|TSM/TDM Science Team Meeting 2016|
|Science Team Meetings|
|How to Submit a Proposal|
|COFUR Price List (scientific use)|
|TanDEM-X Science Service System|
|Title||Grassland management and biomass retrieval in an intensive dairy farm in Ireland using TerraSAR-X Staring Spotlight mode data.|
|Investigator||Barrett, Brian - University of Glasgow, School of Geography and Earth Sciences|
Monitoring grasslands is important for studies of carbon dynamics, conservation and agricultural management. For regions with persistent cloud cover, the use of multi-temporal spaceborne synthetic aperture radar (SAR) data offers an attractive solution for providing up-to-date information on grasslands over large and remote areas. In Ireland, grassland is the dominant land cover, occupying approximately 60% of the country`s terrestrial area and accounts for the majority of agricultural production. Grass growth is variable and it is often difficult to accurately forecast feed supply. XHR TerraSAR-X data has the potential to improve the monitoring of grasslands and this has particular importance for Ireland, given the expected trends for intensification of grassland management through implementation of the Food Harvest 2020 strategy (DAFM, 2011) by the Irish Government and the abolition of milk quotas across the EU-27 in 2015 (after which a significant expansion of the Irish dairy industry is foreseen).These transitions must also be considered in terms of the effect on carbon sequestration and emission processes linked with different grassland management strategies and also the significant implications for biodiversity that will arise.
The objectives of this study are to:
A Teagasc (Irish Agriculture and Food Development Authority) research farm (Moorepark) located near Cork, Ireland will be the selected study site for this analysis. This site (~100ha) is ideal in the context of advancing our understanding on the potential application of TerraSAR-X ST mode data for grasslands monitoring and biomass retrieval. A total of 24 TerraSAR-X ST and 8 dual-polarisation SpotLight (SL) acquisitions are requested between July 2014 and July 2015. This will allow for a thorough investigation of the SAR signal sensitivity to the phenological development of the grasslands and disturbance activities (e.g. grazing, mowing) over a full growing season. By analysing the backscatter signature of grasslands under different (controlled) grazing intensities and management practices, it is envisaged that the results will demonstrate the potential of using SAR data for routine biomass estimation in improved grasslands in the future.The SAR datasets will be co-registered, multi-looked, speckle filtered and geometrically and radiometrically calibrated and converted to decibel (dB) units for analysis with ground measurements. Machine-learning algorithms such as Support Vector Machines (SVM)(Vapnik, 1995) Random Forest (RF)(Breiman, 2001) and Extra Trees (Geurts et al., 2006) will be explored for accurately distinguishing grassland conditions. SAR-derived regression models will be compared to existing grass growth models. The project deliverables will include a SAR-derived regression model for biomass estimation, and temporal classification maps for discriminating between pasture management types.
The proposed research will be jointly funded under an existing Irish Environmental Protection Agency (EPA) research award (duration: 2014-2016) and the Teagasc Walsh Fellowship program (duration: 2013-2016).
Back to list of proposals
|© DLR 2004-2016||