Correlating the field water balance derived crop coefficient (Kc) and canopy reflectance-based NDVI for irrigated sugarcane

Abstract

Studies of crop coefficients as a function of vegetation indices are reported less often for sugarcane than other crops because of the variation in crop evapotranspiration (ETc) and spectral response over the long growth period. In this study, the possibility of correlating crop coefficient (Kc) and ground based normalized difference vegetation index (NDVI) of a sugarcane crop was investigated based on 2 years field experiments conducted in 2015 and 2016 in semi-arid India. The Kc values for the full crop season were determined by the field water balance method and ground NDVI was estimated from spectral reflectance measurements using a field spectro-radiometer. The sugarcane Kc values for the tillering (development stage), grand growth (mid-season) and maturity stages (end season) were 0.70, 1.20 and 0.78, respectively. The results found that the Kc was 16.6% less than that suggested by FAO-56. The sugarcane NDVI ranged from 0.48 to 0.69 at the tillering stage and 0.69 to 0.93 in the grand growth stage. Unlike other crops, sugarcane NDVI at the maturity stage did not reduce from 0.85 even at harvest due to the continued production of fresh green leaves at the top of the plant. Regression equations were developed to estimate the seasonal distribution of Kc with NDVI as the dependent variables and ratio of days $$\left(\frac{t}{T}\right)$$ after planting (t) to the total crop period (T) as the independent variable. The relationship between crop Kc and NDVI was characterized with 2nd order polynomial regression but correlation was moderately strong (r = 0.75, n = 315). Stronger correlations between Kc and NDVI were obtained by splitting growth period into the growth phase (r = 0.98, n = 245) and decline phase (r = 0.99, n = 70). The estimated Kc will be helpful for correcting irrigation scheduling of sugarcane in semi-arid conditions. The Kc-NDVI relationships for sugarcane investigated in this study are important for potential real time irrigation water management in the future.