In the last two lessons you looked at water security and scarcity, and at the techniques societies use to conserve water. In the Models lesson of the Foundation unit you met the planetary boundaries model, and saw that the 2025 Planetary Health Check places freshwater change beyond its boundary.
This lesson asks two questions. How do we know whether freshwater use is sustainable? And who acts on that knowledge? You will work down three scales of measurement, from the whole planet to a single nation to a local stream, and then look at the rules and agreements that govern water use, starting on your own doorstep.
By the end of the lesson you should be able to:
<aside> <img src="/icons/leaf_green.svg" alt="/icons/leaf_green.svg" width="40px" />
4.2.9 Freshwater use is a planetary boundary, with increasing demand for limited freshwater resources causing increased water stress and the risk of abrupt and irreversible changes to the hydrological system.
Consider how the planetary boundary can be measured and possible mitigation strategies to bring it back within the boundary.
</aside>
Freshwater moves through two linked parts of the water cycle, and the boundary now tracks both.
The first version of the planetary boundaries model, published in 2009, measured freshwater use as a single global volume. Humans could safely consume up to about 4,000 km³ of blue water a year. Use at the time was about 2,600 km³, so the boundary looked safe. That measure had two weaknesses. It ignored green water completely, and a global total hides the fact that one river basin can be pumped dry while another has water to spare.
The current version measures change rather than volume. The world's land is divided into a grid of small squares. For each square, scientists use computer models of the water cycle to work out the normal range of monthly streamflow (blue water) and soil moisture (green water) before industrialisation, from 1661 to 1860. A month that is wetter or drier than that normal range counts as a local deviation. Both directions count: unusually wet months bring floods and waterlogging, unusually dry months bring drought, and either pushes the system away from its stable state.
The control variable is the percentage of the world's land area showing deviations. Even before industrialisation, natural variability meant that about one tenth of land showed deviations, so the boundary is set just above that level.
1. Distinguish between blue water and green water.
2. Explain why measuring the percentage of land with deviations is a better test of harm than measuring the total volume of water humans use.
Data from Porkka et al. (2024), Nature Water, as used in Richardson et al. (2023), Science Advances.
The chart shows the values used in the 2023 update of the planetary boundaries model, averaged over 1996 to 2005. On this assessment the blue water boundary was first crossed in 1905 and the green water boundary in 1929.
In 2025 the Planetary Health Check, run by the Potsdam Institute for Climate Impact Research, repeated the assessment with newer data and an updated method. It set the boundaries slightly higher, at 12.9% of land for blue water and 12.4% for green water. It found that about 22% of land now shows deviations for each, roughly twice the pre-industrial level, and dated the first crossing to around 1940.
The Planetary Health Check names three groups of drivers: climate change; water withdrawals and infrastructure, especially irrigation, which takes about 70% of all freshwater withdrawals; and changes in land use and vegetation, such as deforestation, farming and the spread of cities.
Source: Planetary Health Check, Freshwater change
3. Use the chart to calculate how far beyond its boundary each indicator was in 1996 to 2005, in percentage points.
4. The 2023 and 2025 assessments give different boundary values and different dates for the first crossing. Does this change the conclusion? Explain what the difference tells you about the planetary boundaries model.
Source: IPGP (2026), adapted from Larochelle et al. (2026), PNAS, Figure 3F. Licensed under CC BY 4.0. https://doi.org/10.1073/pnas.2526041123
The map shows residual subsidence across the Sacramento Valley from January 2021 to October 2022: the sinking left over once the reversible, elastic squeezing is taken away, so it shows permanent compaction. Red areas were sinking permanently by 20 cm a year or more. The small map on the right shows where the valley lies in California.
Crossing a boundary raises the risk of abrupt and irreversible changes to the hydrological system. One has just been measured beneath California's Sacramento Valley, a major farming region.
When groundwater is pumped out of an aquifer, the layers of sand and clay that held it are squeezed and the land surface sinks. At first this squeezing is elastic: when the aquifer refills, the clay springs back and the ground rises again. Push the aquifer too far, though, and the clay compacts permanently. The ground stays down, and the aquifer can never again hold as much water as before.
A study published in July 2026 in the journal PNAS, involving researchers at the Institut de Physique du Globe de Paris, combined satellite measurements of the ground surface with water levels in monitoring wells from 2016 to 2022. Until 2020, the sinking matched the elastic, reversible pattern. During the 2020 to 2022 drought, as groundwater was pumped heavily, subsidence suddenly sped up, sinking several tens of centimetres a year faster than elastic squeezing could explain. The aquifer had crossed into permanent compaction, and part of its storage capacity is lost for good. The researchers also found that this change could not have been predicted from well water levels alone.
This is the pattern the planetary boundaries model warns about: steady pressure, then a sudden shift that cannot be undone. A 2024 study of about 170,000 wells in nearly 1,700 aquifer systems found that groundwater decline has accelerated over the past four decades in 30% of regional aquifers.
Sources: IPGP, Satellites reveal irreversible damage in a major California aquifer (2026); Jasechko et al. (2024), Nature
5. Using the Sacramento Valley aquifer, explain why its change counts as both abrupt and irreversible.