A Review on Dry and Wet Spell Probability Analysis for Agricultural Crop Planning by using Markov Chain Model

In a predominantly agricultural system, natural rainfall is the primary source of water for agricultural sector. Agricultural crop output during the rainy season is determined by the regional and temporal distribution of rainfall, as well as its intensity and duration. But with alteration in world’sclimate, temperature and rainfall will increases in some places and in other places this situation is inverse. Due to these climatic changes as almost, all crops are season-dependent as well as rainfall-dependent. Climate change is exerting profound effects on global agriculture, disrupting traditional cropping patterns and posing significant challenges to food security. The intricate interplay of increasing temperatures, altered rainfall patterns, and increased frequency of extreme weather events has far-reaching consequences for crop production systems worldwide. Changes in precipitation patterns, including shifts in timing and intensity, significantly impact water availability for crops. Irregular rainfall distribution can lead to droughts or waterlogging, both of which adversely affect crop growth. Regions experiencing increased aridity face challenges in maintaining adequate soil moisture for optimal plant growth. Agricultural productivity in rainfed areas remains low and unstable due to weather anomalies, with major crops affected by monsoon delays, variations in spatial and temporal aberrations, and breaks in the monsoon causing prolonged dry spells and being responsible for early, mid, and terminal drought. These circumstances call for attention of agricultural scientists and planners can develop contingent measures to save the rainfed crops from varied monsoon aberrations.

A scientific approach to agriculture is crucial for optimizing a region'srainfall patterns and maintaining consistent crop output levels. Analyzing dry and rainy spells helps build a crop plan for rainfed areas. Using scientific forecasting to analyze wet and dry spells can help farmers improve their crops and economic conditions (Bora et al., 2022). Knowing when dry and wet seasons occur is crucial for agricultural planning and farming operations since it has a considerable influence on crop output and, as a result, rural populations' lives. The successor failure of any crop, especially in rainfed areas, is influenced not only by total rainfall received during the crop season, but also by how it is distributed during the critical phonologic time of crop growth. The concept of forecasting the likelihood of dry and wet periods, as well as the possibility of consecutive 2/3 days of dry and wet spells based on a threshold amount of rainfall, is very useful for crop planning and agricultural operations. Understanding the occurrence of dry spells and wet spells is critical for mitigating the negative consequences of dry spells during sensitive crop growth phases. The probable behavior of rainfall was studied by many researchers (Kumar et al., 2007); (Chakravorthy and Mandal.,2008); (Chand et al., 2011); and (Jakhar et al., 2011).

The probability concept is commonly used to highlight the significance of dry and wet spells when planning weather-sensitive agricultural activities. (Shrivastav et al., 2004). The purpose of estimating probability with respect to a given amount of rainfall is tremendously helpful for agricultural planning (Subash et al., 2009). Markov chain model has been found suitable to describe the long-term frequency behavior of wet or dry spells. (Victor and Sastri 1979). A Markov Chain Model considers less than 20 mm of rainfall in a week to be a dry week, and 20 mm or more as a wet week. Pandarinath (1991), Dash and Senapati (1992). The Markov chain probability model predicts whether rain will fall on a given week based on the previous week'sweather conditions. (Manikandan et al., 2017). A Markov chain process'sprobability relationship must incorporate the conditional probabilities that the process will transition from any state at period (t) to any subsequence state at period (t+1). As a result, the relation P(Xt+1 = J / Xt = i) represents the conditional probability of transitioning from statei to statej at timet+1. The likelihood of a week being dry or wet is determined by initial probability, however in conditional probability, if a particular periodi is wet or dry, the possibility of the (i+k)thperiod being wet is predicted and expressed as wet/wet or wet/dry. The threshold limits of 10 mm and 20 m of rainfall were chosen as crucial for various agricultural planning purposes. The weekly analysis of rainfall is particularly significant in agricultural planning, and the week period has been deemed the optimal length of time. (Reddy et al., 2008). In general, a Markov process describes only step-by-step dependence, called a first order process. The erratic nature of rainfall has been the primary reason of droughts in India. Drought causes significant yield reductions both for rainfed and irrigated crops. On the area basis, the dry spell analysis would assist in formulation of contingency plan against drought. Knowing the average monthly, seasonal, and yearly rainfall is beneficial in comprehending the overall picture of a certain location, but weekly rainfall data analysis provides more relevant and accurate information for rainfall-based crop planning (Tiwari et al., 1992). Another factor essential for crop planning is the forward and backward accumulation of rainfall, which determines the start and withdrawal of the effective monsoon. The onset and withdrawal features of the monsoon heavily influence the success of rainfed agriculture. The late beginning of the monsoon slows agricultural sowing, resulting in poor yields. Similarly, early removal of rainfall reduces yield owing to severe moisture stress, particularly when kharif crops are in crucial growth phases of grain formation and development (Dixit et al., 2005).