Time Series Trend Analysis of Temperature and Rainfall Using Mann-Kendall Method: A Study of Vadodara City, Gujarat, India
Janak P Joshi *
VSCDL Vadodara, India.
Bindu Bhatt
b Department of Geography, Faculty of Science, The Maharaja Sayajirao University of Baroda, Vadodara Gujarat, India.
*Author to whom correspondence should be addressed.
Abstract
Aims: This study aimed to quantify the variability of monthly temperature and precipitation patterns on a local scale within Vadodara city, Gujarat, India.
Study Design and Data: A retrospective analysis was employed, utilizing historical weather data encompassing a 37-year period from 1981 to 2017 for Vadodara city. The data included monthly minimum and maximum temperatures (TMIN and TMAX) alongside monthly precipitation totals.
Methodology: Non-parametric statistical techniques were implemented to analyze the trends within the temperature and precipitation data. The Mann-Kendall (MK) test was employed to identify statistically significant trends, while Sen's slope estimator was utilized to quantify the magnitude of any trends detected.
Results: The analysis revealed a possible increasing trend in minimum temperature records over the study period, with a positive correlation coefficient (R²) of 0.04. Regarding precipitation, a trend towards increasing rainfall was observed in the month of July (R² = 0.03), while June exhibited a trend towards decreasing rainfall (R² = 0.008).
Keywords: Time series, trend, mann-kendall, sen’s slope, temperature and rainfall