Subsequently, we re-assess global estimates of reservoirs’ C budget (defined as emission-to-burial ratios) by partitioning total emissions into fluxes from the water surface and from the drawdown area, and propagating the uncertainty from the area, and emission and burial rates estimates. In this study, we use that dataset to estimate the extent of reservoir drawdown areas globally and to identify drivers influencing the extent of drawdown areas as well as their spatial and temporal patterns. Recently, a new algorithm was developed providing 30-year-long time series of surface-area variations in thousands of reservoirs worldwide 18. To include emissions from drawdown areas into C budgets of reservoirs and, ultimately, into global C inventories, information about the spatial extent and distribution of drawdown areas is crucial. Extreme water-level drawdown can furthermore create hot moments of GHG emissions and may even offset C burial in the sediments 16, 17. CO 2 emissions from dry inland waters were shown to be controlled mainly by temperature, moisture and organic-matter content of the desiccated sediments 11. Initial studies reported that the drawdown area has the potential to dominate annual CO 2 emissions from reservoirs 12, 14, 15, 16 due to areal CO 2 emissions, which are significantly higher than those from the water surface 11. GHG emissions from drawdown areas are typically dominated by CO 2 while CH 4 is of minor importance 13. It has been shown that dry aquatic sediments in general, and drawdown areas of reservoirs in particular, emit large quantities of GHGs 11, 12.

Although the relevance of these two points has already been acknowledged 1, they have not been considered in reservoir C budgets 10 and global upscalings 1, 6 so far. This has two consequences for upscaling C fluxes from reservoirs: (1) the variable surface area has to be considered in upscaling fluxes based on water surface, and (2) the GHG emissions from the drawdown area have to be included in reservoir C budgets.

Varying portions of reservoirs are temporarily falling dry, forming the so-called drawdown area (Fig. The most obvious consequence of water-level changes, however, is a changing water surface area (Fig. Water-level fluctuations have several consequences for reservoir limnology and biogeochemical cycling 8, 9. They are driven both by natural hydrological dynamics and water management. Water-level changes are a typical feature of reservoirs and a major discriminator to natural lakes 7. Globally, rates of organic carbon (OC) burial in reservoirs have been estimated to exceed rates of C emissions from reservoirs 1, 6. At the same time, reservoirs act as a sediment trap, accumulating organic material from the catchment and in-lake primary production, and they have previously been shown to bury C at higher rates than natural lakes 6. The exact quantification of this source is also important because GHG emissions from reservoirs may affect the carbon (C) footprint of irrigation 2 and hydropower production and thus its perception as C-neutral energy 3, 4, 5. The quantification of this GHG source has been subject to intensive research for at least two decades, and current assessments estimate reservoirs to annually emit 800 Tg carbon dioxide equivalent (CO 2e) to the atmosphere 1. Reservoirs, like inland aquatic ecosystems in general, are recognized as a globally relevant source of greenhouse gases (GHGs) 1. Thus, consideration of drawdown areas overturns our conception of the role of reservoirs in the carbon cycle. This suggests that reservoirs emit more carbon than they bury, challenging the current understanding that reservoirs are net carbon sinks.

Taking into account drawdown areas, the ratio between carbon emissions and carbon burial in sediments is 2.02 (1.04–4.26). The new estimate assigns 26.2 (15–40) (95% confidence interval) TgCO 2-C yr −1 to drawdown areas, and increases current global CO 2 emissions from reservoirs by 53% (60.3 (43.2–79.5) TgCO 2-C yr −1). We re-assessed the global carbon emissions from reservoirs by apportioning CO 2 and methane emissions to water surfaces and drawdown areas using published areal emission rates. Exposure of drawdown areas was most pronounced in reservoirs close to the tropics and shows a complex dependence on climatic (precipitation, temperature) and anthropogenic (water use) drivers. Here we show, on the basis of satellite observations of 6,794 reservoirs between 19, that 15% of the global reservoir area was dry.

However, the global extent of drawdown areas is unknown, precluding an accurate assessment of the carbon budget of reservoirs. Reservoir drawdown areas-where sediment is exposed to the atmosphere due to water-level fluctuations-are hotspots for carbon dioxide (CO 2) emissions.

Rodman reservoir drawdown schedule 2021