Inventory and Phenological Assessment of Apple Orchards Using Various Remote Sensing Techniques for Shopian District of Jammu and Kashmir State, India

Abstract

Horticultural crops are one of the vital role players in the field of economic development in Indian hilly regions. Among all, apple is one of the important horticulture crop in the North-West Himalaya, and 60% of apple production comes from Kashmir state in India. The identification and monitoring of apple becomes essential for proper assessment and planning for large production of apple. Now-a-days, Remote Sensing and GIS have been emerging as the newest and more advanced technique for crop monitoring, crop inventory and the assessment of its production. Keeping in view, the present work deals with the identification of apple-producing areas and assessment of phenology patterns of apple trees in Shopian district of Jammu and Kashmir which is a leading producer of apple in the whole world. The IRS-P6 LISS-IV, Cartosat-1 and LANDSAT-8 data were used for phenology and classification. Top of Atmospheric (TOA) correction and various digital classification algorithms viz., supervised MXL, Unsupervised ISODATA and NDVI threshold were used for estimation of the apple orchard area and NDVI profile for phenology pattern across the years. In comparison to other classification techniques, Supervised classification technique was found most suited with having accuracy ranging between 90 and 95% for all stages of apple trees. On the basis of spectral response and NDVI response, apple orchards were distinguishable amidst the other crops/vegetation types. Generated normalized difference vegetation index value was found to be 0.65 for apple orchards for LISS-IV and the results are used for finding the phenology of apple tree. Using the above NDVI threshold value, the apple orchards are classified with an accuracy range of 75–80%. It shows the unique spectral and phenology feature of apple and highlighted the seasonal change. This method is also very useful for other orchard inventory and estimation of productivity. Besides, this also provides a trend in any abnormal seasonal change on horticulture area which may help in providing pre-information about health and effect on overall production of orchard.KeywordsRemote sensingApple orchards phenologysupervised and unsupervised classificationNDVI