ALBERT

Description: Permafrost is an essential climate variable (ECV). It is a subsurface phenomenon which cannot be directly measured with remotely sensed data. However, many parameters which influence the ground thermal regime and surface indicators can be captured with satellite data. Those are e.g. land surface temperature, land cover and snow parameters, soil moisture and terrain changes. The capabilities of currently available remotely sensed datasets need to be assessed with respect to operational monitoring and their value enhanced by a synergistic use. Within the ESA DUE Permafrost project a wide range of EO datasets are investigated and integrated in an information system with extensive involvement of the permafrost research community. This comprises pan-boreal/arctic to regional and local scale investigations. A number of relevant global datasets do already exist or are currently in development. Land surface temperature is available from MODIS and AATSR on global scale. ESA projects such as GlobCover, GlobCarbon and GlobSnow provide important land cover parameters. An operational soil moisture monitoring system is currently provided by EUMETSAT from MetOp ASCAT. The majority of these datasets has been developed and validated outside of the circumpolar permafrost zone. The assessment of these datasets is thus an integral part of the project. The information system will comprise those global datasets and also regional and local scale monitoring results. The latter cover sites in Alaska, Mackenzie basin, Southern Yakutia and Laptev and East Siberian Sea region. The database setup considers use by permafrost modelers and scientists working on local scale (hydrology, geomorphology, botany etc.). A combination with a WebGIS portal will allow efficient dissemination of results and free access. In this paper we present the current design of the information system.

Description: Time series of a wide range of biogeophysical parameters from satellite data are available to date on a global scale. A few initiatives focus on their improvement and validation in high latitudes. For example the DUE Permafrost and STSE ALANIS-Methane, which are activities funded by the European Space Agency, focus on this issue. ALANIS Methane is a research project to produce and use a suite of relevant earth observation (EO) derived information to validate and improve one of the next generation land-surface models and thus reduce current uncertainties in wetland-related CH4 emissions. The task of the ESA DUE Permafrost project is to build up an Earth observation service for high-latitudinal permafrost applications. Results which are shown in this paper contribute to both. Microwave sensors are of special interest in this context due to their independence on cloud conditions and illumination of the Earth Surface. They can be used for derivation of land surface temperature, snow properties and land surface hydrology. The latter includes near surface soil moisture and inundation. Such parameters are of importance for studies on e.g. permafrost and land-atmosphere exchange. Datasets derived from active microwave instruments operating in C-band have been analysed with respect to their usability at high latitudes. Several examples from western Siberia are discussed.

Description: The ESA DUE Permafrost project (2009-2011) is developing a suite of parameters indicative of the subsurface phenomenon permafrost using satellite remote sensing: Land Surface Temperature (LST), Surface Soil Moisture (SSM), Surface Frozen and Thawed State (Freeze/Thaw), Terrain, Land Cover (LC), and Surface Water (SW). Snow parameters (Snow Extent and Snow Water Equivalent) are being developed through the DUE GlobSnow project, Global Snow Monitoring for Climate Research (2008-2011). The final DUE Permafrost remote sensing products cover the years 2007 to 2011 with a circumpolar coverage that will soon be released (early 2012), and then be used to analyze the temporal dynamics and map the spatial patterns of indicators. Further information is available at www.ipf.tuwien.ac.at/ permafrost. Since the beginning, scientific stakeholders and the International Permafrost Association (IPA) have been involved in the science and implementation plan. Interactive international user workshops took place in 2010 at the Technical University of Vienna, Vienna (AT), and in 2011 at the International Arctic Research Center (IARC), Fairbanks, Alaska (US). This involvement and the ongoing evaluation of the indicators derived from remote sensing for the high-latitude permafrost regions make the DUE Permafrost products trustworthy for the permafrost and the climate research community. The adaption of the remote sensing products for the permafrost and climate modelling is experimental and highly dependent on the users’ involvement. For a few years already, the Geophysical Institute Permafrost Laboratory (GIPL), University of Alaska Fairbanks, US, (http://www.gi.alaska.edu/research/snowicepermafrost/Permafrost) has successfully demonstrated the value of using LST derived from remote sensing data for driving its permafrost models. Further experimental testing of the DUE Permafrost products for use by the modeling community (permafrost and climate) will range from (i) the evaluation of external data of the models, with modifying or providing new external data (e.g. tundra land cover, surface water ratio, soil distribution), to (ii) new drivers for regional models derived from remote sensing (e.g., LST), to (iii) the evaluation of the output data from the models (e.g. spatial patterns of moisture and temperature).

Description: The focus of this research has been on detecting changes in lakes vegetation, land surface temperatures, and snow cover, using data from remote sensing. The study area covers the main (central) part of the Lena River catchment in the Yakutia Region of Siberia (Russia) where continuous permafrost coverage’s is up to 90%. The remote sensing analyses are based on MODIS (NASA) and Landsat (USGS) satellite data. Time series of remote sensing products of MODIS land surface temperature were produced for the study region between 61°N and 65°N, and between 117.5°E and 131.5°E. The MODIS Land Surface Temperature level 3 product, MOD11C3 are configured on a 0.05° latitude/longitude MODIS Climate Model Grid (CMG) raster. The LST product is a monthly composited average and represents clear-sky LST values. The monthly land surface temperature were analyzed over the eleven year interval from May 2000 to April 2011. Linear trend calculations for the 11 year temperature measurement interval were performed separately for each two month interval, in each pixel, using the least squares method. Water bodies were extracted using the Landsat Short Wave Infrared SWIR band 5. Within the study region's 315,000 sq. km, the total area covered by lakes increased by 17.5% between 2002 and 2009. The amount of lake increase differs between 42-11% depending on the region. The overall trend in land surface temperature is around 0.15°C/year, but with seasonal warming trends in April-May of up to 0.45°C/year in some areas and cooling of -0.2 to -0.3°C/year in July-August in other areas. These regional differences and potential causes of the land surface temperature changes will be discussed with respect to land cover changes.

Description: The task of the ESA DUE Permafrost project is to build up an Earth observation service for high-latitudinal permafrost applications with extensive involvement of the permafrost research community. The DUE Permafrost products derived from remote sensing are land surface temperature (LST), surface soil moisture (SSM), surface frozen and thawed state (freeze/ thaw), terrain, land cover, and surface waters. Weekly and monthly averages for most of the DUE Permafrost products will be made available for the years 2007-2010. The DUE Permafrost products are provided for the circumpolar permafrost area (north of 55°N) with 25 km spatial resolution. In addition, regional products with higher spatial resolution (300-1000 m/ pixel) were developed for five case study regions. These regions are: (1) the Laptev Sea and Eastern Siberian Sea Region (RU, continuous very cold permafrost/ tundra), (2) the Yakutsk Region (RU, continuous cold permafrost/ taiga), (3) the Western Siberian transect including Yamal Peninsula and Ob Region (RU, continuous to discontinuous/ taiga-tundra), (4) the Alaska Highway Transect (US, continuous to discontinuous/ taiga-tundra), and (5) the Mackenzie Delta and Valley Transect (CA, continuous to discontinuous/ taiga-tundra). The challenge of the programme is to adapt remote sensing products that are well established and tested in agricultural low and mid-latitudinal areas for highly heterogeneous taiga/ tundra permafrost landscapes in arctic regions. Ground data is essential for the evaluation of DUE Permafrost products and is provided by user groups and global networks. A major part of the DUE Permafrost core user group is contributing to GTN-P, the Global Terrestrial Network of Permafrost. Its main programmes, the Circumpolar Active Layer Monitoring (CALM) and the Thermal State of Permafrost (TSP) have been thoroughly overhauled during the last International Polar Year (2007-2008). Their spatial coverage has been extended to provide a true circumpolar network. Ground data ranges from active layer- and snow depths, to air-, ground-, and borehole temperature data as well as soil moisture measurements and the description of landform and vegetation. The GTN-P sites, with their position in different permafrost zones in the DUE Permafrost case study regions, are highly suitable for the evaluation of DUE Permafrost remote sensing products. Air and surface temperatures with high-temporal resolution are available for three GTN-P sites in Siberia and compared to LST products. Daily average GTN-P borehole- and air temperature data for 22 sites in Alaska and 6 sites in Western Siberia were used to validate surface frozen and thawed state. The preliminary results are promising. In addition, landform and vegetation descriptions of circumpolar GTN-P sites are used for the evaluation of global ‘landcover’ remote-sensing datasets like GlobeCover, Landcover2000 and EcoClimap – global datasets used as input for climate modeling.