This work stands apart: its closest neighbors study either individual regions, short downpours, or the influence of warming on mean precipitation. Here, for the first time, global tropical annual totals over several decades are compared with El Niño phases. This allows us to see how natural variability reshapes the map of wet and dry zones without waiting for the warming signal to become unambiguous.
Two datasets were used: the satellite-based GPCP (2.5° × 2.5°, since 1979) for the entire tropical belt (35°S–35°N) and the gauge-based GPCC (0.5° × 0.5°, since 1950) for land only. Annual precipitation totals at each grid point were compared between epochs: the warm El Niño phase of 1979–1997 and the cold phase of 1998–2013, as well as the cold phase of 1960–1978 and the warm phase of 1979–1997. Histograms of annual totals were constructed, significance of differences was tested with the Kolmogorov–Smirnov test, and then points where totals exceeded 3000 mm or fell below 200 mm at least once during an epoch were mapped.
In the transition from the warm to the cold phase (1979–1997 → 1998–2013), the probability of annual totals below 200 mm and above 3000 mm increased significantly. The Kolmogorov–Smirnov test gave a distance of 0.02 with a p-value close to zero. Geographically, this manifested as the emergence of new extremely wet points near the western part of the Maritime Continent, over the Indian Ocean, and off the west coast of South America, as well as an expansion of dry zones off the western shores of Australia, North America, and South America. In the reverse transition (1960–1978 → 1979–1997), the pattern was almost mirror-like: wet zones contracted, dry zones appeared in southern Africa and the Sahel, and aridity retreated in Pakistan, the western United States, and Australia. This points to the dominant role of El Niño phases rather than gradual warming, since the mean temperature rose in the second case while precipitation extremes declined.
Unlike neighboring studies that link precipitation changes to rising temperatures or focus on individual regions, this work shows that the global pattern of extremes in tropical annual totals follows the natural rhythm of El Niño. This is the first test of the hypothesis of 'waxing and waning' of accumulation extremes based on satellite data over several decades. The result challenges the unambiguous attribution of observed changes in wet and dry zones to anthropogenic warming.
The topic could develop toward separating the contributions of warming and natural variability. This would require comparing epochs with the same El Niño phase but different mean temperatures. It would also be promising to include data on the duration and intensity of individual rainy seasons to link annual totals to short downpours, which are usually studied separately.
The results are important for climate modeling, seasonal forecasting, and risk assessments of droughts and floods in tropical countries. They also affect the interpretation of trends in observational data.
A next step could be to compare precipitation extremes between two cold El Niño phases (1960–1978 and 1998–2013) to assess the contribution of warming. It would also be useful to repeat the analysis with newer data covering recent years.
For climatology, the question remains unresolved as to how much of the observed changes in the hydrological cycle are caused by anthropogenic influence and how much by natural variability. This work shows that for tropical annual precipitation, the natural component can dominate, complicating the extraction of an anthropogenic signal.
The threshold for a 'dry' year in this work is 200 mm. For comparison, Moscow receives on average about 700 mm of precipitation per year, meaning a tropical point with 200 mm is three times drier than the Russian capital.