maximum temperatures of 27°C and 34°C, respectively. The results show that over 76% variation among different crops in SIF observations was explained by variation in vegetation biochemistry and structure. In addition, the changes of vegetation biochemistry and structure explained as much as 73% variation between the two days in far-red SIF, and 40% variation in red SIF. The remaining unexplained variation was mostly attributed to the variability in physiological status. We conclude that reflectance provides valuable information to account for biochemical and structural effects on SIF and to advance analysis of SIF observations. The combination of RTMs, reflectance and SIF opens new pathways to detect vegetation biochemical, structural and physiological changes.〈/p〉〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Zhihui Wang, Philip A. Townsend, Anna K. Schweiger, John J. Couture, Aditya Singh, Sarah E. Hobbie, Jeannine Cavender-Bares〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Foliar functional traits refer to a range of biochemical and physiognomic characteristics of vegetation that control photosynthesis, nutrient and water cycling, and can be used to describe the functional diversity of ecosystems. In this study, we utilize NASA AVIRIS - Next Generation imaging spectroscopy data to map 15 foliar functional traits in a grassland experiment at the Cedar Creek Ecosystem Science Reserve, a Long-Term Ecological Research site in Central Minnesota, across three years. To estimate traits, we compared the widely used partial least squares regression (PLSR) with Gaussian processes regression (GPR) to assess differences in model performance and uncertainty estimates. PLSR is attractive for its straightforward implementation and interpretation, but requires bootstrapping-based methods to estimate and map prediction uncertainties. On the other hand, GPR is more complex to implement and interpret, but provides explicit estimates of uncertainties. Our results indicated that foliar functional traits can be retrieved in these grasslands with moderate to high accuracies using either method. Highest validation accuracies were obtained for leaf mass per area (LMA), soluble cell contents, hemicellulose and cellulose (all with 〈em〉R〈/em〉〈sup〉2〈/sup〉 〉 0.8), and lower accuracies for lignin, nitrogen, and some pigments with both techniques. The estimations for the mass-based pigments were more accurate than the area-based pigments (at least 5% improvement in normalized RMSE). Overall, GPR and PLSR performed comparably with respect to both skill of predictions and the selection of most informative spectral regions. Maps of uncertainties corresponded well between the two models, with highest uncertainties related to low vegetation cover, high diversity levels, or under irrigation and nitrogen treatments (not represented in the field sampling). The maps showed that trait values in each plot were relatively stable across three years of managed species richness. Our results provide a template for mapping foliar traits and their uncertainties in grasslands, and point to the need for extensive ground data across time to properly evaluate performance of trait mapping algorithms.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Guoqing Zhang, Tandong Yao, Wenfeng Chen, Guoxiong Zheng, C.K. Shum, Kun Yang, Shilong Piao, Yongwei Sheng, Shuang Yi, Junli Li, Catherine M. O'Reilly, Shuhua Qi, Samuel S.P. Shen, Hongbo Zhang, Yuanyuan Jia〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Lakes are sensitive indicators of anthropogenic climate change and also respond to direct human activities. Yet, long-term lake inventories and quantitative evaluation of the factors driving observed lake changes across China remain elusive. Here, for the first time, we examined multi-decadal lake area changes in China during 1960s–2015, using historical topographic maps and 〉3831 Landsat satellite images, including lakes as fine as ≥1 km〈sup〉2〈/sup〉 in size. In addition, we quantified the causes of lake changes from climatic and anthropogenic factors. The total area of lakes in China has increased by 5858.06 km〈sup〉2〈/sup〉 (9%) between 1960s and 2015, and with heterogeneous spatial variations. Lake area changes in the Tibetan Plateau, Xinjiang, and Northeast Plain and Mountain regions reveal significant increases of 5676.75, 1417.15, 1134.87 km〈sup〉2〈/sup〉 (≥15%), respectively, but the Inner-Mongolian Plateau shows an obvious decrease of 1223.76 km〈sup〉2〈/sup〉 (22%). We find that 141 new lakes have appeared predominantly in the arid western China; but 333 lakes, mainly located in the humid eastern China, have disappeared over the past five decades. We conclude that climate factors have played a dominant role in lake changes across China, coupled with noticeable anthropogenic contribution of ~35% in the Eastern Plain and Yunnan-Guizhou Plateau. This study has substantial implications to improve decision support regarding water-resource management strategies and land-use planning throughout China.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Jan U.H. Eitel, Andrew J. Maguire, Natalie Boelman, Lee A. Vierling, Kevin L. Griffin, Johanna Jensen, Troy S. Magney, Peter J. Mahoney, Arjan J.H. Meddens, Carlos Silva, Oliver Sonnentag〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Relatively little is known of how the world's largest vegetation transition zone – the Forest Tundra Ecotone (FTE) – is responding to climate change. Newly available, satellite-derived time-series of the photochemical reflectance index (PRI) across North America and Europe could provide new insights into the physiological response of evergreen trees to climate change by tracking changes in foliar pigment pools that have been linked to photosynthetic phenology. However, before implementing these data for such purpose at these evergreen dominated systems, it is important to increase our understanding of the fine scale mechanisms driving the connection between PRI and environmental conditions. The goal of this study is thus to gain a more mechanistic understanding of which environmental factors drive changes in PRI during late-season phenological transitions at the FTE – including factors that are susceptible to climate change (i.e., air- and soil-temperatures), and those that are not (photoperiod). We hypothesized that late-season phenological changes in foliar pigment pools captured by PRI are largely driven by photoperiod as opposed to less predictable drivers such as air temperature, complicating the utility of PRI time-series for understanding climate change effects on the FTE. Ground-based, time-series of PRI were acquired from individual trees in combination with meteorological variables and photoperiod information at six FTE sites in Alaska. A linear mixed-effects modeling approach was used to determine the significance (α = 0.001) and effect size (i.e., standardized slope b*) of environmental factors on late-seasonal changes in the PRI signal. Our results indicate that photoperiod had the strongest, significant effect on late-season changes in PRI (b* = 0.08, p 〈 0.001), but environmental variables susceptible to climate change were also significant (i.e., daily mean solar radiation (b* = −0.03, p 〈 0.001) and daily mean soil temperature (b* = 0.02, p 〈 0.001)). These results suggest that interpreting PRI time-series of late-season phenological transitions may indeed facilitate our understanding of how northern treeline responds to climate change.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Christoph W. Kent, Sue Grimmond, David Gatey, Kohin Hirano〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Urban morphology and aerodynamic roughness parameters are derived from three global digital elevation models (GDEM): Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), Shuttle Radar Topography Mission (SRTM), and TanDEM-X. Initially, each is compared to benchmark elevation data in London (UK). A moving window is used to extract ground heights from the GDEMs, generating terrain models with root-mean-square accuracy of up to 3 m. Subtraction of the ground heights extracted from each GDEM provides roughness-element heights only, allowing for calculation of morphology parameters. The parameters are calculated for eight directional sectors of 1 km grid-squares. Apparent merging of roughness elements in all GDEMs causes height-based parameter underestimation, whilst plan and frontal areas are over- and under-estimated, respectively. Combined, these lead to an underestimation of morphometrically-derived aerodynamic roughness parameters. Parameter errors are least for the TanDEM-X data. Further comparison in five cities (Auckland, Greater London, New York, Sao Paulo, Tokyo) provides basis for empirical corrections to TanDEM-X-derived geometric parameters. These reduce the error in parameters across the cities and for a separate location. Meteorological observations in central London give insight to wind-speed estimation accuracy using roughness parameters from the different elevation databases. The proposed corrections to TanDEM-X parameters lead to improved wind-speed estimates, which combined with the improved spatial representation of parameters across cities demonstrates their potential for use in future studies.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Temesgen Alemayehu Abera, Janne Heiskanen, Petri Pellikka, Miina Rautiainen, Eduardo Eiji Maeda〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Vegetation plays an important role in the climate system. The extent to which vegetation impacts climate through its structure and function varies across space and time, and it is also affected by land cover changes. In areas with both multiple growing periods and significant land cover changes, such as the Horn of Africa, identifying vegetation influence on land surface temperature (LST) through radiative changes needs further investigation. In this study, we used a 13-year time series (2001−2013) of remotely sensed environmental data to estimate the contribution of radiative mechanism to LST change due to growing season albedo dynamics and land cover conversion. Our results revealed that in taller woody vegetation (forest and savanna), albedo increases during the growing period by up to 0.04 compared with the non-growing period, while it decreases in shorter vegetation (grassland and shrubland) by up to 0.03. The warming impact due to a decrease in albedo during the growing period in shorter vegetation is counteracted by a considerable increase in evapotranspiration, leading to net cooling. Analysis of land cover change impact on albedo showed a regional annual average instantaneous surface radiative forcing of −0.03 ± 0.02 W m〈sup〉−2〈/sup〉. The land cover transitions from forest to cropland, and savanna to grassland, displayed the largest mean albedo increase across all seasons, causing an average instantaneous surface radiative forcing of −2.6 W m〈sup〉−2〈/sup〉 and − 1.5 W m〈sup〉−2〈/sup〉 and a decrease in mean LST of 0.12 K and 0.09 K, all in dry period (December, January, February), respectively. Despite the albedo cooling effect in these conversions, an average net warming of 1.3 K and 0.23 K was observed under the dominant influence of non-radiative mechanisms. These results show that the impact of radiative mechanism was small, highlighting the importance of non-radiative processes in understanding the climatic impacts of land cover changes, as well as in delineating effective mitigation strategies.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Huabing Huang, Caixia Liu, Xiaoyi Wang, Xiaolu Zhou, Peng Gong〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Quantification of forest aboveground biomass density (AGB) is useful in forest carbon cycle studies, biodiversity protection and climate-change mitigation actions. However, a finer resolution and spatially continuous forest AGB map is inaccessible at national level in China. In this study, we developed forest type- and ecozone-specific allometric models based on 1607 field plots. The allometric models were applied to Geoscience Laser Altimeter System (GLAS) data to calculate AGB at the footprint level. We then mapped a 30 m resolution national forest AGB by relating the GLAS footprint AGB to various variables derived from Landsat images and Phased Array L-band Synthetic Aperture Radar (PALSAR) data. We estimated the average forest AGB to be 69.88 Mg/ha with a standard deviation of 35.38 Mg/ha and the total AGB carbon stock to be 5.44 PgC in China. Our AGB estimates corresponded reasonably well with AGB inventories from the top ten provinces in the forested area, and the coefficient of determination and root mean square error were 0.73 and 20.65 Mg/ha, respectively. We found that the main uncertainties for AGB estimation could be attributed to errors in allometric models and in height measurements by the GLAS. We also found that Landsat-derived variables outperform PALSAR-derived variables and that the textural features of PALSAR better support forest AGB estimates than backscattered intensity.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Samuel Hislop, Simon Jones, Mariela Soto-Berelov, Andrew Skidmore, Andrew Haywood, Trung H. Nguyen〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉Time series analysis of Landsat data is widely used for assessing forest change at the large-area scale. Various change detection algorithms have been proposed, each employing different techniques to characterise abrupt disturbance events and longer term trends. However, results can vary significantly, depending on the algorithm, parameters and the spectral index (or indices) chosen. This mismatch in results has led to researchers hypothesizing that an ensemble based approach may increase accuracy. In this study we assess two change detection algorithms (LandTrendr and the R package strucchange), each with three indices (the Normalized Difference Vegetation Index or NDVI, the Normalized Burn Ratio or NBR, and Tasseled Cap Wetness or TCW). We test their ability to detect abrupt disturbances in sclerophyll forests over a 29 year time period, and subsequently evaluate a number of ensembles, using simple fusion rules and Random Forests models. A total of 4087 manually interpreted reference pixels, sampled from 9 million ha of forest, were used for training and validation. In addition, we assess the effects of priming the Random Forests classifier with confusing cases (commission errors from the time series algorithms). Our results clearly show that ensembles combining multiple change detection techniques out-perform any one method. Our most accurate Random Forests model, using an ensemble of all 6 algorithm outputs, along with 3 bi-temporal change rasters (change in NBR, NDVI and TCW), had an overall error rate of 7%, compared with the most accurate single algorithm/index approach (LandTrendr with NBR), which had an overall error of 21%. Our findings also indicate that acceptable results (14% error) can be achieved without the use of traditional change detection algorithms, by using robust reference data and Random Forests classification. However, by priming the classifier with confusing cases informed by the change detection algorithms, commission errors decreased substantially, at the expense of slight increases in omission errors. In fact, a Random Forests ensemble, using the primed training data and only 3 bi-temporal change rasters, was more accurate than any one individual algorithm, with an overall error of 11%. By including some additional metrics derived from Landsat time series (e.g. 2 year changes and overall means) this error was further reduced to 8%. Given that most change detection algorithms have large processing requirements, this suggests that algorithms can be applied to a sample of pixels only, for the sole purpose of training a machine learning classifier. We demonstrate the feasibility of this previously unexplored approach, by creating annual disturbance maps over a large area of forest (9 million ha) and long time period (29 years).〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: February 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 221〈/p〉 〈p〉Author(s): Ziti Jiao, Anxin Ding, Alexander Kokhanovsky, Crystal Schaaf, Francois-Marie Bréon, Yadong Dong, Zhuosen Wang, Yan Liu, Xiaoning Zhang, Siyang Yin, Lei Cui, Linlu Mei, Yaxuan Chang〈/p〉 〈div xml:lang="en"〉 〈h5〉Abstract〈/h5〉 〈div〉〈p〉The linear kernel-driven RossThick-LiSparseReciprocal (RTLSR) bidirectional reflectance distribution function (BRDF) model was originally developed from the simplified scenarios of continuous and discrete vegetation canopies, and has been widely used to fit multiangle observations of vegetation-soil systems of the land surface in many fields. Although this model was not developed explicitly for snow surfaces, it can capture the geometric-optical effect caused by the shadowing of rugged or undulating snow surfaces. However, in this study, this model has been further developed to better characterize the scattering properties of snow surface, which can also exhibit strongly forward-scattering behavior. This study proposes a new snow kernel to characterize the reflectance anisotropy of pure snow based on the asymptotic radiative transfer (ART) model that assumes snow can be modeled as a semi-infinite, plane-parallel, weakly absorbing light scattering layer. This new snow kernel adopts a correction term with a free parameter 〈em〉α〈/em〉 to correct the analytic form of the ART model that has been reported to underestimate observed snow reflectance in the forward-scattering direction in the principal plane (PP), particularly in cases of a large viewing zenith angle (〉60°). This snow kernel has now been implemented in the kernel-driven RTLSR BRDF model framework in conjunction with two additional kernels (i.e., the volumetric scattering kernel and geometric-optical scattering kernel) and is validated using observed and simulated multiangle data from three data sources. Pure snow targets were selected from the extensive archive of the Polarization and Directionality of the Earth's Reflectance (POLDER) BRDF data. Antarctic snow field measurements, which were taken from the top of a 32-m-tall tower at Dome C Station and include 6336 spectral bidirectional reflectance factors (BRFs), were also utilized. Finally, a set of simulated BRFs, generated by a hybrid scattering snow model that combines the geometric optics with vector radiative transfer theory, were used to further assess the proposed method. We first retrieve the value of the free parameter 〈em〉α〈/em〉 for a comprehensive analysis using single multiangle snow data with a sufficient BRDF sampling. Then, we determine the optimally fixed value of the 〈em〉α〈/em〉 parameter as prior information for potential users. The new snow kernel method is shown to be quite accurate, presenting a high correlation coefficient (〈em〉R〈/em〉〈sup〉2〈/sup〉 = ~0.9) and a negligible bias between the modeled BRFs and the various snow BRDF validation data. The finding demonstrates that this snow kernel provides an improved potential compared to that of the original kernel-driven model framework for a pure snow surface in many applications, particularly those involving the global water cycle and radiation budget, where snow cover plays an important role.〈/p〉〈/div〉 〈/div〉

Beschreibung: 〈p〉Publication date: January 2019〈/p〉 〈p〉〈b〉Source:〈/b〉 Remote Sensing of Environment, Volume 220〈/p〉 〈p〉Author(s): 〈/p〉