Move the cursor over land. Click to pin a basin.
A precipitationshed is the region where water evaporated before falling as
rain in a selected basin. This map estimates the source region using the
Utrack climatology, which describes average conditions over the study period. The
highlighted land basins are the land sources of rain for the location under
the cursor. Trees return water to the air, and that water can become rain
farther downwind. The sidebar therefore also identifies the main
driver of forest loss in the land portion of the selected location’s
precipitationshed. The driver identifies the activity or disturbance
responsible for forest loss in those supplying basins.
Walkthrough .
Rainfall -
Basin area -
Rain from land -
From its own basin -
From ocean -
Layer detail
Rain the tracking accounts for -
UTrack cells it spans -
Moves between methods by -
Supply the gauges confirm -
Supply a real gauge measures -
Supplied one step further back -
Supplied two steps further back -
Forested and cleared
Over forest -
Over cleared land -
Difference -
Cells compared -
Supplying basins
share of this basin's total rainfall
Exposure to forest loss in the rainshed
Loss in the land that supplies it -
Water risk
Water stress here -
Drought risk here -
Drought risk of the rainshed -
Which way the loss is going
Supply from accelerating land -
Its rate now against 2015-19 -
Driver of forest loss in the rainshed
Governance split
Effective supplier countries -
Foreign terrestrial supply -
Terrestrial rainfall fraction -
highlight
no highlight
mostly land-sourced (over 80% from land)
almost entirely land-sourced (over 90% from land)
mostly ocean-sourced (over 80% from ocean)
almost entirely ocean-sourced (over 90% from ocean)
land-sourced, most forest loss in the rainshed
risk exposure
off
rainshed forest loss × people
rainshed forest loss × cropland
forest loss in the rainshed itself
land supply from abroad
countries the land supply is split between
none of it 80% or more
how much of the land supplying the basin lies
in a cell that lost forest, 2001-2025
This layer shows forest loss in the land that supplies the basin’s rain. The
population and cropland views also account for the share of the basin’s
rainfall that comes from land. A basin can therefore appear pale in those
combined views because its supplying land has little forest loss or because
most of its rain comes from the ocean. In the forest-loss view alone, pale
shading indicates little forest loss in the supplying land.
These layers measure exposure to possible disruption of the land sources of
rainfall. They do not show that rainfall has already decreased. They also do
not estimate how much a particular change in rainfall was caused by
deforestation.
These layers measure exposure to possible disruption of the land sources of
rainfall. They do not show that rainfall has already decreased. They also do
not estimate how much a particular change in rainfall was caused by
deforestation.
more less
less more
i This layer shows exposure to forest loss in the land that supplies rainfall.
It does not show every form of water risk. For example, India appears orange
because many people and croplands depend on rain while relatively little of
the supplying land has lost forest. Much of the North is nevertheless severely
water stressed. Forest-loss exposure and water stress describe different
risks.
These layers measure exposure to possible disruption of the land sources of
rainfall. They do not show that rainfall has already decreased. They also do
not estimate how much a particular change in rainfall was caused by
deforestation.
dryness over cleared land
off
dry season
wet season
all year
0
i The map compares the dryness of air over cleared land with the dryness of air
over land that remains forested inside the same basin. The displayed value is
the cleared-land value minus the forested-land value. The comparison uses the
same basin, month and hundred meter height band, so differences in those
conditions do not explain the displayed difference. Soil, slope and the crops
or other vegetation on cleared ground are not matched. Those remaining
differences mean the map compares the two types of land without isolating the
effect of forest clearing.
Read more
Air dryness is measured as vapor pressure deficit, in kilopascals. Vapor
pressure deficit describes how much more water vapor the air could hold before
reaching saturation. A larger deficit means the air can draw water more
strongly from leaves and soil.
The cleared and forested land are often at different elevations. Across basins
that contain both types of land, cleared land sits a median of 52 meters below
forested land. Lower ground in these basins has drier air regardless of the
vegetation, with a difference of 0.021 kPa per 100 meters. Comparing land
within the same height bands removes most of this elevation difference from
the air-dryness comparison. The remaining gradient is 0.002 kPa per 100
meters. Matching elevation also halves the average difference between cleared
and forested land.
The difference in air dryness varies across the map. At level 5 in the dry
season, the median basin has a difference of +0.001 kPa. For 72 of every 100
basins, the difference is within 0.05 of zero. The 160 tropical basins average
+0.060, while the 239 boreal basins average no difference because their air is
near saturation throughout the year. A basin shaded close to zero has no
detected difference in this comparison. That result does not establish that
clearing has no effect.
Vapor pressure deficit can be separated into a contribution from temperature
and a contribution from the amount of moisture in the air. Both contributions
were calculated from the same data product on the same days. Four fifths of
the tropical difference comes from the moisture contribution, identified here
as lost transpiration. Transpiration is the return of water to the air through
plants. The temperature contribution is no larger than the difference this
method finds when comparing forest with forest, so no temperature signal is
detected. The humidity data come from a quarter degree reanalysis. The
comparison therefore describes differences between separate grid cells and
cannot resolve a clearing smaller than a cell.
how far back to follow the water
where it last evaporated
one step further back
two steps further back
both further steps
where its rain last
evaporated
Read full explanation
Validation layers
shed shading
where the rain came from
tracked rain against gauges
ERA5 against gauges
rain gauges per cell
Measured rainfall is not in this
deployment.
What the gauges can and cannot settle
Read full explanation
where to trust this map
off
tracked rain against ERA5
basin against UTrack's grid
both
movement between methods
The comparison between tracking methods uses annual rainfall totals. The
reconciled moisture flows do not provide separate seasonal estimates, so this
comparison is unavailable for individual seasons.
Rainfall accounting and limitations
Read full explanation
Two additional limitations apply throughout the dataset. The first concerns
the method used to track moisture. Changing that method can change the
estimated sources and travel distances. UTrack's own authors ran Lagrangian
and Eulerian schemes on the same moisture and found the continental recycling
ratio shifting by more than ten percentage points between them, with transport
distances differing by several degrees (Tuinenburg and Staal, 2020). The
tracking model therefore influences the estimated distance traveled by water
and the share that originated on land. These estimates do not depend on
weather alone. The second limitation concerns the size of the basin. A larger
basin includes more of its own moisture sources inside its boundary, so the
proportion of rain supplied from within the basin increases even without a
change in atmospheric transport. In this map, self-supply means the share of
rain supplied by evaporation inside the same basin. Self-supply scales with
basin area to the power 0.8. Differences in basin area account for most of the
variation in self-supply. The median is 0.3% at level 6 and 4.7% at level 3.
Self-supply compares between basins at the same HydroBASINS level, never
across them (van der Ent and Savenije, 2011).
Theeuwen et al. (2023) provide an independent estimate of local moisture
recycling on the same half-degree grid. Their ratio describes the share of
water evaporating from a cell that falls as rain inside a three-by-three block
of cells, about 50 km across. Their ratio begins with the water evaporating
from a place. This map’s self-supply ratio begins with the rain falling into a
place. Because the ratios use different starting quantities, they should
differ by roughly the ratio of land evaporation to rainfall. ERA5 estimates
that ratio as 0.65 over this grid. When the compared source regions have
matching sizes, this map’s estimate is 0.47 to 0.51 times the independent
estimate. The two sets of basin rankings have a Spearman correlation of 0.36.
The estimates have a similar scale and the same direction of association, but
the rankings show limited agreement for individual basins.
The hatched areas do not represent UTrack's
uncertainty. UTrack provides a single climatological estimate for
2008-2017, without uncertainty intervals. Hatching identifies
basins where tracked moisture differs substantially from recorded
rainfall or where the basin is too small relative to the 0.5°
grid to resolve its individual moisture sources. Hatching appears on
the flat map only; the alternative projections show the basin markers
without it.