Zambezi runoff is affected by changes in temperature, evapotranspiration, and rainfall. The Zambezi catchment is characterized by low runoff efficiency, low drainage densities4, and relatively high aridity, indicating a high sensitivity of runoff to climate change. Given the nonlinearity of rainfall-runoff processes, a small change in annual precipitation or annual potential evaporation can have a large impact on annual river flows. Observed impacts of rising temperatures on runoff in other, comparable basins, for example, indicate that an increase of 1° C leads to an approximate 15% reduction in annual flows, exacerbating flow reductions resulting from decreasing rainfall in the catchment (Cai and Cowan 2008).

All Zambezi Basin countries will experience a significant reduction in streamflow. Multiple studies estimate that Zambezi Basin runoff will be reduced by 26-40% by 2050.

Of the 11 African basins reviewed by IPCC (2001)5, the Zambezi exhibited the "worst" effects in response to climate change, due to the resonating effect of increases in temperature and decreases in rainfall on potential evaporation and runoff. Based on ten scenarios, derived by using five different climate models in conjunction with the SRES-A2 and B2 emissions scenarios, Strzepek and McCluskey (2006) indicate that all Zambezi Basin countries will experience a significant reduction in streamflow. Multiple studies cited in IPCC (2001) estimate that Zambezi Basin runoff will be reduced by 26-40% by 2050.

The World Bank (2010) assessed the percentage change in runoff for each of the major Zambezi sub-basins by 2030, relative to the 1961-1990 baseline. Using the mid-range of 23 GCMs with emissions scenario SRES-A1B, they estimated a 16% reduction in runoff from the Upper Zambezi, 24-34% reduction in the Middle Zambezi, and 13-14% reduction in the Lower Zambezi. Norconsult (2003) carried out a sensitivity analysis of climate change on Lake Malawi using a simple water balance to show that small changes in temperature and evaporation could have a significant impact on outflow to the Shire River.

De Wit and Stenkiewicz (2006) assessed changes in surface water supply (especially perennial water availability) across Africa with predicted climate change. They noted that most of Southern Africa (including the Zambezi River Basin) is an "unstable" rainfall region that receives between 400-1000 mm rainfall per annum with high seasonality. Their models examine perennial drainage density and suggest that a 10% drop in rainfall would result in a 17% reduction in surface drainage for regions receiving ~1000 mm rainfall and a shocking 50% reduction in surface drainage for regions receiving 500 mm rainfall. They note also that the Zambezi sub-basins currently receiving 500-600 mm per year could switch from perennial to seasonal surface water supply under climate change forecasts.

Based on average annual rainfall throughout the Zambezi River Basin (about 960 mm), a ~20% reduction in basin-wide runoff is expected. But rainfall is distributed very unevenly across the basin, with the southern and western parts receiving much less rainfall than the northern and eastern parts. Regions around Harare, Zimbabwe and Chipata, Zambia are each predicted to have a 19% reduction in perennial drainage corresponding to a 10% reduction in rainfall. Maun, Botswana, just west of the Zambezi River Basin in the Okavango River basin, is predicted to have a 72% reduction in runoff corresponding to the same 10% reduction in rainfall. Some tributaries of the Middle Zambezi (draining from Zimbabwe) and lower Zambezi (draining the Mozambique highlands) could likewise experience severe reductions in perennial drainage, perhaps shifting to seasonal periods without flow. As the authors note, the extent to which reduced flow in major rivers reflects direct changes in rainfall-runoff discharge and groundwater flow, rather than reduced perennial drainage, requires further study. However, the results indicate that future availability of water, especially in headwater streams, is a serious concern in many parts of the Zambezi Basin.

4. Drainage density is the total length of all the streams in the drainage basin divided by the total area of the drainage basin – a measure of how well or how poorly rainfall-runoff drains from a given catchment.

5. Other African river basins assessed by IPCC (2001) include the Nile, Niger, Volta, Schebeli, Congo, Ogooue, Rufiji, Ruvuma, Limpopo, and Orange.