Deriving the spatial distribution of irrigated-croplands in texas using MODIS satellite imageries
Conference Paper
Overview
Identity
Additional Document Info
View All
Overview
abstract
Spatial distribution of irrigated areas and their temporal variations are a prerequisite for effective planning, management, and resource conservation. The estimation of irrigated agriculture water-use depends on several factors including crops, cropped acreage, and climatic conditions. Crop types and acreage data are based on the reported information by the producers. The data reporting process might be time-consuming and inconsistent causing delay in producing the required annual water-use summary reports. Moreover, it does not provide the spatial pattern of the irrigated areas for any administrative unit. The ultimate goal of this project is to develop and automate unbiased and consistent method of using satellite imageries and image processing techniques to derive and quantify irrigated agricultural areas in Texas. This paper describes preliminary analysis, wherein a few methods, including both supervised and unsupervised classification, were explored to classify irrigated and non-irrigated winter wheat in a selected County in Texas panhandle. Classification using 8-day composite MODIS product is compared with the one conducted using daily time series of MODIS and NDVI signatures for various crops are also presented. Based on the NDVI curves, it is apparent that there is a distinguishable difference between irrigated and non-irrigated wheat. Non-irrigated wheat NDVI typically ranges from 0.2 to 0.4 while irrigated wheat ranges from 0.3 to 0.8. Using a group of pixels as ROIs instead of single pixel produced better classification. More detailed information about irrigation practices in the county would be helpful to validate the classification methodology. It will be important to understand all the variables that affect NDVI values for specific crops.
