Land Use Land Cover Classification and Wheat Yield Prediction in the Lower Chenab Canal System Using Remote Sensing and Gis

Abstract

Reliable and timely information regarding area under wheat and its yield prediction can help in better management of the commodity. The remotely sensed data especially in combination with Geographic Information System (GIS) can provide an important and powerful tool for both, land use land cover (LULC) classification and crop yield prediction. The study objectives include LULC classification and wheat yield prediction. The study was conducted for Rabi Season from Nov. 2011 to April 2012, in the command area of three distributaries i.e. Khurrian Wala, Killian Wala and Mungi of Lower Chennai Canal (LCC) system. The Landsat-7 imagery data with spatial resolution of 30 m was used for this study. Physical features were monitored and assessed using Normalized Difference Vegetative Index (NDVI). LULC classification was done for wheat and non-wheat area which shows wheat proportion and area 87.22% and 28867.95 Ha in Khurrian wala, 71.07% and 22423.20 Ha in Killian Wala and 79.18% and 17974.34 Ha in Mungi distributary, respectively. The correlation values between maximum NDVI value and yield data were 0.45, 0.36 and 0.39 for Khurrian Wala, Killian Wala and Mungi distributary, respectively. On the basis of this correlation, average wheat yield was estimated as 3.48 T/Ha, 3.83 T/Ha and 3.80 T/Ha for Khurrian Wala, Killian Wala and Mungi distributary, respectively.

Wheat Yield Prediction based on Sentinel-2, Regression and Machine Learning Models in Hamedan, Iran

“…As a solution, numerous previous studies have shown the contribution of satellite imagery to wheat yield estimation at larger regions, since satellite images provide precise and continuous information on the phenological status. In this regard, various satellite sensors, e.g., Advanced Very High Resolution Radiometer (AVHRR) [8], Moderate Resolution Imaging Spectroradiometer (MODIS) [1,2,8,9], Sentinel-2, Landsat 8 [2], Landsat TM [10,11], IRS-LISS III [12], Indian geostationary satellite INSAT 3A CCD and IRS (Indian Remote Sensing Satellite) [13], Huan Jing (HJ) satellite HJ1A/B and Landsat 8 Operational Land Imager (OLI) [14], etc., have been used for crop yield prediction in the literature. The most widely used satellite sensors for crop yield prediction provide low spatial resolution and high temporal resolution to capture crop phenological development.…”

“…VIs refers to the values which are often computed from reflectance or radiance of specific bands of satellite images, mostly in the visible and near-infrared bands. Accordingly, researches have been carried out on using indices, such as NDVI [2,11,17], accumulated NDVI [9], adjusted maximum NDVI (MA-NDVI) [1], peak NDVI [4], soil adjusted vegetation index (SAVI) [2], modified SAVI (MSAVI) [2], enhanced vegetation index (EVI) [2], normalized difference drought index (NDDI), normalized difference water index (NDWI), vegetation condition index (VCI), temperature condition index (TCI), vegetation health index (VHI), normalized multi-band drought index (NMDI), visible and shortwave infrared drought index (VSDI), and vegetation supply water index (VSWI) [18] to predict crop yields. Two bands offered by Sentinel 2 from the NIR range, B8 and B8A, can be used for the calculation of the NDVI index.…”