 |
| Titel |
Estimating irrigated areas from satellite and model soil moisture data over the contiguous US |
| VerfasserIn |
Felix Zaussinger, Wouter Dorigo, Alexander Gruber |
| Konferenz |
EGU General Assembly 2017
|
| Medientyp |
Artikel
|
| Sprache |
en
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 19 (2017) |
| Datensatznummer |
250143942
|
| Publikation (Nr.) |
 EGU/EGU2017-7713.pdf |
|
|
|
|
|
| Zusammenfassung |
| Information about irrigation is crucial for a number of applications such as drought- and
yield management and contributes to a better understanding of the water-cycle,
land-atmosphere interactions as well as climate projections. Currently, irrigation is
mainly quantified by national agricultural statistics, which do not include spatial
information. The digital Global Map of Irrigated Areas (GMIA) has been the first effort
to quantify irrigation at the global scale by merging these statistics with remote
sensing data. Also, the MODIS-Irrigated Agriculture Dataset (MirAD-US) was
created by merging annual peak MODIS-NDVI with US county level irrigation
statistics.
In this study we aim to map irrigated areas by confronting time series of various satellite
soil moisture products with soil moisture from the ERA-Interim/Land reanalysis product. We
follow the assumption that irrigation signals are not modelled in the reanalysis product, nor
contributing to its forcing data, but affecting the spatially continuous remote sensing
observations. Based on this assumption, spatial patterns of irrigation are derived from
differences between the temporal slopes of the modelled and remotely sensed time series
during the irrigation season.
Results show that a combination of ASCAT and ERA-Interim/Land show spatial patterns
which are in good agreement with the MIrAD-US, particularly within the Mississippi Delta,
Texas and eastern Nebraska. In contrast, AMSRE shows weak agreements, plausibly due to a
higher vegetation dependency of the soil moisture signal. There is no significant agreement to
the MIrAD-US in California, which is possibly related to higher crop-diversity and lower
field sizes. Also, a strong signal in the region of the Great Corn Belt is observed, which is
generally not outlined as an irrigated area. It is not yet clear to what extent the
signal obtained in the Mississippi Delta is related to re-reflection effects caused by
standing water due to flood or furrow irrigation practices. Consequently, future
research should focus on the specific effects of different irrigation practices and crop
types.
This study is supported by the European Union’s FP7 EartH2Observe “Global Earth
Observation for Integrated Water Resource Assessment” project (grant agreement number
331 603608). |
|
|
|
|
|
|