<<

Groundwater storage dynamics in the world’s large systems from GRACE: uncertainty and role of extreme

Mohammad Shamsudduha1,2,* and Richard G. Taylor1

1 Department of Geography, University College London, London, UK

2 Department of Geography, University of Sussex, Falmer, Brighton, UK

* Corresponding author: M. Shamsudduha ([email protected])

The Supplementary Material contains supplementary Tables S1–S4 and supplementary Figures S1–S80.

1 Supplementary Table S1. Characteristics of the world’s 37 large aquifer systems according to the WHYMAP database including aquifer area, total number of population, proportion of (GW)-fed , mean aridity index, mean annual rainfall, variability in rainfall and total terrestrial water mass (ΔTWS), and correlation coefficients between monthly ΔTWS and precipitation with reported lags.

) 2

2) WHYMAP aquifer number Aquifer name Continent Population (million) Aquifer area (km Correlation between andTWS precipitation (lag in month) GW irrigation (%) Climate zones based on Aridity indices Mean (2002-16)annual precipitation (mm) Rainfall variability (%) varianceTWS (cm Nubian Sandstone Hyper- 1 Africa 86.01 2,176,068 1.6 30 12.1 1.5 0.16 (13) Aquifer System arid Northwestern 2 Sahara Aquifer Africa 5.93 1,007,536 4.4 Arid 69 17.3 1.9 0.19 (8) System Murzuk-Djado Hyper- 3 Africa 0.35 483,817 2.3 8 36.6 1.3 0.20 (-8) Basin arid Taoudeni- Hyper- 4 Africa 0.35 720,991 0.0 66 14.6 1.2 0.46 (1) Tanezrouft Basin arid Senegal- Semi- 5 Africa 17.77 294,613 1.0 540 14.6 31.9 0.73 (2) Mauritanian Basin arid Iullemmeden- 6 Irhazer Aquifer Africa 21.57 580,870 0.4 Arid 302 11.7 6.9 0.66 (2) System 7 Lake Chad Basin Africa 32.25 1,516,233 0.4 Arid 415 9.5 16.8 0.82 (2) Umm Ruwaba Semi- 8 Aquifer (Sudd Africa 10.52 509,174 0.0 789 10.7 44.4 0.86 (2) arid Basin) 9 Ogaden-Juba Basin Africa 20.27 1,080,953 0.3 Arid 393 14.2 2.7 0.48 (1)

10 Congo Basin Africa 34.74 1,487,858 0.0 Humid 1,566 5.6 33.0 0.66 (2)

Upper Kalahari- Semi- 11 Cuvelai-Zambezi Africa 6.02 1,000,812 0.1 819 10.0 145.1 0.66 (2) arid Basin Lower Kalahari- 12 Africa 0.76 426,487 0.6 Arid 297 24.1 5.3 0.47 (2) Stampriet Basin Semi- 13 Karoo Basin Africa 14.53 568,213 2.1 479 17.6 6.5 0.30 (2) arid Northern Great North Sub- 14 6.62 1,507,325 14.5 461 9.7 24.5 0.45 (10) Plains Aquifer America humid Cambro-Ordovician North 15 13.86 129,588 91.7 Humid 880 11.6 35.1 0.39 (10) Aquifer System America California Central North Semi- 16 Valley Aquifer 8.10 71,430 57.8 515 32.0 176.5 0.53 (2) America arid System North Semi- 17 2.44 501,374 86.8 507 16.1 19.7 0.39 (9) (High Plains) America arid

2 Atlantic and Gulf North 18 Coastal Plains 65.38 1,125,968 55.8 Humid 1,230 11.4 43.1 0.43 (8) America Aquifer South 19 Amazon Basin 8.93 2,279,522 1.0 Humid 2,505 8.3 347.2 0.84 (2) America South 20 Maranhao Basin 10.81 592,680 32.6 Humid 1,502 15.7 217.2 0.82 (2) America South 21 System (Parana 47.84 1,834,681 20.5 Humid 1,450 10.6 40.5 0.61 (3) America Basin) Arabian Aquifer 22 Asia 27.43 1,782,954 7.2 Arid 65 15.0 6.1 0.12 (0) System Semi- 23 Indus River Basin Asia 155.85 308,436 31.0 375 16.2 13.0 0.56 (1) arid Ganges- 24 Asia 596.44 616,422 55.8 Humid 1,391 12.1 255.6 0.73 (2) Brahmaputra Basin West Siberian 25 Asia 12.71 2,703,532 3.4 Humid 483 8.1 39.1 0.71 (8) Artesian Basin

26 Tunguss Basin Asia 0.43 1,241,981 0.1 Humid 432 9.8 38.1 0.66 (8)

27 Angara-Lena Basin Asia 2.50 620,591 1.3 Humid 472 7.9 21.6 0.55 (8)

28 Yakut Basin Asia 0.63 861,053 0.7 Humid 386 8.2 19.5 0.58 (-4)

North China Plains 29 Asia 336.70 438,954 37.1 Humid 826 10.0 32.8 0.32 (2) Aquifer System

30 Song-Liao Plain Asia 54.49 279,044 53.5 Humid 514 17.0 19.5 0.20 (1)

31 Tarim Basin Asia 7.69 454,288 2.8 Arid 90 23.6 1.9 0.50 (0)

32 Paris Basin Europe 28.58 178,513 82.3 Humid 727 10.8 32.9 0.24 (6)

East European 33 Europe 100.22 2,859,962 16.6 Humid 606 6.2 35.3 0.54 (7) Aquifer System North Caucasus Semi- 34 Europe 13.30 294,821 4.2 524 11.6 49.2 0.30 (10) Basin arid

35 Pechora Basin Europe 0.19 220,245 0.0 Humid 559 9.5 51.2 0.60 (7)

Great Artesian Semi- 36 Australia 0.20 1,765,830 0.9 444 28.9 20.2 0.55 (2) Basin arid

37 Canning Basin Australia 0.01 432,774 0.4 Arid 443 21.2 50.3 0.47 (2)

3 Supplementary Table S2. Characteristics of the world’s 37 large aquifer systems according to the WHYMAP database including mean annual rainfall, mean aridity index, and variability in total terrestrial water mass (ΔTWS), and correlation coefficients between ΔTWS and annual precipitation. The variance (cm2) in the ensemble GRACE ΔTWS and the proportion of variance explained by and snow water equivalent combined (SWS+SNS), soil moisture storage (SMS) and groundwater storage (GWS) are tabulated.

) 2

Mean (2002–2016) Annual precipitation (mm) Climate zones based on Aridity indices varianceTWS (cm Correlation between annual changes in TWS and annual precipitation varianceTWS explained by SWS+SNS combined (%) varianceTWS explained by (%)SMS varianceTWS explained by GWS (%) WHYMAP aquifer number Aquifer name Nubian Sandstone 1 30 Hyper-arid 1.5 0.32 0.5 32.4 67.1 Aquifer System Northwestern Sahara 2 69 Arid 1.9 0.62 0.9 14.2 84.9 Aquifer System

3 Murzuk-Djado Basin 8 Hyper-arid 1.3 0.08 1.6 1.5 96.9

Taoudeni-Tanezrouft 4 66 Hyper-arid 1.2 0.21 6 14.3 79.6 Basin Senegal-Mauritanian 5 540 Semi-arid 31.9 0.65 14.6 56.6 28.8 Basin Iullemmeden-Irhazer 6 302 Arid 6.9 0.36 12.4 55.7 32 Aquifer System

7 Lake Chad Basin 415 Arid 16.8 0.50 14.8 66.6 18.6

Umm Ruwaba 8 789 Semi-arid 44.4 0.76 1.7 72.9 25.5 Aquifer (Sudd Basin)

9 Ogaden-Juba Basin 393 Arid 2.7 0.14 1.7 35.9 62.3

10 Congo Basin 1,566 Humid 33.0 0.53 15.6 23.7 60.7

Upper Kalahari- 11 Cuvelai-Zambezi 819 Semi-arid 145.1 0.78 0.2 45.6 54.3 Basin Lower Kalahari- 12 297 Arid 5.3 0.68 5.7 44.3 50 Stampriet Basin

13 Karoo Basin 479 Semi-arid 6.5 0.72 3.3 1 95.7

Northern 14 461 Sub-humid 24.5 0.60 4.8 26.5 68.7 Aquifer Cambro-Ordovician 15 880 Humid 35.1 0.75 1.5 60.1 38.3 Aquifer System

4 California Central 16 Valley Aquifer 515 Semi-arid 176.5 0.46 2.8 56 41.2 System Ogallala Aquifer 17 507 Semi-arid 19.7 0.80 2.6 14.6 82.8 (High Plains) Atlantic and Gulf 18 Coastal Plains 1,230 Humid 43.1 0.61 33.8 51.6 14.6 Aquifer

19 Amazon Basin 2,505 Humid 347.2 0.28 30.6 48.9 20.5

20 Maranhao Basin 1,502 Humid 217.2 0.92 27.5 64.2 8.3

Guarani Aquifer 21 System (Parana 1,450 Humid 40.5 0.38 21 52.3 26.7 Basin) Arabian Aquifer 22 65 Arid 6.1 0.57 5.2 56.7 38.1 System

23 Indus River Basin 375 Semi-arid 13.0 0.39 4 32.4 63.5

Ganges-Brahmaputra 24 1,391 Humid 255.6 0.69 17.7 62.7 19.5 River Basin West Siberian 25 483 Humid 39.1 0.71 36.8 11.8 51.4 Artesian Basin

26 Tunguss Basin 432 Humid 38.1 0.71 57.2 1.9 40.9

27 Angara-Lena Basin 472 Humid 21.6 0.71 23.3 27.1 49.6

28 Yakut Basin 386 Humid 19.5 0.47 26 12 61.9

North China Plains 29 826 Humid 32.8 0.22 0.3 65.4 34.3 Aquifer System

30 Song-Liao Plain 514 Humid 19.5 0.62 9.4 59.5 31.1

31 Tarim Basin 90 Arid 1.9 0.32 3 8.3 88.7

32 Paris Basin 727 Humid 32.9 0.86 8.9 57.4 33.7

East European 33 606 Humid 35.3 0.75 39.3 41.2 19.5 Aquifer System North Caucasus 34 524 Semi-arid 49.2 0.86 2.7 59.1 38.2 Basin

35 Pechora Basin 559 Humid 51.2 0.56 70 1.8 28.1

36 444 Semi-arid 20.2 0.89 17.6 55 27.3

37 Canning Basin 443 Arid 50.3 0.85 3 49.1 47.9

5 Supplementary Table S3. Characteristics of the relationships between CRU precipitation and GRACE ΔGWS in the world’s 37 large aquifer systems.

GWS Δ percentile) over th percentile precipitation th 90 ≥ WHYMAP aquifer number Aquifer name Mean (1901 to 2016) annual (hydrological year) precipitation (mm) 90 (period: 1901–2016) Rainy season: calendar year (Jan–Dec) hydrologicalor year (Aug–Jul) Extreme hydrological years ( GRACE period (2002–2016) Positive change in non-linear trends in GRACE- over the period of2002–2016 and reported values between two consecutive years (cm) Nubian Sandstone 2014 (0.43) 1 33 41 Jan–Dec 2014 (51st) Aquifer System 2015 (0.63) Northwestern Sahara 2003–2004 2005–2006 (0.25) 2 72 92 Aug–Jul Aquifer System 2005–2006 2015–2016 (0.41) 2015 (0.83) 3 Murzuk-Djado Basin 8 11 Jan–Dec 2015 (78th) 2016 (0.70) Taoudeni-Tanezrouft 2014 (0.30) 4 63 75 Jan–Dec 2010 Basin 2015 (0.31) Senegal-Mauritanian 2010 (1.96) 5 548 703 Jan–Dec 2010 Basin 2011 (1.53) Iullemmeden-Irhazer 2010 (0.75) 6 301 377 Jan–Dec 2010 (68th) Aquifer System 2011 (0.86) 2012 (0.45) 7 Lake Chad Basin 392 460 Jan–Dec 2012 2013 (0.69) 2007 (87th) 2008 (0.83) Umm Ruwaba 8 788 889 Jan–Dec 2008 (78th) 2013 (0.98) Aquifer (Sudd Basin) 2014 (60th) 2014 (1.27) 2006 (0.92) 9 Ogaden-Juba Basin 376 465 Jan–Dec 2013 2013 (0.69) 2013–2014 (5.15) 10 Congo Basin 1632 1740 Aug–Jul 2015–2016 (56th) 2015–2016 (2.93) Upper Kalahari- 2008-2009 (74th) 2008–2009 (3.05) 11 Cuvelai-Zambezi 800 935 Aug–Jul 2010-2011 (70th) 2010–2011 (5.12) Basin Lower Kalahari- 2006–2007 (0.80) 12 286 386 Aug–Jul 2010–2011 Stampriet Basin 2010–2011 (0.78) 2010–2011 (1.86) 13 Karoo Basin 500 623 Aug–Jul 2010–2011 2011–2012 (1.50) 2005 (1.71) Northern Great Plains 14 429 487 Jan–Dec 2005, 2010, 2013 2010 (2.70) Aquifer 2011 (2.42) Cambro-Ordovician 2010 (2.70) 15 819 960 Jan–Dec 2010, 2013, 2015 Aquifer System 2011 (2.42) California Central 2004–2005 (3.92) 16 Valley Aquifer 567 772 Aug–Jul 2005–2006 2005–2006 (4.15) System 2010–2011 (2.05) Ogallala Aquifer 2008 (2.26) 17 489 578 Jan–Dec 2015 (High Plains) 2009 (1.83) 2009 (1.40) Atlantic and Gulf 18 1205 1356 Jan–Dec 2009, 2013, 2015 2013 (0.80) Coastal Plains Aquifer 2015 (0.55)

6 2005–2006 2007–2008 (2.58) 19 Amazon Basin 2416 2627 Aug–Jul 2007–2008 2011–2012 (3.31) 2008–2009 2008–2009 2003–2004 (2.45) 20 Maranhao Basin 1470 1762 Aug–Jul 2010–2011 2009–2010 (1.83) Guarani Aquifer 2009–2010 2010–2011 (1.65) 21 System (Parana 1384 1573 Aug–Jul 2015–2016 2015–2016 (5.78) Basin) Arabian Aquifer 2003 (0.15) 22 74 87 Jan–Dec 2004 (40th) System 2004 (0.12) 2014 (0.93) 23 Indus River Basin 330 452 Jan–Dec 2010, 2015 2015 (0.82) Ganges-Brahmaputra 24 1483 1663 Jan–Dec 2011 (59th) 2011 (1.44) Basin 2013 (2.57) West Siberian 25 452 505 Jan–Dec 2013, 2014, 2015 2014 (3.08) Artesian Basin 2015 (2.17) 2005 (3.25) 26 Tunguss Basin 415 461 Jan–Dec 2007, 2008, 2015 2006 (3.25) 2013 (4.49) 2006 (1.71) 27 Angara-Lena Basin 456 512 Jan–Dec 2006, 2008, 2009 2013 (1.80) 2005, 2006, 2007, 2005 (1.89) 28 Yakut Basin 342 402 Jan–Dec 2008, 2013 2006 (2.15) North China Plains 2003 (3.21) 29 793 916 Jan–Dec 2003 Aquifer System 2004 (3.73) 2013 (2.14) 30 Song-Liao Plain 531 629 Jan–Dec 2012, 2013 2014 (1.57) 2003, 2005, 2010, 2015 (0.74) 31 Tarim Basin 78 105 Jan–Dec 2013 2016 (1.04) 2013–2014 (2.34) 2006–2007 32 Paris Basin 724 862 Aug–Jul 2014-2015 (1.84) 2015–2016 2015–2016 (1.53) East European 2005 (1.84) 33 585 638 Jan–Dec 2004, 2012 Aquifer System 2006 (2.17) 2004 (3.32) 34 North Caucasus Basin 514 602 Jan–Dec 2004 2005 (3.21) 2013 (3.37) 35 Pechora Basin 490 583 Jan–Dec 2003, 2007, 2014 2014 (3.21) 2009–2010 2009–2010 (1.72) 36 Great Artesian Basin 438 584 Aug–Jul 2010–2011 2010–2011 (3.34) 2005–2006 2010–2011 (1.46) 37 Canning Basin 377 547 Aug–Jul 2010–2011 2011–2012 (2.37) 2013–2014

7 Supplementary Table S4. Characteristics of the relationships between GPCC precipitation and GRACE ΔGWS in the world’s 37 large aquifer systems.

GWS Δ percentile) over th percentile precipitation th 90 ≥ WHYMAP aquifer number Aquifer name Mean (1901 to 2016) annual (hydrological year) precipitation (mm) 90 (period: 1901–2016) Rainy season: calendar year (Jan–Dec) hydrologicalor year (Aug–Jul) Extreme hydrological years ( GRACE period (2002–2016) Positive change in non-linear trends in GRACE- over the period of2002–2016 and reported values between two consecutive years (cm) Nubian Sandstone 2014 (61st) 2014 (0.43) 1 32 44 Jan–Dec Aquifer System 2015 (72nd) 2015 (0.63) 2003–2004 Northwestern Sahara 2005–2006 (0.25) 2 68 94 Aug–Jul 2005–2006 Aquifer System 2015–2016 (0.41) 2008–2009 2013 (74th) 2015 (0.83) 3 Murzuk-Djado Basin 15 23 Jan–Dec 2015 (34th) 2016 (0.70) Taoudeni-Tanezrouft 2014 (0.30) 4 91 125 Jan–Dec 2010 Basin 2015 (0.31) Senegal-Mauritanian 2010 (1.96) 5 511 655 Jan–Dec 2010 Basin 2011 (1.53) Iullemmeden-Irhazer 2010 (0.75) 6 275 350 Jan–Dec 2010 (75th) Aquifer System 2011 (0.86) 2012 (0.45) 7 Lake Chad Basin 385 452 Jan–Dec 2012 2013 (0.69) 2006, 2007 (66th), 2008 (0.83) Umm Ruwaba 2008 (83rd), 2009 8 778 882 Jan–Dec 2013 (0.98) Aquifer (Sudd Basin) (87th), 2011 (84th) 2014 (1.27) 2014 (81st) 2006 (0.92) 9 Ogaden-Juba Basin 393 481 Jan–Dec 2006 2013 (0.69) 2013–2014 (5.15) 10 Congo Basin 1657 1750 Aug–Jul 2015–2016 (48th) 2015–2016 (2.93) Upper Kalahari- 2005–2006 2008–2009 (3.05) 11 Cuvelai-Zambezi 840 952 Aug–Jul 2008–2009 (87th) 2010–2011 (5.12) Basin 2010–2011 (84h) Lower Kalahari- 2006–2007 2006–2007 (0.80) 12 303 409 Aug–Jul Stampriet Basin 2010–2011 2010–2011 (0.78) 2010–2011 (1.86) 13 Karoo Basin 500 620 Aug–Jul 2010–2011 2011–2012 (1.50) 2005 (1.71) Northern Great Plains 14 427 484 Jan–Dec 2005, 2010 2010 (2.70) Aquifer 2011 (2.42) Cambro-Ordovician 2010 (2.70) 15 834 969 Jan–Dec 2008, 2010 Aquifer System 2011 (2.42) California Central 2004–2005 (3.92) 16 Valley Aquifer 552 749 Aug–Jul 2005–2006 2005–2006 (4.15) System 2010–2011 (2.05) Ogallala Aquifer 2008 (2.26) 17 501 602 Jan–Dec 2015 (High Plains) 2009 (1.83)

8 2009 (1.40) Atlantic and Gulf 18 1215 1372 Jan–Dec 2004, 2009, 2015 2013 (0.80) Coastal Plains Aquifer 2015 (0.55) 2005–2006 2007–2008 2007–2008 (2.58) 19 Amazon Basin 2500 2661 Aug–Jul 2008–2009 2011–2012 (3.31) 2013–2014 2003–2004 (2.45) 20 Maranhao Basin 1421 1703 Aug–Jul 2008–2009 2009–2010 (1.83) 2003–2004 Guarani Aquifer 2009–2010 2010–2011 (1.65) 21 System (Parana 1378 1535 Aug–Jul 2014–2015 2015–2016 (5.78) Basin) 2015–2016 2003 (66th) Arabian Aquifer 2003 (0.15) 22 81 99 Jan–Dec 2005 (52nd) System 2004 (0.12) 2013 (88th) 2014 (0.93) 23 Indus River Basin 336 466 Jan–Dec 2011, 2015 2015 (0.82) 2007 (89th) Ganges-Brahmaputra 24 1490 1650 Jan–Dec 2008 (78th) 2011 (1.44) Basin 2011 (39th) 2013 (2.57) West Siberian 2002, 2007, 25 443 497 Jan–Dec 2014 (3.08) Artesian Basin 2014, 2015 2015 (2.17) 2005 (3.25) 2002, 2007, 2008, 26 Tunguss Basin 389 434 Jan–Dec 2006 (3.25) 2011, 2014, 2015 2013 (4.49) 2004, 2005, 2006, 2006 (1.71) 27 Angara-Lena Basin 422 471 Jan–Dec 2008, 2009 2013 (1.80) 2004, 2005, 2006, 2005 (1.89) 28 Yakut Basin 316 363 Jan–Dec 2007, 2008, 2012 2006 (2.15) 2013 North China Plains 2003 (3.21) 29 795 933 Jan–Dec 2003 Aquifer System 2004 (3.73) 2013 (2.14) 30 Song-Liao Plain 534 636 Jan–Dec 2012, 2013 2014 (1.57) 2003, 2005, 2010, 2015 (0.74) 31 Tarim Basin 48 66 Jan–Dec 2013, 2015 2016 (1.04) 2013–2014 (2.34) 2006–2007 32 Paris Basin 750 914 Aug–Jul 2014-2015 (1.84) 2015–2016 2015–2016 (1.53) East European 2005 (1.84) 33 561 624 Jan–Dec 2004, 2012 Aquifer System 2006 (2.17) 2004 (3.32) 34 North Caucasus Basin 488 584 Jan–Dec 2004 2005 (3.21) 2002, 2003, 2005, 2013 (3.37) 35 Pechora Basin 470 565 Jan–Dec 2006, 2007, 2012 2014 (3.21) 2009–2010 2009–2010 (1.72) 36 Great Artesian Basin 429 568 Aug–Jul 2010–2011 2010–2011 (3.34) 2011–2012 2004–2005 2010–2011 (1.46) 37 Canning Basin 438 587 Aug–Jul 2010–2011 2011–2012 (2.37)

9 Supplementary Figure captions

Fig. S1-S36. Time-series data of terrestrial water storage anomaly (ΔTWS) from GRACE and individual water stores from GLDAS Land Surface Models (LSMs): (a) Ensemble monthly GRACE ΔTWS from three solutions (CSR, Mascons, GRGS), (b-c) ensemble monthly ΔSMS and ΔSWS + ΔSNS from four GLDAS LSMs (CLM, Noah, VIC, Mosaic), (d) computed monthly ΔGWS and (e) monthly precipitation from August 2002 to July 2016 for WHYMAP large aquifer systems from no. 1 to 37, except for the High Plains Aquifer System in the USA (no. 17) which is shown as Fig. 2. Values in the Y-axis of the top four panels show monthly water-storage anomalies (cm) and the bottom panel shows monthly precipitation (cm).

Fig. S37. Seasonal-trend decomposition (STL) of an ensemble monthly GRACE ΔTWS time-series data from three solutions (CSR, Mascons, GRGS) for the High Plains Aquifer System in the USA (no. 17)

Fig. S38. Comparison of ΔGWS time series datasets from GRACE and in situ piezometric observations. ΔGWS time-series records in the unconsolidated, alluvial aquifer within the humid Bengal Basin of Bangladesh (a), and a weathered crystalline rock aquifer in the semi- arid Limpopo Basin in South Africa (b). Shaded bands in light-pink and light-blue around ΔGWS lines denote the standard deviation of the estimated mean value. Monthly precipitation records are shown as stair-step lines in blue on both panels with a dashed blue line indicating the long-term monthly mean value.

Fig. S39. Time-series data of terrestrial water storage anomaly (ΔTWS) from GRACE and individual water stores from GLDAS Land Surface Models (LSMs): (a) Ensemble monthly GRACE ΔTWS from three solutions (CSR, Mascons, GRGS), (b-c) ensemble monthly ΔSMS and ΔSWS + ΔSNS from four GLDAS LSMs (CLM, Noah, VIC, Mosaic), (d) computed monthly ΔGWS and (e) monthly precipitation from August 2002 to July 2016 for the Bengal Basin of Bangladesh.

Fig. S40. Time-series data of terrestrial water storage anomaly (ΔTWS) from GRACE and individual water stores from GLDAS Land Surface Models (LSMs): (a) Ensemble monthly GRACE ΔTWS from three solutions (CSR, Mascons, GRGS), (b-c) ensemble monthly

10 ΔSMS and ΔSWS + ΔSNS from four GLDAS LSMs (CLM, Noah, VIC, Mosaic), (d) computed monthly ΔGWS and (e) monthly precipitation from August 2002 to July 2016 for the Limpopo Basin of South Africa.

Fig. S41-S77. Decomposition of time-series components of GRACE ΔGWS data for all 37 large aquifer systems using a non-parametric, Seasonal-Trend decomposition procedure based on Loess (STL) technique. The graph on the top shows monthly anomaly data and decomposed time-series components of the series into seasonal, trend and remainder components. The graph also shows monthly precipitation records from CRU data along with annual total precipitation values (cm) written next to each annual cycle. The pie chart at the bottom shows the proportion of variance (%) in GRACE ΔGWS explained by the time-series components (S=seasonal, T=trend, R=residual/remainder).

Fig. S78. Time series of ensemble mean GRACE ΔTWS (red), GLDAS ΔSMS (green), ΔSWS + ΔSNS (blue) and computed GRACE ΔGWS (black) showing the calculation of anomalously negative or positive values of GRACE ΔGWS that deviate substantially from underlying trends. Cases for the Upper Kalahari-Cuvelai-Zambezi Basin (11) under a semi- arid climate, the Congo Basin (10) under a tropical humid climate, and the Angara-Lena Basin (27) in given in Fig. 6.

Fig. S79. Uncertainty in the estimates of GRACE-derived GWS from 20 realisations using 3 GRACE (CSR, JPL-Mascons, GRGS) products and 4 LSMs (CLM, Noah, VIC, Mosaic) and their ensemble means for the WHYMAP large aquifer systems from no. 1 to 37, except for the High Plains Aquifer System in the USA (no. 17) which is shown as Fig. 2.

Fig. S80. Long-term in-situ groundwater-level records from Makutapora Wellfield in central Tanzania (n=5) and Limpopo Basin in South Africa (n=4). Groundwater levels of from Makutapora site are referenced to a fixed datum (mean sea level, msl) and thus indicate piezometric heads; records in Limpopo Basin are plotted below groundwater level (bgl), hence the negative sign. The frequency of these multi-decadal groundwater-level records varies from daily to monthly with irregular time intervals. Annual rainfall time-series records are shown below each graph with a dashed, grey line denoting the mean of the time series.

Supplementary Fig. S1–S80:

11 Water storage (cm) −0.5 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0.0 0.1 0.2 0.3 0.4 0.0 0.5 −3 −2 −1 −2 0 2 0 1 2 3 Basin−averaged 2004 Fig. S1:NubianSandstoneAquifer System(1) 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 12 2010 2012 Linear trend 2014 2016 Water storage (cm) −1.0 −0.5 0.00 0.05 0.10 0.15 0.20 0.0 0.5 1.0 1.5 2.0 2.5 0.0 0.5 1.0 1.5 2.0 −4 −3 −2 −1 −4 −2 0 2 0 1 2 Basin−averaged 2004 Fig. S2: Northwestern Sahara Aquifer System(2) Fig. S2:Northwestern 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 13 2010 2012 Linear trend 2014 2016 Water storage (cm) −0.6 −0.4 −0.2 0.0 0.2 0.4 0.6 0.8 0.0 0.1 0.2 0.3 0.4 0.5 0.0 0.2 0.4 0.6 −4 −2 −2 0 2 0 2 Basin−averaged 2004 2006 Fig. S3:Murzuk−DjadoBasin(3) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 14 2010 2012 Linear trend 2014 2016 Water storage (cm) 0.00 0.05 0.10 0.15 0.20 0.25 −3 −2 −1 −3 −2 −1 −1 Basin−averaged 0 1 2 3 4 0 1 2 0 1 2 0 1 2 2004 2006 Fig. S4:Taoudeni−Tanezrouft Basin(4) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 15 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 0.0 0.5 1.0 1.5 2.0 2.5 −5 −5 −5 10 15 20 25 10 10 10 15 0 5 0 5 0 5 0 5 Basin−averaged 2004 2006 Fig. S5: Senegal−Mauritanian Basin(5) Fig. S5:Senegal−Mauritanian (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 16 2010 2012 Linear trend 2014 2016 Water storage (cm) 0.0 0.2 0.4 0.6 0.8 1.0 −2 −5 −2 10 15 0 5 0 2 4 0 2 4 6 0 5 Basin−averaged 2004 Fig. S6:Iullemmeden−Irhazer Aquifer System(6) 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 17 2010 2012 Linear trend 2014 2016 Water storage (cm) −0.2 0.0 0.2 0.4 0.6 0.8 1.0 −4 −2 −2 −5 10 15 10 0 2 4 6 8 0 5 0 5 0 2 4 6 Basin−averaged 2004 2006 Fig. S7:Lake ChadBasin(7) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 18 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −10 0.0 0.5 1.0 −4 −2 −5 −5 10 15 20 10 10 15 0 5 0 5 0 5 0 2 4 6 8 Basin−averaged 2004 Fig. S8:UmmRuwaba Aquifer (SuddBasin)(8) 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 19 2010 2012 Linear trend 2014 2016 Water storage (cm) 0.0 0.5 1.0 −2 −4 −2 −4 −2 10 0 5 0 2 0 2 4 0 2 4 Basin−averaged 2004 2006 Fig. S9:Ogaden−Juba Basin(9) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 20 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −15 −10 −10 −5 −5 −5 10 15 20 25 10 15 10 15 10 0 5 5 0 5 0 1 2 0 5 Basin−averaged 2004 2006 Fig. S10:CongoBasin(10) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 21 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −20 −10 −30 −20 −10 10 15 20 10 20 10 10 20 10 20 30 0 0 5 0 0 5 0 Basin−averaged 2004 Fig. S11:UpperKalahari−Cuvelai−Zambezi Basin(11) 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 22 2010 2012 Linear trend 2014 2016 Water storage (cm) 0.0 0.2 0.4 0.6 0.8 1.0 −4 −2 −4 −2 −4 −2 10 0 5 0 2 4 0 2 4 6 0 2 4 6 8 Basin−averaged 2004 Fig. S12: Lower Kalahari−Stampriet Basin(12) Fig. S12:Lower Kalahari−Stampriet 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 23 2010 2012 Linear trend 2014 2016 Water storage (cm) −0.2 0.0 0.2 0.4 0.6 0.8 −5 −4 −2 10 0 2 4 0 5 0 5 0 5 Basin−averaged 2004 2006 Fig. S13:KarooBasin(13) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 24 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −4 −2 −5 −5 10 12 10 10 15 0 5 0 2 4 6 8 0 5 0 2 4 6 0 2 4 6 Basin−averaged 2004 Fig. S14: Northern GreatPlainsAquifer (14) Fig. S14:Northern 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 25 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −15 −10 −10 −5 −5 −5 10 15 20 10 10 15 10 0 5 5 0 5 0 2 4 6 0 5 Basin−averaged 2004 Fig. S15:Cambro−Ordovician Aquifer System(15) 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 26 2010 2012 Linear trend 2014 2016 Water storage (cm) −20 −30 −20 −10 0.0 0.5 1.0 1.5 2.0 −5 10 15 20 10 10 15 20 0 0 5 0 0 5 Basin−averaged 2004 Fig. S16:California Central Valley Aquifer System(16) 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 27 2010 2012 Linear trend 2014 2016 Water storage (cm) −0.5 −10 −15 −10 0.0 0.5 1.0 1.5 2.0 −5 −5 −5 10 15 20 10 10 15 0 5 0 5 5 0 5 Basin−averaged 2004 Fig. S17:AtlanticandGulfCoastalPlainsAquifer (18) 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 28 2010 2012 Linear trend 2014 2016 Water storage (cm) −15 −10 −20 −10 −2 −5 10 20 30 40 10 20 10 20 0 0 0 2 4 0 5 Basin−averaged 2004 2006 Fig. S18:Amazon Basin(19) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 29 2010 2012 Linear trend 2014 2016 Water storage (cm) −20 −10 −20 −15 −10 −5 10 20 30 40 10 15 10 10 20 30 20 40 0 0 0 5 0 5 0 Basin−averaged 2004 2006 Fig. S19:Maranhao Basin(20) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 30 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −5 −5 10 15 20 25 10 10 10 20 0 5 0 5 0 2 4 0 5 Basin−averaged 2004 Fig. S20:Guarani Aquifer System(Parana Basin)(21) 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 31 2010 2012 Linear trend 2014 2016 Water storage (cm) −1.0 −0.5 0.5 1.0 1.5 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.5 1.0 1.5 −6 −4 −2 −6 −4 −2 0 2 0 2 Basin−averaged 2004 2006 Fig. S21:(22) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 32 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −10 0.0 0.5 1.0 1.5 −2 −5 −5 10 15 10 0 5 0 5 0 5 0 2 4 Basin−averaged 2004 2006 Fig. S22:IndusRiver Basin(23) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 33 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −20 −30 −20 −10 −5 10 20 30 40 50 10 20 10 15 20 0 0 0 0 2 4 6 0 5 Basin−averaged 2004 Fig. S23:Ganges−Brahmaputra River Basin(24) 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 34 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −10 −10 −10 −5 −5 −5 10 20 10 15 20 10 10 15 0 5 2 4 6 8 0 0 5 0 5 Basin−averaged 2004 Fig. S24: West Siberian Artesian Basin(25) Fig. S24:West Artesian Siberian 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 35 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −10 −10 −5 −5 −5 10 20 10 10 15 0 5 2 4 6 8 0 0 5 0 5 Basin−averaged 2004 2006 Fig. S25:Tunguss Basin(26) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 36 2010 2012 Linear trend 2014 2016 Water storage (cm) −15 −10 −10 −6 −4 −2 −5 −5 10 10 10 0 5 0 2 4 6 8 0 5 0 5 0 2 4 Basin−averaged 2004 2006 Fig. S26:Angara−Lena Basin(27) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 37 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −5 −5 −5 10 10 0 5 0 2 4 6 8 0 5 0 5 0 5 Basin−averaged 2004 2006 Fig. S27:Yakut Basin(28) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 38 2010 2012 Linear trend 2014 2016 Water storage (cm) −15 −10 −15 −10 −5 −5 −5 10 15 20 25 10 0 5 0 5 0 5 0 1 2 3 0 5 Basin−averaged 2004 Fig. S28: North ChinaPlainsAquifer System(29) Fig. S28:North 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 39 2010 2012 Linear trend 2014 2016 Water storage (cm) −4 −2 −5 −4 −2 10 15 20 10 0 5 0 2 4 6 0 1 2 3 4 5 0 2 4 6 8 0 5 Basin−averaged 2004 2006 Fig. S29:Song−LiaoPlain(30) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 40 2010 2012 Linear trend 2014 2016 Water storage (cm) −0.5 0.0 0.5 1.0 0.0 0.5 1.0 −4 −2 −2 0 2 0 1 2 3 0 2 Basin−averaged 2004 2006 Fig. S30:Tarim Basin(31) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 41 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −10 −10 0.0 0.5 1.0 1.5 2.0 −5 −5 −5 10 15 10 15 10 15 0 5 5 0 5 0 5 Basin−averaged 2004 2006 Fig. S31:Paris Basin(32) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 42 2010 2012 Linear trend 2014 2016 Water storage (cm) −15 −10 −15 −10 −15 −10 −5 −5 −5 −5 10 10 10 15 10 10 0 5 2 4 6 8 0 5 0 5 0 5 Basin−averaged 2004 Fig. S32:EastEuropeanAquifer System(33) 2006 (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 43 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −10 −10 −5 −5 10 10 10 0 2 4 6 8 0 5 0 2 4 6 8 0 5 Basin−averaged 2004 2006 Fig. S33: North CaucasusBasin(34) Fig. S33:North (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 44 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −20 −10 −10 −10 10 10 20 10 20 30 10 10 0 2 4 6 8 0 0 0 Basin−averaged 2004 2006 Fig. S34:Pechora Basin(35) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 45 2010 2012 Linear trend 2014 2016 Water storage (cm) 0.0 0.5 1.0 −4 −2 −5 10 15 10 10 15 0 5 0 5 0 2 4 6 0 5 Basin−averaged 2004 2006 Fig. S35: Great Artesian Basin(36) Fig. S35:GreatArtesian (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 46 2010 2012 Linear trend 2014 2016 Water storage (cm) −10 −10 0.0 0.5 1.0 1.5 −5 −5 10 15 20 10 10 10 20 0 0 5 0 5 0 5 Basin−averaged 2004 2006 Fig. S36:CanningBasin(37) (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 47 2010 2012 Linear trend 2014 2016 Fig. S37: STL decomposition of ensemble GRACE−TWS signal High Plains Aquifer System (17) 5 0 data −5 −10 3 2 1 0 −1 seasonal −2 −3 4 2 0 −2 trend −4 −6 2 0 −2 remainder −4

2004 2006 2008 2010 2012 2014 2016 time

48 Fig. S38 (a)

(b)

49 Water storage (cm) −10 −20 −30 −20 −10 100 20 40 60 80 10 20 10 15 10 20 40 0 0 0 0 5 0 Basin−averaged 2004 2006 (c) Ensemble (SWS+SNS) Fig. S39:BengalBasin 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 50 2010 2012 Linear trend 2014 2016 Water storage (cm) 0.0 0.5 1.0 1.5 −4 −2 −5 −4 −2 10 15 20 10 0 5 0 2 4 6 0 2 4 6 0 5 Basin−averaged 2004 2006 Fig. S40:LimpopoBasin (c) Ensemble (SWS+SNS) 2008 Non−linear trend (d) Ensemble GWS (a) Ensemble TWS (b) Ensemble SMS (e) Precipitation 51 2010 2012 Linear trend 2014 2016 Groundwater storage (cm)

−3 −2 −1 0 1 2 3 3 2004 3 4 Fig. S41:NubianSandstoneAquifer System(1) 2006 data 3 T= 30% Seasonal−Trend−Remainder (GRACE GWS) 3 2008 3 Proportion ofvariance trend 4 52 2010 R= 49% 3 seasonal 3 2012 3 S= 21% 4 remainder 2014 4 3 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−4 −2 0 2 8 2004 10 8 T= 56% Fig. S42: Northwestern SaharaAquifer System(2) Fig. S42:Northwestern 2006 data 10 Seasonal−Trend−Remainder (GRACE GWS) 8 2008 7 Proportion ofvariance trend 10 53 2010 5 seasonal 7 2012 R= 29% 7 S= 15% 5 remainder 2014 7 6 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−4 −2 0 2 1 2004 1 1 T= 31% 2006 data 1 Seasonal−Trend−Remainder (GRACE GWS) Fig. S43:Murzuk−DjadoBasin(3) 1 2008 1 Proportion ofvariance trend 1 54 2010 1 seasonal R= 39% 1 2012 S= 29% 1 2 remainder 2014 1 1 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−4 −3 −2 −1 0 1 2 7 2004 7 6 2006 T= 32% data 6 Fig. S44:Taoudeni−Tanezrouft Basin(4) Seasonal−Trend−Remainder (GRACE GWS) 7 2008 6 Proportion ofvariance trend 7 55 2010 9 R= 46% seasonal 6 2012 6 S= 22% 6 remainder 2014 7 7 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−5 0 5 10 59 2004 47 57 T= 52% 2006 data 51 Fig. S45:Senegal−MauritanianBasin(5) Seasonal−Trend−Remainder (GRACE GWS) 45 2008 58 Proportion ofvariance trend 61 56 2010 71 seasonal 48 2012 58 S= 27% R= 21% 57 remainder 2014 47 51 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−4 −2 0 2 4 37 2004 29 30 Fig. S46: Iullemmeden−Irhazer AquiferFig. S46:Iullemmeden−Irhazer System(6) 2006 data 28 T= 53% Seasonal−Trend−Remainder (GRACE GWS) 31 2008 32 Proportion ofvariance trend 28 57 2010 32 seasonal 26 R= 37% 2012 37 29 remainder S= 10% 2014 28 31 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−6 −4 −2 0 2 4 6 8 38 2004 40 43 2006 data 42 Seasonal−Trend−Remainder (GRACE GWS) 46 T= 13% Fig. S47:LakeChadBasin(7) S= 62% 2008 44 Proportion ofvariance trend 35 58 2010 46 seasonal 39 2012 48 R= 25% 38 remainder 2014 40 42 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−15 −10 −5 0 5 10 15 84 2004 66 71 S= 82% Fig. S48:UmmRuwabaAquifer (Sudd Basin)(8) 2006 data 81 Seasonal−Trend−Remainder (GRACE GWS) 89 2008 87 Proportion ofvariance trend 67 59 2010 86 seasonal 77 2012 T= 7% 77 80 remainder R= 11% 2014 82 75 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−4 −2 0 2 4 37 2004 45 36 2006 T= 42% data 42 Seasonal−Trend−Remainder (GRACE GWS) Fig. S49:Ogaden−Juba Basin(9) 41 2008 38 Proportion ofvariance trend 37 60 2010 36 seasonal R= 40% 41 2012 39 S= 18% 53 remainder 2014 40 44 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−10 −5 0 5 10 15 20 166 2004 154 162 2006 162 data T= 45% Seasonal−Trend−Remainder (GRACE GWS) 164 Fig. S50:CongoBasin(10) 2008 168 Proportion ofvariance trend 165 61 2010 168 R= 42% seasonal 150 2012 166 161 S= 13% remainder 2014 161 159 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−30 −20 −10 0 10 20 30 81 2004 88 73 T= 63% Fig. S51:UpperKalahari−Cuvelai−ZambeziBasin(11) 2006 data 89 Seasonal−Trend−Remainder (GRACE GWS) 82 2008 81 Proportion ofvariance trend 87 62 2010 88 seasonal 86 2012 85 R= 22% S= 15% 78 remainder 2014 87 75 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−4 −2 0 2 4 6 26 2004 31 29 2006 Fig. S52:Lower Kalahari−StamprietBasin(12) data 39 T= 41% Seasonal−Trend−Remainder (GRACE GWS) 24 2008 32 Proportion ofvariance trend 27 63 2010 38 R= 44% seasonal 41 2012 29 S= 15% 21 remainder 2014 36 24 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−5 0 5 10 44 2004 47 47 T= 83% 2006 data 51 Seasonal−Trend−Remainder (GRACE GWS) 51 Fig. S53:Karoo Basin(13) 2008 53 Proportion ofvariance trend 46 64 2010 53 seasonal 67 2012 41 48 remainder R= 11% S= 6% 2014 50 45 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−5 0 5 10 15 41 2004 46 T= 76% 50 2006 data 39 Fig. S54: Northern GreatPlainsAquifer (14) Fig. S54:Northern Seasonal−Trend−Remainder (GRACE GWS) 48 2008 46 Proportion ofvariance trend 44 65 2010 52 seasonal 49 2012 43 52 S= 13% remainder R= 11% 2014 46 46 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−15 −10 −5 0 5 10 15 72 2004 88 T= 42% 70 Fig. S55:Cambro−Ordovician Aquifer System(15) 2006 data 86 Seasonal−Trend−Remainder (GRACE GWS) 91 2008 97 Proportion ofvariance trend 90 66 2010 99 seasonal 87 2012 R= 29% S= 29% 73 105 remainder 2014 95 97 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−30 −20 −10 0 10 20 62 2004 56 78 Fig. S56:California CentralValley Aquifer System(16) T= 64% 2006 data 84 Seasonal−Trend−Remainder (GRACE GWS) 33 2008 44 Proportion ofvariance trend 43 67 2010 62 seasonal 65 2012 44 S= 24% 42 remainder R= 12% 2014 30 47 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−10 −5 0 5 42 2004 58 52 T= 77% 2006 data 47 Fig. S57:OgallalaAquifer (High Plains)(17) Seasonal−Trend−Remainder (GRACE GWS) 55 2008 55 Proportion ofvariance trend 54 68 2010 55 seasonal 44 2012 33 R= 14% 48 remainder S= 9% 2014 55 67 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−5 0 5 131 2004 134 113 Fig. S58:AtlanticandGulfCoastalPlainsAquifer (18) 2006 112 data T= 51% Seasonal−Trend−Remainder (GRACE GWS) 106 2008 122 Proportion ofvariance trend 137 69 2010 104 seasonal 111 R= 39% 2012 125 140 remainder S= 10% 2014 125 149 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−20 −10 0 10 20 30 269 2004 252 248 2006 283 data S= 72% Seasonal−Trend−Remainder (GRACE GWS) 253 Fig. S59:AmazonBasin(19) 2008 290 Proportion ofvariance trend 275 70 2010 237 seasonal T= 16% 235 2012 228 236 remainder R= 12% 2014 246 238 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−20 −10 0 10 136 2004 169 139 T= 14% 2006 153 data Seasonal−Trend−Remainder (GRACE GWS) 151 Fig. S60:MaranhaoBasin(20) 2008 163 Proportion ofvariance trend 179 71 2010 140 R= 48% seasonal 178 S= 39% 2012 135 147 remainder 2014 160 134 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−5 0 5 10 15 158 2004 135 137 Fig. S61:GuaraniAquifer System(Parana Basin)(21) 2006 126 data T= 57% Seasonal−Trend−Remainder (GRACE GWS) 151 2008 130 Proportion ofvariance trend 126 72 2010 178 seasonal 136 R= 36% 2012 136 138 remainder S= 7% 2014 157 160 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−6 −4 −2 0 2 7 2004 8 T= 90% 7 2006 data 7 Fig. S62:ArabianAquifer System(22) Seasonal−Trend−Remainder (GRACE GWS) 9 2008 6 Proportion ofvariance trend 7 73 2010 7 seasonal 7 2012 7 7 remainder R= 7% S= 3% 2014 8 7 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−10 −5 0 5 42 2004 28 37 T= 34% 2006 data 44 Seasonal−Trend−Remainder (GRACE GWS) 40 Fig. S63:IndusRiverBasin(23) 2008 42 Proportion ofvariance trend 28 74 2010 49 R= 43% seasonal 39 2012 32 S= 23% 38 remainder 2014 33 46 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−30 −20 −10 0 10 20 30 147 2004 137 124 Fig. S64: Ganges−Brahmaputra RiverBasin(IGB)(24) Fig. S64:Ganges−Brahmaputra 2006 131 data Seasonal−Trend−Remainder (GRACE GWS) 147 T= 45% 2008 148 Proportion ofvariance trend 124 75 2010 138 S= 48% seasonal 152 2012 149 151 remainder R= 7% 2014 126 146 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−20 −10 0 10 20 44 2004 50 45 2006 data 48 Fig. S65:West Basin(25) SiberianArtesian Seasonal−Trend−Remainder (GRACE GWS) 52 2008 50 S= 55% Proportion ofvariance trend T= 40% 46 76 2010 48 seasonal 49 2012 43 51 remainder R= 5% 2014 54 57 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−10 0 10 20 30 43 2004 45 T= 84% 41 2006 data 38 Seasonal−Trend−Remainder (GRACE GWS) 49 Fig. S66:Tunguss Basin(26) 2008 49 Proportion ofvariance trend 40 77 2010 46 seasonal 45 2012 36 42 remainder S= 9% R= 7% 2014 46 48 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−10 −5 0 5 10 45 2004 51 50 T= 49% 2006 data 52 Seasonal−Trend−Remainder (GRACE GWS) Fig. S67:Angara−LenaBasin(27) 43 2008 52 Proportion ofvariance trend 54 78 2010 47 seasonal 45 R= 35% 2012 46 S= 16% 52 remainder 2014 43 45 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−10 −5 0 5 10 15 36 2004 40 43 2006 T= 54% data 41 Seasonal−Trend−Remainder (GRACE GWS) 43 Fig. S68:Yakut Basin(28) 2008 43 Proportion ofvariance trend 38 79 2010 37 seasonal 37 S= 35% 2012 40 43 remainder R= 11% 2014 35 34 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−20 −15 −10 −5 0 5 10 111 2004 90 89 T= 55% Fig. S69: North ChinaPlainsAquifer System(29) Fig. S69:North 2006 data 75 Seasonal−Trend−Remainder (GRACE GWS) 92 2008 83 Proportion ofvariance trend 88 80 2010 86 seasonal 81 2012 75 S= 25% R= 20% 72 remainder 2014 79 91 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−5 0 5 58 2004 48 55 T= 49% 2006 data 53 Seasonal−Trend−Remainder (GRACE GWS) 41 Fig. S70:Song−LiaoPlain(30) 2008 50 Proportion ofvariance trend 53 81 2010 62 seasonal 47 R= 31% 2012 64 S= 20% 67 remainder 2014 37 51 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−4 −2 0 2 4 13 2004 8 14 T= 15% 2006 data 8 Seasonal−Trend−Remainder (GRACE GWS) 10 Fig. S71:Tarim Basin(31) 2008 9 Proportion ofvariance trend 7 82 2010 12 S= 47% seasonal 6 R= 38% 2012 10 11 remainder 2014 7 10 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−10 −5 0 5 10 78 2004 65 T= 60% 70 2006 data 66 Seasonal−Trend−Remainder (GRACE GWS) 88 Fig. S72:Paris Basin(32) 2008 77 Proportion ofvariance trend 73 83 2010 68 seasonal 71 2012 81 S= 21% R= 19% 83 remainder 2014 83 68 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−20 −15 −10 −5 0 5 10 15 61 2004 67 57 2006 Fig. S73:EastEuropean Aquifer System(33) data 61 S= 68% Seasonal−Trend−Remainder (GRACE GWS) 62 2008 63 Proportion ofvariance trend 62 84 2010 57 T= 18% seasonal 59 2012 69 62 R= 14% remainder 2014 55 60 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−15 −10 −5 0 5 10 15 52 2004 66 57 T= 39% 2006 data 53 Fig. S74: North CaucasusBasin(34) Fig. S74:North Seasonal−Trend−Remainder (GRACE GWS) 43 2008 51 Proportion ofvariance trend 59 85 2010 56 seasonal S= 39% 56 2012 46 R= 23% 57 remainder 2014 48 49 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−20 −10 0 10 20 30 61 2004 53 58 2006 data 57 T= 52% Seasonal−Trend−Remainder (GRACE GWS) 68 Fig. S75:Pechora Basin(35) 2008 57 Proportion ofvariance trend 53 86 2010 58 seasonal S= 40% 47 2012 57 52 remainder R= 8% 2014 61 59 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−5 0 5 10 31 2004 44 34 T= 71% 2006 data 45 Seasonal−Trend−Remainder (GRACE GWS) Fig. S76: Great Artesian Basin(36) Fig. S76:GreatArtesian 37 2008 45 Proportion ofvariance trend 53 87 2010 60 seasonal 70 2012 53 R= 23% 33 remainder S= 6% 2014 32 41 2016

0 10 20 30 40 50 60 Rainfall (cm) Groundwater storage (cm)

−15 −10 −5 0 5 10 15 49 2004 57 24 T= 81% 2006 data 58 Seasonal−Trend−Remainder (GRACE GWS) 45 Fig. S77:CanningBasin(37) 2008 40 Proportion ofvariance trend 39 88 2010 33 seasonal 66 2012 43 R= 15% 37 remainder S= 4% 2014 57 35 2016

0 10 20 30 40 50 60 Rainfall (cm) Storage anomaly (cm) Storage anomaly (cm) Storage anomaly (cm) Storage anomaly (cm) Storage anomaly (cm) −10 −5 0 5 −4 0 2 4 6 8 −4 −2 0 2 4 −4 0 2 4 6 8 −3 −1 0 1 2 2004 2004 2004 2004 2004 TWS 2006 2006 2006 2006 2006 Iullemmeden−Irhazer Aquifer System(6) Lower Kalahari−Stampriet Basin(12) Lower Kalahari−Stampriet Ogallala Aquifer (HighPlains)(17) Taoudeni−Tanezrouft Basin(4) 2008 2008 2008 2008 2008 Ogaden−Juba Basin(9) SMS 2010 2010 2010 2010 2010 2012 2012 2012 2012 2012 2014 2014 2014 2014 2014 Legend SWS+SNS 89 2016 2016 2016 2016 2016

0 5 10 15 20 25 30 0 5 10 15 20 25 0 5 10 15 20 25 0 5 10 15 20 25 30 0 5 GWS Precipitation (cm) Precipitation (cm) Precipitation (cm) Precipitation (cm) Precipitation (cm) Iullemmeden−Irhazer Aquifer System(6) Lower Kalahari−Stampriet Basin(12) Lower Kalahari−Stampriet 2013.0 2010.0 2014.0 2008.0 2010.0 −10 −5 0 5 Ogallala Aquifer (HighPlains)(17) −4 −2 0 2 4 −4 −2 0 2 4 −2 0 2 4 6 −3 −2 −1 0 1 2 Taoudeni−Tanezrouft Basin(4) Ogaden−Juba Basin(9) Precipitation 2014.0 2011.0 2015.0 2009.0 2011.0 2015.0 2012.0 2016.0 2010.0 2012.0

0 5 10 15 20 25 30 0 5 10 15 0 5 10 15 0 5 10 15 20 25 30 0 5 GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm)

−5 0 5 10 −15 −10 −5 0 5 10 15 −5 0 5 −5 0 5 10 2004 2004 2004 2004 Nubian SandstoneAquifer System(1) Senegal−Mauritanian Basin(5) Senegal−Mauritanian 2006 2006 2006 2006 Murzuk−Djado Basin(3) Lake Chad Basin(7) 2008 2008 2008 2008 2010 2010 2010 2010 2012 2012 2012 2012 2014 2014 2014 2014 2016 2016 2016 2016 5% 50% 95% 5% 50% 95% 5% 50% 95% 5% 50% 95% 90

GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm)

−25 −15 −5 0 5 10 15 20 −5 0 5 10 −5 0 5 −5 0 5 2004 2004 2004 2004 Northwestern Sahara Aquifer System(2) Northwestern Iullemmeden−Irhazer Aquifer System(6) Umm Ruwaba Aquifer (SuddBasin)(8) Taoudeni−Tanezrouft Basin(4) 2006 2006 2006 2006 2008 2008 2008 2008 2010 2010 2010 2010 2012 2012 2012 2012 2014 2014 2014 2014 2016 2016 2016 2016 5% 50% 95% 5% 50% 95% 5% 50% 95% 5% 50% 95% GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm)

−20 −10 0 5 10 15 20 25 −10 −5 0 5 10 15 −50 −35 −20 −5 5 15 25 35 −5 0 5 Upper Kalahari−Cuvelai−Zambezi Basin(11) 2004 2004 2004 2004 Cambro−Ordovician Aquifer System(15) 2006 2006 2006 2006 Ogaden−Juba Basin(9) Karoo Basin(13) 2008 2008 2008 2008 2010 2010 2010 2010 2012 2012 2012 2012 2014 2014 2014 2014 2016 2016 2016 2016 5% 50% 95% 5% 50% 95% 5% 50% 95% 5% 50% 95% 91

GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm)

−50 −35 −20 −5 5 15 25 35 −15 −5 0 5 10 15 20 25 −10 −5 0 5 10 −35 −25 −15 −5 5 15 25 California Central Valley Aquifer System(16) 2004 2004 2004 2004 Lower Kalahari−Stampriet Basin(12) Lower Kalahari−Stampriet Northern GreatPlainsAquifer (14) Northern 2006 2006 2006 2006 Congo Basin(10) 2008 2008 2008 2008 2010 2010 2010 2010 2012 2012 2012 2012 2014 2014 2014 2014 2016 2016 2016 2016 5% 50% 95% 5% 50% 95% 5% 50% 95% 5% 50% 95% GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm)

−50 −35 −20 −5 5 15 25 35 −10 −5 0 5 −45 −30 −15 −5 5 15 25 35 −15 −10 −5 0 5 10 15 20 Ganges−Brahmaputra River Basin(IGB)(24) Atlantic andGulfCoastalPlainsAquifer (18) 2004 2004 2004 2004 Arabian Aquifer System(22) 2006 2006 2006 2006 Maranhao Basin(20) 2008 2008 2008 2008 2010 2010 2010 2010 2012 2012 2012 2012 2014 2014 2014 2014 2016 2016 2016 2016 5% 50% 95% 5% 50% 95% 5% 50% 95% 5% 50% 95% 92

GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm)

−30 −20 −10 0 10 20 30 40 −20 −15 −10 −5 0 5 10 15 20 −15 −5 0 5 10 15 20 25 −30 −20 −10 0 5 15 25 35 Guarani Aquifer System(Parana Basin)(21) 2004 2004 2004 2004 West Siberian Artesian Basin (25) West Artesian Siberian 2006 2006 2006 2006 Indus River Basin(23) Amazon Basin(19) 2008 2008 2008 2008 2010 2010 2010 2010 2012 2012 2012 2012 2014 2014 2014 2014 2016 2016 2016 2016 5% 50% 95% 5% 50% 95% 5% 50% 95% 5% 50% 95% GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm)

−25 −15 −5 0 5 10 15 20 −10 −5 0 5 10 −15 −5 0 5 10 15 20 25 30 −20 −10 0 10 20 30 40 50 2004 2004 2004 2004 2006 2006 2006 2006 Song−Liao Plain(30) Tunguss Basin(26) Yakut Basin(28) Paris Basin(32) 2008 2008 2008 2008 2010 2010 2010 2010 2012 2012 2012 2012 2014 2014 2014 2014 2016 2016 2016 2016 5% 50% 95% 5% 50% 95% 5% 50% 95% 5% 50% 95% 93

GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm) GWS anomaly (cm)

−20 −10 0 5 10 20 30 −15 −10 −5 0 5 10 −30 −20 −10 0 5 10 15 20 −15 −10 −5 0 5 10 15 20 2004 2004 2004 2004 North ChinaPlainsAquifer System(29) North East EuropeanAquifer System(33) 2006 2006 2006 2006 Angara−Lena Basin(27) Tarim Basin(31) 2008 2008 2008 2008 2010 2010 2010 2010 2012 2012 2012 2012 2014 2014 2014 2014 2016 2016 2016 2016 5% 50% 95% 5% 50% 95% 5% 50% 95% 5% 50% 95% GWS anomaly (cm) GWS anomaly (cm)

−15 −10 −5 0 5 10 15 20 −30 −20 −10 0 5 10 15 20 2004 2004 2006 2006 North CaucasusBasin(34) North Great Artesian Basin(36) Great Artesian 2008 2008 2010 2010 2012 2012 2014 2014 2016 2016 50% 95% 5% 50% 95% 5% 94

GWS anomaly (cm) GWS anomaly (cm)

−20 −10 −5 0 5 10 15 20 −35 −25 −15 −5 5 15 25 35 45 2004 2004 2006 2006 Canning Basin(37) Pechora Basin(35) 2008 2008 2010 2010 2012 2012 2014 2014 2016 2016 5% 50% 95% 5% 50% 95% Fig. S80  !"# $%& '() "0 1234"5 36 32

B B B B  B  789@A8     

         

Fig. S80 E!"F2G'('("0 1234"H(%&I"PQ)2R

ST ST ST ST  CD 789@A8 B B B      

    

95