Data Mining is an analytic process designed to explore data or big data in search of consistent patterns or systematic relationships between variables, and then to validate the findings by applying the detected patterns to new subsets of data .In this study use explorative data analysis (EDA) of data mining techniques that include basics statistics, time series & probability distribution for the study of dry spell in Marathwada region. This region situated between 170 -35 N and 200- 40 N latitude and 740 - 40 E and 780 – 15 E longitudes in Maharashtra state. In this region consist of eight districts such as, Beed, Hingoli, Jalna, Latur, Nanded, Osmanabad and Parbhani. The monthly dry spell data of 55 years (1960 to 2015) was used for study. The primary objectives of study firstly to investigate the general trends of dry spells, secondly to fitting the probability distribution &estimate the return period of dry spell. The result show that all selected metrological station of Marathwada region was increasing trend on basis of result of Mann-Kendall Test in this test statistics value is positive & p-value is less than the significant level alpha (0.05). Discrete probability distribution such as, Geometric , Neg. Binomial, D. Uniform and possion was used for study. Kolmogorov–Smirnov test (K-S) of goodness fit test was used to select best fit probability distribution on the basis of minimum value of test statistic. The result shows that Aurangabad, Beed, Latur, Parabhani & Hingoli it was fit possion distribution, Osmanabad & Nanded it was fit Dis.Unifrom distribution, Jalna it was fit Neg.Binmoial distribution. Maximum likelihood method was used to estimate the parameters of fitted probability distributions. Regarding estimation of return period of dry spell was used to fitting probability distribution and their parameter such as estimation of expected monthly 10 dry day in Aurangabad station by using passion probability distribution was very high (68%) occurs at every one year and 20 dry day was only 1.5 % of occurs of every 66.47 years .This result is also helpful to prediction and estimation of rainfall for farmers and agriculture department of the Government of Maharashtra for planning cropping pattern of the Marathwada region.
