Author
 |
Starks, Patrick |
|
Jackson, Thomas |
|
Submitted to: Journal of Range Management
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 12/30/2001 Publication Date: 9/1/2002 Citation: N/A Interpretive Summary: An objective of a major multi-agency field campaign entitled the Southern Great Plains 1997 Hydrology Experiment was to examine the feasibility of estimating water content of the soil by combining remotely sensed data, field measurements and modeling techniques. To address this objective, field data were used to evaluate the ability of a simple two-layer soil water budget model to simulate surface and root zone soil water content fo the climate, soil and vegetative conditions at four tallgrass prairie sites in central and south central Oklahoma. In an analogous manner, remotely sensed estimates of soil water content were used to initialize the model and the output was compared to measured values. A small bias in modeled soil water content was observed at two sites having sandy soils when field data were used to initialize the model. However, modeled soil water content compared well to measurements overall. When remotely sensed data were used to initialize the model, modeled soil water content at two sites were very different from measured values. One of these two sites had a dense litter layer covering the soil surface and the other had a large number of trees in the area which caused an underestimation of remotely sensed soil water contents used to initialize the model. Results from the evaluation indicate that the model produces reasonable estimates of surface and root zone water content when given good initialization values of soil water content. The model's simplicity, coupled with its overall good performance, makes it suitable for remote sensing studies of soil water content. Technical Abstract: A two-layer soil water budget model was evaluated for use in remote sensing studies of volumetric soil water content at four tallgrass prairie sites located in central and south central Oklahoma. Field measurements and remotely sensed estimates of antecedent soil water content were used separately to initialize the model. Although the model exhibited a small bias at two sites with sandy soils, modeled soil water content compared well to measurements when the model was initialized with field data. At two of the four sites, modeled soil water content underestimated measured values when remotely sensed estimates of antecedent soil water content was used to initialize the model. One of these two sites had a dense layer of litter on the soil surface while the second site had a large number of trees in the vicinity, which probably caused an underestimation of remotely sensed soil water content. Modeled soil water content at the remaining two osites compared well to measured data. The evaluation indicated that the model produced reasonable estimates of surface and root zone water content when given good initialization values of antecedent soil water content. The model's simplicity, coupled with its generally acceptable performance, makes it suitable for implementation in remote sensing studies of spatial and temporal dynamics of soil water content. |
