ALBERT

Beschreibung: Global food security is critical to eliminating hunger and malnutrition. In the changing climate, farmers in developing countries must adopt technologies and farming practices such as precision agriculture (PA). PA-based approaches enable farmers to cope with frequent and intensified droughts and heatwaves, optimising yields, increasing efficiencies, and reducing operational costs. Biophysical parameters such as Leaf Area Index (LAI), Leaf Chlorophyll Content (LCab), and Canopy Chlorophyll Content (CCC) are essential for characterising field-level spatial variability and thus are necessary for enabling variable rate application technologies, precision irrigation, and crop monitoring. Moreover, robust machine learning algorithms offer prospects for improving the estimation of biophysical parameters due to their capability to deal with non-linear data, small samples, and noisy variables. This study compared the predictive performance of sparse Partial Least Squares (sPLS), Random Forest (RF), and Gradient Boosting Machines (GBM) for estimating LAI, LCab, and CCC with Sentinel-2 imagery in Bothaville, South Africa and identified, using variable importance measures, the most influential bands for estimating crop biophysical parameters. The results showed that RF was superior in estimating all three biophysical parameters, followed by GBM which was better in estimating LAI and CCC, but not LCab, where sPLS was relatively better. Since all biophysical parameters could be achieved with RF, it can be considered a good contender for operationalisation. Overall, the findings in this study are significant for future biophysical product development using RF to reduce reliance on many algorithms for specific parameters, thus facilitating the rapid extraction of actionable information to support PA and crop monitoring activities.

Beschreibung: The Yangtze River Delta (YRD) is one of the regions with the most intensive human activities. The eutrophication of lakes in this area is becoming increasingly serious with consequent negative impacts on the water supply of the surrounding cities. But the spatial-temporal characteristics and driving factors of the trophic state of the lake in this region are still not clearly addressed. In this study, a semi-analytical algorithm for estimating the trophic index (TSI) using particle absorption at 645 nm based on MODIS images is proposed to monitor and evaluate the trophic state of 41 large lakes (larger than 10 km2) in the YRD from 2002 to 2020. The performance of the proposed algorithm is evaluated using an independent dataset. Results showed that the root-mean-square error (RMSE) of the algorithm is less than 6 and the mean absolute percentage error (MAPE) does not exceed 8%, indicating that it can be applied for remotely deriving the TSI in the YRD. The spatial-temporal patterns revealed that there were significantly more lakes with moderate eutrophication in the Lower Yangtze River (LYR) than in the Lower Huaihe River (LHR). The overall average value of the TSI reaches a maximum in summer and a minimum in winter. The TSI value in the YRD over the period 2002–2020 showed a downward trend, especially after 2013. Individually, 33 lakes showed a downward trend and 8 lakes showed an upward trend. Furthermore, marked seasonal and interannual temporal variations can be clearly observed in the LYR and LHR and the sum of the variance contributions of seasonal and interannual components is more than 50%. Multiple linear regression analysis showed that human activities can explain 65% of the variation in the lake TSI in the YRD.

Beschreibung: Modeling forest fire spread is a very complex problem, and the existing models usually need some input parameters which are hard to get. How to predict the time series of forest fire spread rate based on passed series may be a key problem to break through the current technical bottleneck. In the process of forest fire spreading, spread rate and wind speed would affect each other. In this paper, three kinds of network models based on Long Short-Term Memory (LSTM) are designed to predict fire spread rate, exploring the interaction between fire and wind. In order to train these LSTM-based models and validate their effectiveness of prediction, several outdoor combustion experiments are designed and carried out. Process data sets of forest fire spreading are collected with an infrared camera mounted on a UAV, and wind data sets are recorded using a anemometer simultaneously. According to the close relationship between wind and fire, three progressive LSTM based models are constructed, which are called CSG-LSTM, MDG-LSTM and FNU-LSTM, respectively. A Cross-Entropy Loss equation is employed to measure the model training quality, and then prediction accuracy is computed and analyzed by comparing with the true fire spread rate and wind speed. According to the performance of training and prediction stage, FNU-LSTM is determined as the best model for the general case. The advantage of FNU-LSTM is further demonstrated by doing comparison experiments with the normal LSTM and other LSTM based models which predict both fire spread rate and wind speed separately. The experiment has also demonstrated the ability of the model to the real fire prediction on the basis of two historical wildland fires.

Beschreibung: In the Alps, understanding how climate change is affecting evapotranspiration (ET) is relevant due to possible implications on water availability for large lowland areas of Europe. Here, changes in ET were studied based on 20 years of MODIS data. MOD16 and operational Simplified Surface Energy Balance (SSEBop) products were compared with eddy-covariance data and analyzed for trend detection. The two products showed a similar relationship with ground observations, with RMSE between 0.69 and 2 mm day−1, and a correlation coefficient between 0.6 and 0.83. A regression with the potential drivers of ET showed that, for climate variables, ground data were coherent with MOD16 at grassland sites, where r2 was 0.12 for potential ET, 0.17 for precipitation, and 0.57 for air temperature, whereas ground data agreed with SSEBop at forest sites, with an r2 of 0.46 for precipitation, no correlation with temperature, and negative correlation with potential ET. Interestingly, ground-based correlation corresponded to SSEBop for leaf area index (LAI), while it matched with MOD16 for land surface temperature (LST). Through the trend analysis, both MOD16 and SSEBop revealed positive trends in the south-west, and negative trends in the south and north-east. Moreover, in summer, positive trends prevailed at high elevations for grasslands and forests, while negative trends dominated at low elevations for croplands and grasslands. However, the Alpine area share with positive ET trends was 16.6% for MOD16 and 3.9% for SSEBop, while the share with negative trends was 1.2% for MOD16 and 15.3% for SSEBop. A regression between trends in ET and in climate variables, LST, and LAI indicated consistency, especially between ET, temperature, and LAI increase, but low correlation. Overall, the discrepancies in the trends, and the fact that none of the two products outperformed the other when compared to ground data, suggest that, in the Alps, SSEBop and MOD16 might not be accurate enough to be a robust basis to study ET changes.

Beschreibung: In the northeastern United States, widespread deforestation occurred during the 17–19th centuries as a result of Euro-American agricultural activity. In the late 19th and early 20th centuries, much of this agricultural landscape was reforested as the region experienced industrialization and farmland became abandoned. Many previous studies have addressed these landscape changes, but the primary method for estimating the amount and distribution of cleared and forested land during this time period has been using archival records. This study estimates areas of cleared and forested land using historical land use features extracted from airborne LiDAR data and compares these estimates to those from 19th century archival maps and agricultural census records for several towns in Massachusetts, a state in the northeastern United States. Results expand on previous studies in adjacent areas, and demonstrate that features representative of historical deforestation identified in LiDAR data can be reliably used as a proxy to estimate the spatial extents and area of cleared and forested land in Massachusetts and elsewhere in the northeastern United States. Results also demonstrate limitations to this methodology which can be mitigated through an understanding of the surficial geology of the region as well as sources of error in archival materials.

Beschreibung: Urban greenspace provides essential benefits and often depends on its distribution and spatial relationship with residents. Many cities set ambitious goals to increase the coverage of greenspace. In addition, to increase the total amount of greenspace, spatial patterns of greenspace supply and demand also need to be taken into account to make sure its ecosystem services can reach the residents. While previous research has examined greenspace distribution, its association with various ecosystem services, and its spatial relationship with residents’ socioeconomic characteristics, relatively few studies have considered the spatial pattern of greenspace demand to assess its supply change over time. To fill this gap, we evaluated the greenspace change of Beijing between 2005 and 2015 using 2.5 m and 0.5 m high resolution remote sensing images. We first identified all of the greenspace changes, then evaluated the improvement of greenspace that was accessible to residents, and finally, we examined whether such improvement met different levels of demand estimated by neighborhood population, age structure, and economic status. The results showed a net increase of 1472 ha (7.8%) from 2005 to 2015. On average, percent greenspace within 500 m of the neighborhood boundary increased from 21% to 24%. Areas with low greenspace supply had a significantly higher increase. The standard deviation reduced from 8% to 7%, which indicated a smaller disparity of accessible greenspace. However, results showed that greenspace increase had little variation among neighborhoods with different demand levels. Our findings indicated that the greening efforts improved spatial distribution and reduced inequality in accessibility but failed to address different demand levels among neighborhoods. Furthermore, we identified neighborhoods with low supply/high demand and that lost greenspace between 2005–2015. These neighborhoods need to be given attention in future greening projects.

Beschreibung: The detection and characterisation of the radar Bright Band (BB) are essential for many applications of weather radar quantitative precipitation estimates, such as heavy rainfall surveillance, hydrological modelling or numerical weather prediction data assimilation. This study presents a new technique to detect the radar BB levels (top, peak and bottom) for Doppler radar spectral moments from the vertically pointing radars applied here to a K-band radar, the MRR-Pro (Micro Rain Radar). The methodology includes signal and noise detection and dealiasing schemes to provide realistic vertical Doppler velocities of precipitating hydrometeors, subsequent calculation of Doppler moments and associated parameters and BB detection and characterisation. Retrieved BB properties are compared with the melting level provided by the MRR-Pro manufacturer software and also with the 0 °C levels for both dry-bulb temperature (freezing level) and wet-bulb temperature from co-located radio soundings in 39 days. In addition, a co-located Parsivel disdrometer is used to analyse the equivalent reflectivity of the lowest radar height bins confirming consistent results of the new signal and noise detection scheme. The processing methodology is coded in a Python program called RaProM-Pro which is freely available in the GitHub repository.

Beschreibung: In dealing with the problem of target detection in high-resolution Synthetic Aperture Radar (SAR) images, segmenting before detecting is the most commonly used approach. After the image is segmented by the superpixel method, the segmented area is usually a mixture of target and background, but the existing regional feature model does not take this into account, and cannot accurately reflect the features of the SAR image. Therefore, we propose a target detection method based on iterative outliers and recursive saliency depth. At first, we use the conditional entropy to model the features of the superpixel region, which is more in line with the actual SAR image features. Then, through iterative anomaly detection, we achieve effective background selection and detection threshold design. After that, recursing saliency depth is used to enhance the effective outliers and suppress the background false alarm to realize the correction of superpixel saliency value. Finally, the local graph model is used to optimize the detection results. Compared with Constant False Alarm Rate (CFAR) and Weighted Information Entropy (WIE) methods, the results show that our method has better performance and is more in line with the actual situation.

Beschreibung: Point cloud classification plays a significant role in Light Detection and Ranging (LiDAR) applications. However, most available multi-scale feature learning networks for large-scale 3D LiDAR point cloud classification tasks are time-consuming. In this paper, an efficient deep neural architecture denoted as Point Expanded Multi-scale Convolutional Network (PEMCNet) is developed to accurately classify the 3D LiDAR point cloud. Different from traditional networks for point cloud processing, PEMCNet includes successive Point Expanded Grouping (PEG) units and Absolute and Relative Spatial Embedding (ARSE) units for representative point feature learning. The PEG unit enables us to progressively increase the receptive field for each observed point and aggregate the feature of a point cloud at different scales but without increasing computation. The ARSE unit following the PEG unit furthermore realizes representative encoding of points relationship, which effectively preserves the geometric details between points. We evaluate our method on both public datasets (the Urban Semantic 3D (US3D) dataset and Semantic3D benchmark dataset) and our new collected Unmanned Aerial Vehicle (UAV) based LiDAR point cloud data of the campus of Guangdong University of Technology. In comparison with four available state-of-the-art methods, our methods ranked first place regarding both efficiency and accuracy. It was observed on the public datasets that with a 2% increase in classification accuracy, over 26% improvement of efficiency was achieved at the same time compared to the second efficient method. Its potential value is also tested on the newly collected point cloud data with over 91% of classification accuracy and 154 ms of processing time.

Beschreibung: This article presents the methodology for an improved estimation of the sea surface wind speed measured by the cyclone global navigation satellite system (CYGNSS) constellation of satellites using significant wave height (SWH) information as external reference data. The methodology consists of a correcting 2D look-up table (LUT) with inputs: (1) the CYGNSS wind speed given by the geophysical model function (GMF); and (2) the collocated reference SWH given by the WW3 model, which is forced by winds from the European Centre for Medium-Range Weather Forecasts (ECMWF) organization. In particular, the analyzed CYGNSS wind speeds are the fully developed seas (FDS) obtained with the GMF 3.0, and the forcing winds are the ECMWF forecast winds. Results show an increase in sensitivity to large winds speeds and an overall reduction in the root mean square difference (RMSD) with respect to the ECMWF winds from 2.05 m/s to 1.74 m/s. The possible influence of the ECWMF winds on the corrected winds (due to their use in the WW3 model) is analyzed by considering the correlation between: (1) the difference between the ECMWF winds and those from another reference; and (2) the difference between the corrected CYGNSS winds and those from the same reference. Results using ASCAT, WindSat, Jason, and AltiKa as references show no significant influence.