Verifying Shifts in the Equatorial African Precipitation Cycle Using a New Seasonal Rainfall Model

Growing concern surrounding food security in various regions of equatorial Africa has motivated recent efforts to better understand the seasonal precipitation cycle on the continent. Global climate models and energetic theories predict shifts in large-scale drivers of precipitation as a response to anthropogenic climate change; the Intertropical Convergence Zone (ITCZ), a global climate phenomenon connected to equatorial rainfall, is perhaps the most prominent example. While many studies have noted real and present changes in regional seasonal cycles, a theorized shift in the ITCZ present in many global climate models has yet to be identified in the observational record.

To address this discrepancy, we apply a novel double-Gaussian model to three precipitation data products—the Global Historical Climate Network v. 2 (GHCN) rain gauge dataset (Peterson & Vose 1997), the University of Delaware precipitation record (Willmot & Matsuura 2001), and the Tropical Rainfall Measurement Mission 3B43_V6 (TRMM) satellite product (Huffman et. al. 2007)—isolating metrics for seasonal timing, magnitude, and duration. This model represents the observational data extremely well, achieving mean correlations of 0.92 for the GHCN and University of Delaware records and 0.88 for the TRMM record. Using the derived seasonality metrics, we track change in seasonal precipitation over the last century. Along the equator, we find that the timing of peak rainfall in the boreal spring season has shifted earlier in the annual cycle by 6.68 days over the last century and that the boreal fall season has shifted later by 4.43 days in the same timeframe. These broadly synchronous trends support the energetic theory of a northward-shifting ITCZ residing longer in the Northern Hemisphere under anthropogenic climate change. This result underscores the importance of considering both global and regional influences while trying to understand precipitation behavior on any scale. We believe our results to be particularly relevant to the equatorial African region, where changes in seasonal timing have very real consequences on agriculture. The success of rain-fed subsistence crops that support most of the region’s population depends as much upon proper timing of planting as quantity of rainfall in a season. As such, understanding changes related to seasonal rainfall over the continent may better inform future agricultural practices.