MAPPING CROPS FROM THEIR TEMPORAL BEHAVIOR BY INTEGRATING SENTINEL-2 SATELLITE AND GROUND DATA

Abstract

Monitoring of crop is important for food security and to improve agricultural productivity. Crop monitoring is also important to understand that how changing climate is affecting the cropping pattern.  The ability of remote sensing to provide increased spatial granularity across the globe extensively helped in mapping crop types and for the estimation of biophysical parameters. The launch of Sentinel satellite (Copernicus programme) gives a unique opportunity to monitor crops systematically every 5 to 10 days. The development of operational crop monitoring methods involves an understanding of the temporal variations in spectral properties of each crop type. This study aims to analyze the temporal trajectory of vegetation indices for the identification of a variety of crops including winter and summer crops (wheat, mustard, carrot and alfalfa) and their health during the season in Bannu District. Sentinel-2 satellite images have been acquired from October 2017 to April 2018. Phenological profiles of each crop were derived based on the vegetation health index including NDVI and crop calendar. These phenological profiles along with environmental conditions were interpreted physically and mapped at 10-meter resolution. The results of the crop area and its health obtained are much promising and highlight the efficiency of the adopted techniques in delivering quick, cost-effective, traceable and reliable data compared to the conventional crop data collection methods.

Index Terms— Crop Type Mapping, crop monitoring phenology, Sentinel-2