Dry and Wet Spell Monthly Frequency and Severity in the Jos Plateau, Nigeria: Insights From Mann-kendall Trend Analysis, Hidden Markov Models, and Climatic Covariate Assessment (1995–2025)
Year: 2026
Type: article
Abstract: Understanding how often, how long, and how severe dry and wet spells are is key for managing climate risks, especially for farming and water resources on the Jos Plateau. In this study, we looked at monthly patterns of dry and wet spells over 31 years (1995–2025; 372 months) using several methods. These included classifying spells by monthly percentiles, calculating Dry Spell Severity Index (DSSI) and Wet Spell Severity Index (WSSI) from rainfall data, running Mann-Kendall trend analysis with Sen slope, and applying a four-state Gaussian Hidden Markov Model (HMM) that considered rainfall, temperature, and relative humidity together. Over the study period, there were 68 dry spell events and 63 wet spell events. On average, dry spells lasted 2.53 months, with the longest lasting 7 months (January–July 1995). Wet spells averaged 1.30 months, with a maximum of 3 months. The number of dry spells per year showed a slight downward trend (MK Z = −1.955, p = 0.051; Sen slope = −0.044 month per year), while wet spell frequency stayed about the same (p = 0.773). Mean temperature rose significantly (+0.0163°C per year; p = 0.014), but relative humidity did not change much. DSSI was significantly linked to relative humidity in February (ρ = −0.367, p = 0.043) and April (ρ = −0.368, p = 0.042), and to temperature in March (ρ = −0.362, p = 0.046), showing these climate factors affect spell severity at certain times. The HMM found four main climate states: Dry (33.6% of months), Transitional (10.5%), Wet (19.4%), and Very Wet (36.6%). These states tended to persist from month to month, with self-transition probabilities between 0.57 and 0.89. The Dry State, with almost no rainfall (mean = 0.1 mm) and low humidity (29%), matched 92% of dry spell months, supporting this method. Over the decades, the transitional state has become less common in the last decade (2015–2025), suggesting that moisture patterns are becoming more polarised, likely due to ongoing warming.
Language: English
Subfield: Global and Planetary Change
Field: Environmental Science
Domain: Physical Sciences
SDG: Climate action
Open Access status: green
OpenAlex ID: W7204937786
FWCI:
Cites:
Cited by:
Related to: