Estimation of LAI across phenological stages of wheat using google earth engine

Abstract

The Leaf Area Index (LAI) is a measure of photosynthesis and transpiration, and it has become the common currency for agro-climatic researchers. The non-destructive technique of LAI estimation using remote sensing has immense potential. The challenge lies in estimating LAI at the field scale for implementing research results for crop management using Google Earth Engine (GEE) integrated with Python. Sentinel-2A datasets empowered by high spatial, spectral, and temporal resolution over an arid region of southwest Punjab, India were used to estimate LAI at field and district level. Wheat LAI was estimated for two consecutive years, 2016–2017 and 2017–2018. The comprehensive data analysis approach comprised of processing and estimation of LAI, designed for four significant phenological stages followed by validation with in situ field observed LAI collected from the experimental plots as well as with the Moderate Resolution Imaging Spectroradiometer (MODIS)’s LAI data products. The results showed a strong positive co-relationship between observed field LAI and Sentinel-2A estimated LAI as 0.64 and 0.47, with MODIS dataset as 0.24 and 0.19 for both years. Therefore, it can be concluded that field-level LAI can be estimated from Sentinal-2A satellite images with moderate accuracy by agricultural specialists and practitioners.