Includes bibliographical references.

Nitrogen is a limiting macro-nutrient essential for crop growth but its excessive amount critically influences ecosystem biodiversity. The precise knowledge of the amount of available nitrogen (Avl. N) in soil under distinct crop species and topographical settings allows farmers to decide the fertilization amount required at different locations on the farm. Leaf nitrogen concentration (LNC) checks the assimilation and translocation of available nitrogen to leaves, the photosynthetically active part of crops. Since their conventional estimation in the laboratory happens to be laborious, expensive and time-consuming, this study uses geospatial technique and chlorophyll meter reflectance data as a rapid and real-time method of nitrogen estimation. A total of 121 composite soil and leaf samples were collected, each composite sample at 0-15 cm depth was prepared from 10 random samples from 10m X 10m of ginger cultivated areas under varying slope, aspect and elevation classes. SPAD (Soil Plant Analysis Development) readings were also taken before extracting the respective leaf samples. The sites were recorded using a handheld Global Positioning System (GARMIN GPSMAP-64). The samples were analyzed in the laboratory using the standard protocol. A cloud-free (< 10 %) Sentinel -2 image was accessed from the official website of Copernicus Open Access Hub. Models were developed for nitrogen prediction in soils and leaves of ginger. The result shows a strong correlation between Avl. N with NDVI (r = 0.81) and LNC with NDVI, TVI and OSAVI (r = 0.83). Other indices like MSAVI, SAVI, EVI, RDVI, GNDVI and GRVI were also found to be highly correlated (r > 0.7), whereas WDVI was moderately correlated (r = 0.3 – 0.7) with Avl. N and LNC. The Partial Least Square Regression (PLSR) (R2 = 0.72, RMSE = 0.114) model was found to be superior in performance which is followed by Principal Component Regression (PCR) (R2 = 0.69, RMSE = 0.140) and Stepwise- Multiple Linear Regression (Stepwise-MLR) (R2 = 0.67, RMSE = 0.154). Further, the SPAD chlorophyll meter has provided a better quadratic (R2 = 0.82) and linear (R2 = 0.79) regression model than individual spectral indices for LNC prediction in ginger. It is concluded that the use of spectral indices and SPAD meters has the potential to provide an indication of LNC against lab methods which are time-consuming and expensive. Avl. N has also provided a good prediction model with spectral indices and therefore can be used as an alternative for laboratory analysis.