Figure 5. Tide gauge predictions and observations. (a) Maximum yearly change at each PSMSL RLR tide gauge site. Six sites have a predicted a maximum yearly change of 5 mm or more. These six sites are distinguished by a larger symbol; also indicated are the tide gauge name, the maximum predicted yearly increase, and the year in which that maximum occurred. (b) Predicted sea level change due to reservoir construction (blue) versus RLR PSMSL observed (black dots) and atmospheric corrected sea level (Piecuch et al., 2019; red points) at the Baie Comeau, Québec, Canada tide gauge, near the Manicouagan Reservoir. The reservoir was filled from 1970 to 1978. Units on the vertical axis are millimeters, but the absolute value is arbitrary. (c) As in (b), but for tide gauge in Point-au-Père, Québec, Canada.
4.2 Variability in Reservoir Storage
Seepage is not the only mechanism that can alter the mass of a reservoir after impoundment. An additional post-construction signal comes from variations in the amount of water impounded, either seasonally, or in long-term draw-down of the water. We assess both of these for a handful of the largest reservoirs. Records of lake levels can be difficult to obtain in a uniform manner, so we use changes in the surface height as observed by satellite altimetry, provided by the USDA Global Reservoir and Lake Elevation Database (https://ipad.fas.usda.gov/cropexplorer/global_reservoir/) to estimate the mass change. The satellites generally observed the largest reservoirs two to three times a month, beginning in 1992 and continuing through the study period. Because fluctuations in lake level are small over the ~10-day timescale between satellite observations, the behavior of the reservoir volume can be reasonably well constrained by this method. Dramatic increases of impounded water that happen over a shorter timescale are generally due to large precipitation events, in which case the signal from the reservoir is likely to be masked by a significant increase in groundwater and surface water unrelated to the reservoir itself.
From the surface area and volume provided for each reservoir in the GRanD database we calculate a mean depth. The altimetry data show changes in altitude, but not absolute altitude, so we cannot use these data to verify when the reservoir is full. Instead, we set the maximum datum from the altimetry as the mean depth of the reservoir, assuming that this represents a full reservoir. We then use every other altimetric data point to estimate the fractional change of the mean depth, and approximate this as the fractional change of the volume of the reservoir. We discuss both long-term drawdown and seasonal changes in water storage.
Lake Powell in the United States is an excellent target for detailed investigation. First, high-resolution data are available to verify the accuracy of the GRanD dataset. Second, satellite altimetry data show that Lake Powell is characterized by a decline in lake level of slightly more than 12 m between 2000 and 2011. Therefore, we can use the data set to investigate the effects of a long-term reduction in water storage, and whether the use of satellite altimetry is appropriate in determining storage changes in large reservoirs.
GRanD lists Lake Powell as the 43rd largest reservoir in the database, with a capacity of 25.07 km3 and a surface area of 120.7 km2. In contrast, the Bureau of Reclamation (BoR; part of the United States Department of the Interior), which manages Lake Powell and the Glen Canyon Dam that impounds it, lists the capacity as 32.3 km3 and the surface area as 688.9 km2 (https://www.usbr.gov/uc/rm/crsp/gc/). These two sets of estimates produce very different values for the mean depth of the reservoir, 208 m and 47 m respectively. The 12 m decline shown in the satellite altimetry data, then, represents a reduction in impounded water of either 6% or 25%. The BoR also provide daily calculations of reservoir storage based on the observed elevation of the water level and bathymetric surveys. For the time period 2000 to 2011, these surveys (https://www.usbr.gov/rsvrWater/HistoricalApp.html) indicate that the elevation in Lake Powell decreased 13 m from 1122 m to 1109 m, and that storage in Lake Powell decreased from 27.1 km3 to 19.7 km3, a decrease of 27%. The satellite altimetry data are thus consistent with the water impoundment reported by the BoR, and additionally support the BoR-reported parameters for reservoir size, rather than the values in the GRanD database.
While a decline of 27% in the volume of the reservoir does contribute meaningfully to the GRD fingerprint through time, it is similar in magnitude—but opposite in sign—to the seepage we assume for Lake Powell (Fig. S4b). This time series indicates that significant, long-term drawdown in impoundment can change the magnitude of local sea level change associated with the unloading. However, for reservoirs that do not experience significant drawdown in storage, periods of low water levels will be characterized by a relatively small perturbation in local sea level, and will have a second-order effect on the global fingerprint of artificially impounded water.
Lake Guri, a reservoir in Venezuela with a volume of 135 km3, experiences seasonal variations in surface height that routinely exceed 10 m, representing almost 30% of the reservoir’s capacity as calculated from the GRanD-reported volume and surface area. We compare sea level predictions—those that do and do not include variability in Lake Guri—at the closest grid point to Lake Guri at in Fig. S4a. Note that this location is not on the coast; because the coast is at a greater distance, the impoundment signal will be smaller there. The perturbations in sea level in this case are on the order of a few millimeters, with higher sea level occurring when the reservoir is full, as expected. If the seasonality is due to precipitation, i.e ., a full reservoir occurs during a rainy season, then GRD effects associated with the higher level of impoundment could enhance the hazard of coastal flooding at precisely the time when increasing rain makes flooding more likely.
4.3 Future Changes in Sea Level
Our projections of future changes in sea level due to water impoundment are significantly larger than previous estimates. Following Rahmstorfet al ., (2012), Kopp et al . (2014) use impoundment data from the WRD (Chao et al ., 2008) and population projections from the United Nations Department of Economics and Social Affairs (2013) to derive a relationship between cumulative impoundment and population (see Supporting Information); their analysis implies a maximum additional water impoundment corresponding to an additional GMSL fall of 6 mm. We plot the future construction suggested by Zarfl et al . (2015) alongside this relationship in Fig. S5, and show that the extrapolation to 2040 based on Kopp et al . (2014) underestimates impoundment by 8.4 mm equivalent GMSL fall. There are uncertainties in our estimates based on the Zarfl et al . (2015) database. However, it is clear that the previously derived relationship between cumulative impoundment and world population is likely to significantly underestimate future artificial impoundment contributions to sea level changes.
Finally, we show in detail predictions for two areas that the Zarflet al . (2015) database indicates will experience significant increases in reservoir impoundment: Southeast Asia and southeastern Brazil (Fig. 4). These relatively low-elevation, highly populated areas near the coast will experience a sea level rise due to impoundment of as much as ~10 mm in the next two decades that will enhance the hazard associated with global mean sea level rise. Vitousek et al . (2017) suggest that in tropical areas, sea level rise of as little as 50 mm can double the coastal flooding hazard. Thus, major reservoir construction near such coasts could change the frequency of flooding in these populated areas.
5 Conclusions
We present an estimate of spatially and temporally resolved sea level change due to the impoundment of water in artificial reservoirs from 1900 to 2011, and a projection of sea level change to the year 2040 due to the same effects. For each year over the period 1900 – 2011, our predicted global GRD fingerprints are available by download from doi.org/10.5281/zenodo.3751986.
Our analysis of historical data (Lehner et al. , 2011) is consistent with previous studies (Chao et al ., 2008; Fiedler and Conrad, 2010) that have estimated that globally averaged sea level fell between 21 mm and 30 mm over the course of the twentieth century, representing a significant fraction of the sea level budget. Our spatial analysis is generally consistent with previous work (Fiedler and Conrad, 2010), but the spatio-temporal resolution of the predictions we present allows for comparison of predictions to records from specific tide gauge sites. Our estimate that reservoir impoundments could have raised sea level by as much as 40 mm should motivate such studies.
Our analysis of reservoirs that are in some form of planning or construction (Zarfl et al. , 2015) shows that the era of reservoir construction has not ceased, that this continued impoundment will contribute significantly to changes in sea level, and that the spatial pattern of sea level change will be different from that of the majority of the twentieth century. We show that current best projections may underestimate by nearly a centimeter the effect of water impoundment on GMSL in the next 20 years. It is difficult to accurately predict the sea level fingerprint of reservoirs without precise estimates of the volumetric capacity. Nonetheless, our calculations demonstrate that in areas characterized by a high density of reservoir construction, artificial impoundment is predicted to cause a local rise in sea level, which may change the hazard of coastal flooding. This suggests that an analysis of the local effects on sea level should be performed prior the impoundment of large volumes of water in tropical areas of low elevation, including Southeast Asia and southeastern Brazil.
Finally, our study highlights the importance of establishing a comprehensive database of water impoundment that is volumetrically complete, and geospatially and temporally referenced. Artificial water impoundment is a crucial piece of the sea level budget (Cazenaveet al ., 2018; Kopp et al ., 2014, 2015), and accurately accounting for its spatially and temporally varying contribution is imperative. The database would be important not only for the community concerned with the impacts of sea level rise, but also to a number of others, including those focused on assessments of global electricity generation (Zarfl et al. , 2015), the impact of river fragmentation on ecosystems (Grill et al. , 2015), and hazards related to reservoir-induced seismicity (e.g ., Gupta, 2002). In the absence of such a database, our analysis represents the most complete estimate to date of sea level change due to artificial impoundment of water from 1900 to 2040.
Acknowledgments, Samples, and Data
The authors wish to thank C. Zarfl for the database of future hydroelectric construction projects; and C. Piecuch, J. Fiedler, C. Conrad, and B. Chao for supplying their models that we use to contextualize our work. Satellite altimetry products courtesy of the USDA/NASA G-REALM program at https://ipad.fas.usda.gov/cropexplorer/global_reservoir/. Data for reservoirs maintained by the U.S. Bureau of Reclamation, and accessed via their website at https://www.usbr.gov. GRD Fingerprints for the years 1900 – 2011 are provided at doi.org/10.5281/zenodo.3751986. GMT was used to create many of the figures in this manuscript (Wesselet al ., 2013). Support for this work was provided by the NSF (DGE-1106400 to WBH; ICER-1663807 to REK) and NASA (NSSC17K0698 to JXM; 80NSSC17K0698 to REK). Harvard University also provided funding to JXM.
References
Cazenave, A., & WCRP Global Sea Level Budget Group (2018). Global sea-level budget 1993–present. Earth System Science Data ,10 (3), 1551–1590. https://doi.org/10.5194/essd-10-1551-2018
Chao, B. F., Wu, Y. H., & Li, Y. S. (2008). Impact of artificial reservoir water impoundment on global sea level. Science ,320 (5873), 212–214. https://doi.org/10.1126/science.1154580
Chepurin, G. A., Carton, J. A., & Leuliette, E. (2014). Sea level in ocean reanalyses and tide gauges. Journal of Geophysical Research: Oceans , 119 (1), 147–155. https://doi.org/10.1002/2013JC009365
Compo, G. P., Whitaker, J. S., Sardeshmukh, P. D., Matsui, N., Allan, R. J., Yin, X., et al. (2011). The Twentieth Century Reanalysis Project.Quarterly Journal of the Royal Meteorological Society ,137 (654), 1–28. https://doi.org/10.1002/qj.776
Dangendorf, S., Hay, C., Calafat, F. M., Marcos, M., Piecuch, C. G., Berk, K., & Jensen, J. (2019). Persistent acceleration in global sea-level rise since the 1960s. Nature Climate Change ,9 (9), 705–710. https://doi.org/10.1038/s41558-019-0531-8
Dziewonski, A. M., & Anderson, D. L. (1981). Preliminary reference Earth model. Physics of the Earth and Planetary Interiors ,25 (4), 297–356. https://doi.org/10.1016/0031-9201(81)90046-7
Fiedler, J. W., & Conrad, C. P. (2010). Spatial variability of sea level rise due to water impoundment behind dams. Geophysical Research Letters , 37 (12), 1–6. https://doi.org/10.1029/2010GL043462
Gregory, J. M., Griffies, S. M., Hughes, C. W., Lowe, J. A., Church, J. A., Fukimori, I., et al. (2019). Concepts and Terminology for Sea Level: Mean, Variability and Change, Both Local and Global. Surveys in Geophysics , 40 (6), 1251–1289. https://doi.org/10.1007/s10712-019-09525-z
Grill, G., Lehner, B., Lumsdon, A. E., MacDonald, G. K., Zarfl, C., & Reidy Liermann, C. (2015). An index-based framework for assessing patterns and trends in river fragmentation and flow regulation by global dams at multiple scales. Environmental Research Letters ,10 (1), 015001. https://doi.org/10.1088/1748-9326/10/1/015001
Gupta, H. K. (2002). A review of recent studies of triggered earthquakes by artificial water reservoirs with special emphasis on earthquakes in Koyna, India. Earth-Science Reviews , 58 (3–4), 279–310. https://doi.org/10.1016/S0012-8252(02)00063-6
Harr, M. E. (Ed.). (1962). Groundwater and Seepage . New York: McGraw-Hill.
Holgate, S. J., Matthews, A., Woodworth, P. L., Rickards, L. J., Tamisiea, M. E., Bradshaw, E., et al. (2013). New Data Systems and Products at the Permanent Service for Mean Sea Level. Journal of Coastal Research , 288 , 493–504. https://doi.org/10.2112/JCOASTRES-D-12-00175.1
Kendall, R. A., Mitrovica, J. X., & Milne, G. A. (2005). On post-glacial sea level - II. Numerical formulation and comparative results on spherically symmetric models. Geophysical Journal International , 161 (3), 679–706. https://doi.org/10.1111/j.1365-246X.2005.02553.x
Kopp, R. E., Horton, R. M., Little, C. M., Mitrovica, J. X., Oppenheimer, M., Rasmussen, D. J., et al. (2014). Probabilistic 21st and 22nd century sea-level projections at a global network of tide-gauge sites. Earth’s Future , 2 (8), 383–406. https://doi.org/10.1002/2014ef000239
Kopp, R. E., Hay, C. C., Little, C. M., & Mitrovica, J. X. (2015). Geographic Variability of Sea-Level Change. Current Climate Change Reports , 1 (3), 192–204. https://doi.org/10.1007/s40641-015-0015-5
Lehner, B., Liermann, C. R., Revenga, C., Vörösmarty, C. J., Fekete, B. M., Crouzet, P., Döll, P., Endejan, M., & Frenken, K. (2011). Global Reservoir and Dam (GRanD) database. European Environment , (March), 12. https://doi.org/10.7927/H4HH6H08
Lehner, B., Liermann, C. R., Revenga, C., Vörösmarty, C. J., Fekete, B. M., Crouzet, P., Döll, P., Endejan, M., Frenken, K., et al. (2011). High-resolution mapping of the world’s reservoirs and dams for sustainable river-flow management. Frontiers in Ecology and the Environment , 9 (9), 494–502. https://doi.org/10.1890/100125
Loo, T. (2007). Disturbing the Peace: Environmental Change and the Scales of Justice on a Northern River. Environmental History ,12 (4), 895–919. https://doi.org/10.1093/envhis/12.4.895
Mitrovica, J.X. & Milne, G.A. (2003). On post-glacial sea level: I. General theory. Geophys. J. Int., 154 , 253-267.
Mitrovica, J.X., Gomez, N., Morrow, E., Hay, C., Latychev, K. , & Tamisiea, M.E. (2011). On the robustness of predictions of sea-level fingerprints. Geophys. J. Int., 187 , 729-742.
Piecuch, C. G., Calafat, F. M., Dangendorf, S., & Jordà, G. (2019).The Ability of Barotropic Models to Simulate Historical Mean Sea Level Changes from Coastal Tide Gauge Data . Surveys in Geophysics (Vol. 40). Springer Netherlands. https://doi.org/10.1007/s10712-019-09537-9
Piecuch, C. G., Huybers, P., & Tingley, M. P. (2017). Comparison of full and empirical Bayes approaches for inferring sea-level changes from tide-gauge data. Journal of Geophysical Research: Oceans ,122 (3), 2243–2258. https://doi.org/10.1002/2016JC012506
Poli, P., Hersbach, H., Tan, D., Dee, D., Thépaut, J.-N., Simmons, A., et al. (2013). The data assimilation system and initial performance evaluation of the ECMWF pilot reanalysis of the 20th-century assimi- lating surface observations only (ERA-20C). ERA Report series 14.
Rahmstorf, S., Perrette, M., & Vermeer, M. (2012). Testing the robustness of semi-empirical sea level projections. Climate Dynamics , 39 (3–4), 861–875. https://doi.org/10.1007/s00382-011-1226-7
Vitousek, S., Barnard, P. L., Fletcher, C. H., Frazer, N., Erikson, L., & Storlazzi, C. D. (2017). Doubling of coastal flooding frequency within decades due to sea-level rise. Scientific Reports ,7 (1), 1399. https://doi.org/10.1038/s41598-017-01362-7
Vörösmarty, C. J., Sharma, K. P., Fekete, B. M., Copeland, A. H., Holden, J., Marble, J., & Lough, J. A. (1997). The Storage and Aging of Continental Runoff in Large Reservoir Systems of the World.Ambio , 26 (4), 210–219.
Wessel, P., Smith, W. H. F., Scharroo, R., Luis, J., & Wobbe, F. (2013). Generic Mapping Tools: Improved Version Released. Eos, Transactions American Geophysical Union , 94 (45), 409–410. https://doi.org/10.1002/2013EO450001
United Nations, Department of Economic and Social Affairs, Population Division (2013). World Population Prospects: The 2012 Revision, Highlights and Advance Tables. Working Paper No. ESA/P/WP.228.
Zarfl, C., Lumsdon, A. E., Berlekamp, J., Tydecks, L., & Tockner, K. (2015). A global boom in hydropower dam construction. Aquatic Sciences , 77 (1), 161–170. https://doi.org/10.1007/s00027-014-0377-0