Time Series Analysis of Temperature and Rainfall in the Savannah Region in Togo, West Africa

This study investigates the trend in monthly and annual rainfall, and minimum and maximum temperature (Tmin and Tmax) in the Savannah region of Togo. The historical data of Mango and Dapaong weather stations from 1981 to 2019 were used. A serial correlation test was applied to all time series to identify serially independent series. A Non-parametric Mann-Kendall (MK) test was applied to serially independent series. The magnitude of the trend was calculated using the Sen's slope (SS) method. For the data influenced by serial correlation, a modified version of the Mann-Kendall test was applied. An open-source library package was developed in the R language, namely, mkmodified. For annual rainfall, results showed a significant increasing trend at Dapaong (p < 0.05) and a non-significant decreasing trend at Mango (p > 0.05) at 95%. There was an increasing trend in the Tmin both at Mango and Dapaong. No statistically significant trend was found at Mango (p > 0.05), while at Dapaong (p < 0.05), a significant trend was found at 95%. Simlarly, there was a statistically increasing trend in the Tmax both at Mango and Dapaong. Rainfall in Dapaong has increased (5.50 mm/year) whereas in Mango, it has decreased (-0.93 mm/year). Tmn increased by 0.04 and 0.008 degrees C per year in Mango and Dapaong, respectively. Tmax increased by 0.03 and 0.02 degrees C per year in Mango and Dapaong, respectively. A Rainfall Anomaly Index (RAI) was also used to present a temporal variation in rainfall; the historical series presented drier years. Many studies have analysed the trend of climate parameters in northern Togo, but none of them has specifically targeted the Savannah region that is considered the poorest region of the country.

Automated summary

This study examines the changes in rainfall and temperature in the Savannah region of Togo. Historical weather data from Mango and Dapaong stations between 1981 and 2019 were analyzed. The Mann-Kendall test and Sen's slope method were used to identify trends in the data. Results show an increasing trend in annual rainfall at Dapaong and a non-significant decreasing trend at Mango. Both Tmin and Tmax exhibit an increasing trend at both stations. The study highlights the need for specific research on the Savannah region, which is the poorest in the country.