Article Preview
Top1. Introduction
The agricultural sector shares an important part in the progress of the Indian economy and shares approx. 17% share in the country’s GDP (Kumar, et al., 2021). Due to the dynamic nature of the Indian climate, the agricultural sector becomes more and more sensitive. Some crucial factors such as landholdings, soil nature, massive use of pest & fertilizers, and unpredicted weather conditions, are exposed more agricultural sector in India (Bushra & Sharma, 2019). Due to climate changes increase in the frequency of droughts & floods that result in substantial damages in crop production. This will lead to more challenges and threatens in food security in the countries like India (Patnaik, 1996). Research shows that climate changes like change in temperature or variation in precipitation plays important role in damaging crops. The variability in rainfall especially during summer monsoons shows a major impact on agricultural productivity which directly affects the country’s economy (Kumar & Gautam, 2016). India's agricultural sector has a large impact by receiving 70% of total rainfall by months of summer monsoon i.e. June to September. Previous studies clearly show that monsoon rainfall with long-term variability over Indian and its subdivisions also observed a significant trend for India annual rainfall/southwest monsoon. The local and global phenomena like northern hemisphere temperature, snow cover, and sea surface temperatures directly affected the monsoon rainfall (Wu, Chau, & Fan, 2010). The food crops like wheat, maize, rice, pulses, groundnut, and sugarcane, etc are mostly rain-dependent crops that would be directly affected by the early and late monsoon onset and from the variability of southwest monsoon rainfall (Rajeevan, Pai, Kumar, & Lal, 2007). Kharif crops are monsoon crops like maize and rice. Climate variability has been reported as major documented effects in wide-ranging land regions over agricultural areas like crop growth & development, crop production, and agricultural water resource in the world.
The researchers are continuously exploring the relationship between crop and climate, statistics of regional geographical data, and other considerable expected field studies showing the harvesting of the rice and wheat productivity approach based on various simulation models, focusing on the Indian subcontinental (Chmielewski & Potts, 1995; Xiong, Holman, Conway, Lin, & Li, 2008). And especially in the Uttarakhand (India) region, due to the increasing division of land, farmers are also taking agriculture as an impossible way to achieve food security. In this region, the crop productivity is low when compared to other regions due to natural circumstances such as the high erosion rates, constant threat of landslides, and soil erosion in the rainy season. In Uttarakhand, a flat area (up to 2400 m), about 80% of agricultural yielding dependent on the rainy season (Saxena, et al., 2021). Regular crop rotation and the following methods help to maintain the variability depending on irrigation conditions, height, soil type, humidity, location information, and the method and degree of inclination.
Much research has been done on this feature of powerful and flexible computer models to predict rainfall production in the Uttarakhand region (Basistha, Arya, & Goel, 2008). Therefore, the purpose of the study is to create a link between crop production shifts and summer rainfall variations over the Indian subcontinent to predict long distances. In this contribution, DRN best suited to predicting rainfall is proposed to monitor the situation with a small set of climate limits data collected in Dehradun, India. The presented approach attempted for predictation of yielding for the higher crop, stabilization and decline in migration, and increased employment, in the Uttarakhand region through data transmission methods (Chen, et al., 2020). Comparative testing of DRN with existing parameters and non-parametric methods was performed to confirm the validity of DRN in rainfall forecasts (Hernández, Sanchez-Anguix, Julian, Palanca, & Duque, Rainfall prediction: A deep learning approach, 2016).