Climate story · Kénya, Mali
Growing up, we more or less knew when the rains would come. Now it feels different. Sometimes we wait much longer, and then the rain comes all at once.
For me, climate change is that the seasons don’t feel as reliable as they used to.
The climate data behind this story
About this data, sources, methods & scenarios in detail →
Each stripe is one year's average temperature, . Blue = cooler, red = warmer than the 1961–1990 average. The color scale is identical for every location (saturating at ±2.5 °C), so charts are directly comparable.
Bars show how much warmer (red) or cooler (blue) each year was. Lines: this location’s 10-year average and, for comparison, the global average (HadCRUT5) on the same baseline.
The zero line is the 1961–1990 average — “what used to be normal”. Each bar shows how much warmer or cooler a year was than that. Single years bounce around naturally (that is weather); the dark line smooths them into a 10-year average (that is climate). The dashed line shows the same for the whole planet — land areas like this usually warm faster than the global average, because oceans, two thirds of the planet, heat up more slowly.
Number of days per year where the daily maximum reached 30 °C or more.
Simply the number of days per year on which the thermometer reached 30 °C or more. These days strain bodies, buildings, and crops. Watch how the bars cluster: in many places, what used to be a rare summer event has become an every-year expectation.
Total yearly precipitation. The dashed line is the 1961–1990 average.
Total rain and snow per year, with the old normal as a dashed line. Rainfall naturally varies a lot from year to year, so single bars mean little. One subtle point: even where yearly totals barely change, the landscape can still get drier — because warmer air evaporates more water. That effect is what the drought index below measures.
The single wettest day of each year (Rx1day). Warmer air holds ~7 % more moisture per degree, so rain increasingly arrives in concentrated bursts — this can rise even where yearly totals fall. Dark bars: well above the typical heaviest day of 1961–1990 (dashed line).
Each bar is the biggest 24-hour rainfall of that year. Where these bars grow while yearly totals shrink, “more drought” and “more floods” are both true at once — long dry spells, interrupted by harder downpours — and several stories on this site describe exactly that. One honest limit: an actual cloudburst is a sub-hourly event a few kilometres wide, which this ~25 km daily data smooths over. The chart shows the trend that makes such events more likely, not the individual disaster.
The standard drought metric: 12 months of rain and snow minus what the warming atmosphere evaporates, compared with 1961–1990. Brown bars below −1 mean drought, below −2 extreme drought; teal means wetter than normal.
The drought index asks: over the last 12 months, did the land receive more water (rain, snow) than the atmosphere took back (evaporation)? Zero means a normal balance for this place. Below −1 counts as drought, below −2 as extreme drought — like a water account sliding into overdraft. Because warmer air is “thirstier”, droughts can deepen even when rainfall stays the same. SPEI is a standard scientific measure, so these values are comparable with published drought research.
Longest run of days with less than 1 mm of rain (the standard "consecutive dry days" index). The dashed line is the 1961–1990 average.
For each year: the longest unbroken run of days with less than 1 mm of rain — essentially, the longest stretch the landscape had to survive on stored water. Long dry spells stress soils, gardens, and forests even in years whose rainfall total looks normal.
Nights that never cool below 20 °C, when homes stop recovering from the day's heat. The dashed line is the 1961–1990 average.
Nights that never cooled below 20 °C. Cool nights are when bodies and buildings shed the day’s heat — when they disappear, heat stress accumulates, sleep suffers, and heatwaves become dangerous, especially for older people. In much of Europe these nights barely existed a generation ago.
The year you just lived through, day by day. Daily mean temperature against the 1961–1990 normal for that calendar day (dashed line and grey band = the normal and its typical range; the line's color shows each day's departure); rain adding up over the year vs a normal year; and the drought index, month by month.
The top chart shows every single day of the past year against what used to be typical for that calendar day: the grey band covers the range where roughly two of three days used to land. Days above it were unusually warm for their date, days below unusually cool — the line’s color shows the departure. The middle chart adds up rain through the year: if the solid line runs below the dashed one, the year has been drier than a normal year. The bottom row shows the drought account balance for each of the last 12 months.
Where is this heading? Observed temperature (ERA5) next to the average of three high-resolution CMIP6 climate models, bias-corrected against ERA5-Land; the shaded band spans the models. Projections follow a high-emission pathway (SSP5-8.5) after 2015, a "without stronger climate action" trajectory, not a prediction, and up to 2050 the emission pathways have barely diverged yet, so this is a fair approximation of any future short of rapid decarbonization. High-resolution CMIP6 runs end in 2050 by experiment design. On Global you can see where this pathway sits among all the futures still open, and what happens after 2050.
Climate models are excellent at physics but cannot know future human choices — how much CO₂ the world will emit depends on policy and technology. So scientists compute “what if” storylines instead. This chart follows the high-emission storyline (“no further climate action”): the grey line is what actually happened; the orange line is where three high-resolution models say this place would head under that storyline, and the band shows how much the models disagree. Individual future years are plausible weather, not forecasts — only the trend is meaningful. Which pathway humanity takes is a choice, not a physics problem.