Temporal changes in spatial variability of plant available water at the watershed scale
Abstract
Identification of soil moisture distribution patterns by remote-sensing at the basin scale has become a major challenge for variable source hydrology. On time scales of a few days, water uptake from plants and evaporation can change significantly as a result of soil moisture dynamics, while on a time scales of a few weeks vegetation dynamics may represent a strong relationship with soil moisture. This study explores the relationship between vegetation dynamics and soil water content/ subsurface storage. Multi-temporal, multi-spectral Remotely Sensed Landsat images are used to identify spatial differences and temporal changes of vegetative cover over a subbasin of the Cannonsville reservoir, in the Catskills mountains region of New York state. Vegetation indices are processed and compared for six months (April,May June, July, September and October) in 2001. For each month, three classes of vegetation indices were determined from the frequency distribution of indices over the study area. The histogram of the vegetation indices revealed hypothetical Gaussians corresponding with generic land uses (forest, open grass/shrublands, pasture/crops and plowed land), and were well correlated with land uses estimated by from other sources. Comparison from one month to another of the actual position in the landscape of these three index classes led to the identification of different zones sharing the same index distribution. These zones were also seen to follow the temporal growth curve characteristic of its particular vegetation types. The spatial variations patterns of vegetation indices within each land use zone were then compared with the patterns of soil moisture distribution, as output by a fully distributed hydrological model, SMDR.
- Publication:
-
AGU Spring Meeting Abstracts
- Pub Date:
- May 2004
- Bibcode:
- 2004AGUSM.H54A..05D
- Keywords:
-
- 1640 Remote sensing;
- 1866 Soil moisture