Accelerated water loss over Canada’s landmass in 2002–2024

Accelerated water loss over Canada’s landmass in 2002–2024 | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Accelerated water loss over Canada’s landmass in 2002–2024 Shusen Wang, Fuqun Zhou, Farzam Fatolazadeh, Hazen Russell, Melissa Bunn This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6253460/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Terrestrial water storage (TWS) changes significantly influence the global water cycle and the development of better-informed water policies. Here we investigate TWS variations across Canada’s landmass using the GRACE and GRACE-FO satellites observations. We show that Canada’s TWS exhibited an accelerated downward trend in 2002–2024, resulting in a total water loss of 2430 km 3 that corresponds to a global sea-level rise of 6.9 mm. The loss was mainly driven by glacier and snow melt over the Pacific Cordillera and the Arctic Cordillera, as well as permafrost degradation in central Canada. For ice-free regions, the dry areas became drier, and the wet areas became wetter. For ice-covered regions, the water loss rate accelerated over the Pacific Cordillera but slowed down over the Arctic Cordillera and the permafrost regions. As water fluxes accelerate in a warming climate, it is expected that future TWS trends may experience further changes. Earth and environmental sciences/Hydrology Earth and environmental sciences/Climate sciences/Climate change Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Terrestrial Water Storage (TWS) is the sum of the water stored in terrestrial ecosystems, including surface water in rivers and lakes, soil water held in soil zones, groundwater stored in aquifers, and snow and ice covering land surface. Given its importance to the global water cycle and sensitivity to climate change impacts, TWS was added as one of 54 Essential Climate Variables (ECVs) by the Global Climate Observing System (GCOS) in 2022. Information on the status and trends of TWS plays a crucial role not only in understanding the global climate and water cycle but also in better-informed water resource management and policy development at various scales. The Gravity Recovery and Climate Experiment (GRACE) satellites and their successor, GRACE Follow-on (GRACE-FO), have been providing direct measurements of TWS changes since 2002 (Landerer et al., 2020). These measurements have been used to characterize the dynamics of TWS at various scales, including changes in its components such as groundwater storage, glacier loss and snow melt, and basin discharge (e.g., Famiglietti 2014; Rodell et al., 2018; Ciracì et al., 2020; Duvvuri et al., 2023; Li and Rodell, 2024). Canada has a vast landmass of nearly ten million square kilometers with a diverse range of cold region hydrological processes, including glaciers distributed over the Pacific Cordillera and Arctic Cordillera, permafrost in the north, a humid climate in the east, and a semi-arid climate in the Prairies. Frozen soil covers almost the entire country during winter. Snow accumulation and snowmelt-driven surface runoff are important components of the water cycle. The region is experiencing the fastest rates of climate change, which has highly impacted its water conditions. GRACE observations have significantly contributed to our understanding of cold-region hydrological processes in Canada, including estimating snow mass and snowmelt-driven river flows (Wang and Russell 2016; Wang et al., 2017), modelling groundwater discharge and freezing temperature control on aquifer conductivity (Wang 2019; Wang et al., 2021), quantifying groundwater storage change (Li and Wang, 2022), and assessing water budget closures (Wang et al., 2014; Wang et al., 2015). At a national scale, Wang and Li (2016) characterized Canada’s TWS climatology by retrieving a suite of parameters. However, this study was based on a relatively short GRACE record of 13 years (2002–2014), leaving open questions about which trends in TWS are persistent and indicative of longer-term changes, and which are driven by short-term climate variability. Here we present the TWS changes for Canada’s landmass using the latest GRACE and GRACE-FO TWS products over a 22-year period from April 2002 to March 2024, and revisit issues relating to long-term TWS changes. A total of six GRACE/GRACE-FO TWS products from both Spherical Harmonics and Mascon solutions were used in this study as detailed in the Methods section. Unless noted, the results represent the assemble means of the six products. We also evaluated the differences among these six TWS products over Canada’s landmass, with the results provided in the Supplementary Information. Results and Discussion Terrestrial water storage (TWS) trends in 2002–2024 The TWS for Canada’s entire landmass trended down almost continuously over 2002–2024 at an average rate of 11.1 mm year − 1 (Fig. 1 ). Over the 22-year study period, Canada’s landmass lost 2430 km 3 of freshwater, equivalent to 1.5 times the total water volume of Lake Ontario and accounting for 6.9 mm or 8.5% of the global sea level rise observed over the 22 years (Hamlington et al., 2024). As shown below, the loss of water was driven mainly by melting of glaciers and ice sheets, as well as the loss of ground ice and subsurface water due to permafrost degradation. Notably, the decreasing rate of TWS appeared to accelerate after 2020, averaging more than 26 mm year − 1 for the last four years. This accelerated water loss in Canada is associated with the recent development of severe drought conditions in south-central Canada. The TWS trends over Canada’s landmass showed large variations geographically. Figure 2 shows the Sen’s slope values calculated for each grid. Sen’s slope is a nonparametric estimate of the slope for a time-series dataset. A negative Sen’s slope value indicates an overall downward trend, signifying water loss over the entire 22-year study period, and vice versa. The most pronounced feature in the results is the large negative trends observed over the western Pacific Cordillera and the Arctic Cordillera, with Sen’s slope values in some areas exceeding − 200 mm year − 1 . Relatively, the decreasing trend in the western Pacific Cordillera, particularly in its northern part, was more pronounced than that in the Arctic Cordillera. Figure 3 shows the monthly TWS time series data for the Pacific Coastal Drainage Region, which covers an area of over 320×10 3 km 2 and largely represents the western Pacific Cordillera (see Fig. S5 in the Supplementary Information for location). The average downward trend for this drainage region reached about − 60 mm year − 1 . The data also revealed that the TWS loss accelerated in the latter part of the study period, starting in 2012, when the rate of loss exceeded − 85 mm year − 1 . Analyses of precipitation and water yield data (Li and Wang, 2021) over the Pacific Cordillera and the Arctic Cordillera regions indicate that these factors could not explain the downward TWS trends. Both regions have widespread distributions of glaciers, ice sheets, and permanent snow cover, all of which have experienced significant climate warming. Our results suggest that the TWS decreasing trends were mainly driven by the significant water loss from the melting of ice and snow cover, followed by decreases in sub-surface water storage once the ice or snow cover disappeared, as evidenced by numerous in situ glacier and ice mass balance studies (e.g., The GlaMBIE Team, 2025; Burgess et al., 2024). Another region that experienced significant large-scale decreasing trends in TWS is the central part of the landmass around latitude 60°N, including the far north of Quebec. The water loss rates in some parts of this region reached over − 10 mm year − 1 . The TWS downward trends are not explained by corresponding precipitation and water yield data (Li and Wang, 2021), which showed slightly positive trends during our study period. This region lies in the transition zone from permafrost to seasonal frozen soil, where significant soil thaw and permafrost degradation have occurred over the past decades due to climate change (Thibault and Payette, 2009; Pelletier et al., 2019). Several studies have revealed that permafrost thaw has resulted in changes of water cycle and energy balance over this region (Kurylyk et al., 2016; Göckede; et al., 2019). Permafrost thaw has activated soil drainage and aquifer discharge processes (Sergeant et al., 2021; Li and Wang, 2022). Aquifer conductivity and groundwater discharge can significantly increase with climate warming (Wang, 2019). Moreover, increased soil water drainage leads to drier soil surface, resulting in more surface energy being partitioned to sensible heat flux rather than latent heat flux. This change in surface energy balance leads to the increase in land surface temperature, which positively feeds back and further accelerates permafrost thaw processes. All these processes would contribute to the water loss over the region, as reported by ground-based studies (Bouchard et al., 2013). Regions around Hudson Bay are experiencing significant post glacial rebound. Some areas of the permafrost transition zone mentioned earlier are located over regions with the most dramatic isostatic rebound. The Glacial Isostatic Adjustment (GIA) in this region plays an important role in retrieving TWS trends from GRACE observations (Peltier et al., 2015, 2017). Due to the uncertainties in GIA models, the trends observed in the region likely have larger uncertainty compared to other regions. Regions exhibiting positive TWS trends at a large scale were mainly distributed in Eastern Canada, stretching from southern Ontario northeastward to Newfoundland. The magnitudes of these trends over some regions exceeded 10 mm year − 1 . The positive trends over this region were found to be associated with increasing precipitation and water yield, indicating that Eastern Canada experienced a net gain in TWS due to higher precipitation during the study period. Figure 4 shows the time series of monthly TWS variations for the St. Lawrence Drainage Region (see Fig. S5 in the Supplementary Information for location). It shows that the TWS rate of increase over the region was fairly consistent over the 22 years of 2002–2024. The TWS trends over other regions of the landmass were relatively small. This includes the south-central part of the country, notably the Canadian Prairies. The detailed interannual variations of TWS over this region will be discussed in the following section. Shift of trends The Prairie region over south-central Canada had overall small trends from 2002–2024, as shown in Fig. 2 . However, this region was reported to have significant positive trends in an earlier study using the GRACE data for 2002–2014 (Wang and Li, 2016). This suggests that there was a change or a shift in trends over different periods (sub-trends) within the 22-year timeframe. Figure 5 shows the monthly TWS time series for the South Saskatchewan Drainage Region, located in the centre of the Prairies (see Fig. S5 in the Supplementary Information for location). It is worth noting that the Prairies have a semi-arid climate, with annual total precipitation typically below 400 mm, while potential evaporation can exceed 1000 mm year − 1 (Li and Wang, 2020). The region experienced one of its most severe droughts in 2000 − 2003 since instrumental record was available. It resulted in large deficits in groundwater, surface water, and soil water (Hanesiak et al. 2011). Consequently, TWS was at a very low level at the beginning of the GRACE mission. During the subsequent 10 years following this drought event, the region saw substantial recovery as precipitation returned to normal levels, leading to a positive TWS trend. However, in the past 12 years, a new severe drought has emerged, causing a consistent decline in TWS. This prolonged dry period has completely reversed the water gains from the previous decade. By the end of our study period in March 2024, TWS in the region had dropped more than 50 mm below its April 2002 level. The small overall trend obtained for the full 22-year study period reflects the opposing effects of the initial positive sub-trend and the later negative sub-trend, which largely canceled each other out. Another notable feature of TWS over the Prairies is its large interannual variability. The region went through another short drought period in the seasons around 2009 which led to a substantial drop in TWS even during the first 10 years when it had an overall positive sub-trend. Similarly, the region received above-normal precipitation around 2017, which led to a significant increase in TWS, despite the region being in a period of overall decline. These findings highlight the importance of accounting for short-term sub-trends and extreme climate events when assessing long-term TWS trends in arid and semi-arid regions. To evaluate the possible shifts in TWS trends for the entire landmass in the study period, we calculated Sen’s slope for each grid over the period of 2002–2014. We then compared it with that from 2002–2024 to identify changes over time. Figure 6 shows the difference in Sen’s slope between these periods. A negative difference indicates that the region became “drier” in the later years, either due to accelerated water loss or reduced water gain. Conversely, a positive difference suggests the region became “wetter”, which could result from either a slowdown in water loss or an increase in water accumulation. The results reveal two distinct TWS change patterns across Canada. Southwestern Canada experienced a negative trend shift, while the rest of the country showed a positive trend shift (Fig. 6 ). It suggests that water loss in the southwest accelerated in the later years of the study period. Specifically, the results indicate ( 1 ) an acceleration of glacier melt over the Pacific Cordillera region, and ( 2 ) a reversal of positive trends over the Prairie region, where early water gains transitioned into water losses in recent years. In contrast, for the rest of the country, the results indicate that ( 1 ) glacier melt in the Arctic Cordillera, snow cover reduction in the Arctic, and surface/subsurface water discharge in the permafrost transition zone have slowed, and ( 2 ) water gain rates in eastern Canada were enhanced in the later years compared with the early years of the study period due to heightened precipitation. In summary, the TWS for Canada’s entire landmass continuously trended down over 2002–2024, with an average loss rate of 11.1 mm year − 1 . This water loss rate appeared to have accelerated in recent years, coinciding with the recent drought development in south-central Canada. Compared to the global TWS decline rate reported by Li and Rodell (2024), Canada’s landmass water loss rate over the 22-year period was approximately 10% higher. Moreover, while Li and Rodell (2024) noted a global slowdown in TWS loss since 2020, Canada’s landmass TWS loss has instead accelerated. Geographically, there are three major regions that experienced significant large-scale water loss. They include the western Pacific Cordillera, the Arctic Cordillera, and the permafrost transition area in central Canada. Drivers of water loss in these regions involve consistent ice sheets melt and permafrost degradation. In contrast, eastern Canada experienced significant water gains, primarily driven by increased trend in precipitation. The Canadian Prairies experienced a trend reversal, shifting from an increasing sub-trend in the early years to a decreasing sub-trend in the later years, resulting in an overall near-zero trend for 2002–2024. Our results show that, for the ice-free regions in Canada, the dry areas (i.e., south-central Canada) have become drier, and the wet areas (i.e., eastern Canada) have become wetter, aligning with the paradigm that has been used to summarize the expected trends of the global hydrological cycle under climate change (Xiong et al., 2022). For ice-covered regions, water loss rates have accelerated over the Pacific Cordillera but have slowed over the Arctic Cordillera and permafrost zones in northern Canada. The 22-year record of TWS data from the GRACE and GRACE-FO satellite missions provides an unprecedented opportunity to understand the cold region dynamics of TWS for Canada. As the GRACE-FO observations continue and the GRACE-C continuity mission is expected to be launched in 2028, the TWS data records will be further extended. This will enable more robust characterization of the TWS climatology, particularly in assessing long-term trends, decadal variations, and extreme events. With the impact of accelerated water fluxes in a warmer climate, future TWS changes are likely to become even more dynamic. Improved TWS information will be increasingly critical for enhancing water resources management and informed decision-making. Methods GRACE and GRACE-FO TWS datasets The GRACE and Follow-On (GRACE-FO) TWS datasets used in this study include a total of six products from four research institutions: the Center for Space Research at the University of Texas (CSR), GeoForschungsZentrum (GFZ), Jet Propulsion Laboratory (JPL), and NASA Goddard Space Flight Center (GSFC). Three products are derived from Spherical Harmonics (SH) solutions, while the other three are based on Mass Concentration Blocks (Mascon) solutions. These datasets were obtained from the respective data provider’s websites, as listed in Table 1 , and represent the latest available versions at the time of downloading (June 2024). These datasets consist of Level 3 gridded land fields of time-series GRACE and GRACE-FO TWS, provided at a monthly temporal resolution, covering a 22-year period from April-2002 to March-2024. Although previous studies have applied gap-filling techniques and downscaled TWS data to higher spatial resolutions for Canada (He et al., 2021; Yu et al., 2021; Zhong et al., 2021; Fatolazadeh et al., 2022), our analysis in this study relies solely on the original GRACE and GRACE-FO TWS data. One consideration is that all these gap-filling and downscaling studies used data from additional sources (e.g., land surface models) in addition to GRACE and GRACE-FO observations, which may introduce potential uncertainties in the data products. Another consideration is that this study primarily focuses on large scale and trend analyses, where the impact of data gaps and coarse spatial resolution on our results is minimized. All the downloaded GRACE TWS datasets were subset for Canada’s landmass, resampled at a 5-km grid cell, and re-projected into the Lambert Conformal Conic (LCC) projection. Scale factors were applied to the corresponding datasets ( https://grace.jpl.nasa.gov/data/get-data/monthly-mass-grids-land/ ). For grid cells where the scale factor had a null value (mainly in the Arctic Coast-Islands for the SH datasets), the scale factor was set to 1.0. All reported results are anomalies calculated relative to the baseline (average) of the entire TWS time series dataset (April 2002 – March 2024). Note that the baseline for the original datasets is the average of TWS over the period of January 2004 – December 2009. Table 1 Data information for the GRACE and GRACE-FO TWS products used in this study. Solutions Spherical Harmonics (SH) Mass Concentration Blocks (Mascon) Producer CSR JPL GFZ CSR JPL GSFC Version RL06v4.0 RL06v4.0 RL06v4.0 RL06v2.0 RL06v3.0 RL06v2.0 Grid size 1.0°×1.0° 1.0°×1.0° 1.0°×1.0° 0.25°×0.25° 0.5°×0.5° 0.5°×0.5° GRACE and GRACE-FO Datasets download Website (accessed in 2024/06) CSR-SH GRACE: https://podaac.jpl.nasa.gov/dataset/TELLUS_GRAC_L3_CSR_RL06_LND_v04 GRACE-FO: https://podaac.jpl.nasa.gov/dataset/TELLUS_GRFO_L3_CSR_RL06.1_LND_v04 JPL-SH GRACE: https://podaac.jpl.nasa.gov/dataset/TELLUS_GRAC_L3_JPL_RL06_LND_v04 GRACE-FO: https://podaac.jpl.nasa.gov/dataset/TELLUS_GRFO_L3_JPL_RL06.1_LND_v04 GFZ-SH GRACE: https://podaac.jpl.nasa.gov/dataset/TELLUS_GRAC_L3_GFZ_RL06_LND_v04 GRACE-FO: https://podaac.jpl.nasa.gov/dataset/TELLUS_GRFO_L3_GFZ_RL06.1_LND_v04 CSR-Mascon GRACE and GRACE-FO: https://www2.csr.utexas.edu/grace/RL06_mascons.html JPL-Mascon GRACE and GRACE-FO: https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/TELLUS_GRAC-GRFO_MASCON_CRI_GRID_RL06.1_V3/ GSFC-Mascon GRACE and GRACE-FO: https://earth.gsfc.nasa.gov/geo/data/grace-mascons Evaluations of TWS datasets The differences among the six GRACE TWS datasets used in this study were assessed using the Mean Absolute Difference (MAD), calculated as: $$\:MAD(i,\:j)=\frac{1}{M}\sum\:_{t=1}^{t=M}\left(\frac{1}{N}\sum\:_{n=1}^{n=N}|{TWS}_{n}\left(t,\:i,j\right)-{TWS}_{mean}\left(t,\:i,j\right)|\right)$$ 1 where TWS mean is the mean for a TWS dataset group (i.e., SH or Mascon), TWS n is the TWS for an individual dataset n, t is the time (month), ( i, j ) are grid cell coordinates, N is the total number of TWS products in a group (3 for both SH and Mascon), and M is the length of the TWS data time series. The TWS differences were assessed for three groups of datasets: ( 1 ) the three TWS datasets by the Spherical Harmonics solution (SH), ( 2 ) the three TWS datasets by the Mass Concentration Blocks solution (Mascon), and ( 3 ) the product means between the two TWS solutions. Results for the intercomparisons are provided in the Supplementary Information. In summary, the MAD for the three TWS products by SH solution is very small over Canada’s landmass, varying mostly under 15 mm. The difference among the three TWS products by Mascon solution is much larger than that by SH solution. Regions with MAD greater than 30 mm account for almost 33% of the landmass. The highest MAE exceeded 200 mm. High MAD values are found mainly over regions where ice cover is distributed, namely the Pacific Cordillera and the Arctic Cordillera. The differences of TWS between the SH products and the Mascon products are found to be larger than the differences among the products within each of the two solutions. Regions with MAD exceeding 30 mm accounts for more than 43% of the landmass. High MAD values are distributed over regions where the high MAD values for the Mascon products are found. In addition to the grid-level assessment, the results were also compared at the scale of the Canadian Drainage Regions and are reported in the Supplementary Information. Canada has a total of 25 Drainage Regions that drain water to the Pacific Ocean, Arctic Ocean, Atlantic Ocean, Hudson Bay, and Gulf of Mexico. The impacts of the differences among the TWS products are found to be small at the Drainage Region scale. The correlation coefficient of the TWS trends between SH and Mascon solutions is 0.94 for the 25 Drainage Regions. At the national scale, the TWS products from SH and Mascon solutions agreed well (Fig. 1 ). The correlation coefficient between the two TWS time series is greater than 0.99, with negligible bias. The TWS trends generated by the SH and Mascon solutions also have negligible differences at the national scale. Our evaluation is based on GRACE observations only, and no benchmark data is available for direct comparison. As such, a strict validation of the TWS products is not yet possible. However, it is worth noting that the TWS trends by the Mascon solution were found to have outperformed the SH solution over regions with ice covers when compared to the trends of high-resolution precipitation and water yield data over the study period. This suggests the advantages of Mascon solution in reducing leakage errors and other uncertainties in TWS from post-processing over Spherical Harmonics solution, making it increasingly preferrable for applications, particularly in regional studies. With new observation methods becoming available for improving the quantifications of various TWS components (for example, the Surface Water and Ocean Topography (SWOT) mission for quantifying surface water storage changes), it is expected that water budget closure analyses will advance. This progress could potentially help evaluate the data quality across different GRACE TWS products. Declarations Acknowledgements This study was supported by the Earth Observation for Cumulative Effects (EO4CE) Project of the Natural Resources Canada (NRCan). The EO4CE project is part of NRCan’s Status and Trends Program and is funded as a component of the Government of Canada’s initiative for the assessment and monitoring of cumulative effects. Data availability Data used in this study are openly available and can be obtained in the websites listed in Table 1. Competing interests The authors declare no competing interests. Additional information This article contains Supplementary Information as attached. References Bouchard, F., K. W. Turner, L. A. MacDonald, et al. 2013. Vulnerability of shallow subarctic lakes to evaporate and desiccate when snowmelt runof is low. Geophys. Res. Lett. 40, 6112–6117, https://doi.org/10.1002/2013GL058635. Burgess, D., G. Wolken, B. Wouters, L. M. Andreassen, C. Florentine, J. Kohler, B. Luks, F. Pálsson, L. Sass, L. Thomson, and T. Thorsteinsson, 2024: Glaciers and ice caps outside Greenland [in “State of the Climate in 2023”]. Bull. Amer. Meteor. Soc. , 105 (8), S307–S310, https://doi.org/10.1175/BAMS-D-24-0101.1. Ciracì, E., Velicogna, I., and Swenson, S. 2020. Continuity of the mass loss of the world's glaciers and ice caps from the GRACE and GRACE Follow-On missions. Geophysical Research Letters , 47 , e2019GL086926. https://doi.org/10.1029/2019GL086926. Duvvuri, B., and E. Beighley, 2023. Estimating Monthly River Discharges from GRACE/GRACE-FO Terrestrial Water Storage Anomalies. Remote Sens. 15, 4516. https://doi.org/10.3390/rs15184516. Famiglietti, J. S. 2014. The global groundwater crisis. Nature Clim. Change, 4(11), 945–948. https://doi.org/10.1038/nclimate2425 Fatolazadeh, F.; Eshagh, M.; Goïta, K. and Wang, S. 2022. A New Spatiotemporal Estimator to Downscale GRACE Gravity Models for Terrestrial and Groundwater Storage Variations Estimation. Remote Sens. 14, 5991. https://doi.org/10.3390/rs14235991. Göckede, M., Kwon, M.J., Kittler, F., Heimann, M., Zimov, N., Zimov, S., 2019. Negative feedback processes following drainage slow down permafrost degradation. Global Change Biology 25, 3254–3266. https://doi.org/10.1111/gcb.14744 Hamlington, B.D., Bellas-Manley, A., Willis, J.K. et al., 2024, The rate of global sea level rise doubled during the past three decades. Commun Earth Environ 5, 601 (2024). https://doi.org/10.1038/s43247-024-01761-5 Hanesiak, J.M., R.E. Stewart, B.R. Bonsal, et al., 2011, Characterization and Summary of the 1999–2005 Canadian Prairie Drought. Atmosphere-Ocean, 2011, 1–32 http://dx.doi.org/10.1080/07055900.2011.626757. He, H., K. Yang, S. Wang, H.A. Petrosian, M. Liu, J. Li, J. M. Junior, W. Gonçalves, L. Wang, and J. Li, 2021. A deep learning approach to spatial downscaling of GRACE terrestrial water storage using EALCO model over Canada. Canadian Journal of Remote Sensing, 47, 657–675. DOI: https://doi.org/10.1080/07038992.2021.1954498. Kurylyk, B. L., M. Hayashi,W. L. Quinton, J. M. McKenzie, andC. I. Voss (2016), Influence of vertical and lateral heat transfer on permafrost thaw, peatland landscape transition, and groundwater flow, Water Resour.Res., 52, 1286–1305, doi:10.1002/2015WR018057. Landerer, F.W., Flechtner, F., et al., 2020, Extending the global mass change data record: GRACE Follow-On instrument and science data performance, Geophys. Res. Lett., https://doi.org/10.1029/2020GL088306. Li, B., Rodell, M. 2024, Terrestrial water storage in 2023. Nat Rev Earth Environ 5, 247–249. https://doi.org/10.1038/s43017-024-00545-x. Li, J. and S. Wang, 2022, Seasonal variations and long-term trends of groundwater in Canada. Hydrogeology J. , 30:401–415, https://doi.org/10.1007/s10040-022-02460-1. Li, Z. & S. Wang, 2021, Water yield variability and response to climate change across Canada, Hydrological Sciences Journal, 66:7, 1169-1184, DOI: https://doi.org/10.1080/02626667.2021.1925122. Li, Z., Wang, S., Li, J., 2020, Spatial variations and long-term trends of potential evaporation in Canada. Scientific Reports, 10, 22089, www.nature.com/articles/s41598-020-78994-9. Pelletier, M., Allard, M., Levesque, E., 2019. Ecosystem changes across a gradient of permafrost degradation in subarctic Québec (Tasiapik Valley, Nunavik, Canada). Arctic Science 5, 1–26. https://doi.org/10.1139/as-2016-0049 Peltier, W. R., Argus, D. F. & Drummond, R. Comment on “An Assessment of the ICE-6G_C (VM5a) glacial isostatic adjustment model by Purcell et al. J. Geophys. Res. Solid Earth 122, 2019–2028 (2017). https://doi.org/10.1002/2016JB013844 Peltier, W. R., Argus, D. F. & Drummond, R. Space geodesy constrains ice age terminal deglaciation: the global ICE-6G_C (VM5a) model. J. Geophys. Res. Solid Earth 120, 450–487 (2015). https://doi.org/10.1002/2014JB011176 Rodell, M., Famiglietti, J.S., Wiese, D.N. et al. Emerging trends in global freshwater availability. Nature 557, 651–659 (2018). https://doi.org/10.1038/s41586-018-0123-1. Sergeant, F., Therrien, R., Oudin, L., Jost, A., Anctil, F., 2021. Evolution of Arctic rivers recession flow: Global assessment and data-based attribution analysis. Journal of Hydrology 601, 126577. https://doi.org/10.1016/j.jhydrol.2021.126577 The GlaMBIE Team. Community estimate of global glacier mass changes from 2000 to 2023. Nature (2025), https://doi.org/10.1038/s41586-024-08545-z. Thibault, S., Payette, S., 2009. Recent permafrost degradation in bogs of the James Bay area, northern Quebec, Canada. Permafrost and Periglacial Processes 20, 383–389. https://doi.org/10.1002/ppp.660 Wang, S. and H. Russell, 2016, Forecasting snowmelt-induced flooding using GRACE satellite data: A case study for the Red River watershed. Canadian Journal of Remote Sensing, 42, 203-213, doi: https://doi.org/10.1080/07038992.2016.1171134. Wang, S. and J. Li, 2016, Terrestrial water storage climatology for Canada from GRACE satellite observations in 2002-2014. Canadian Journal of Remote Sensing, 42, 190-202, doi: https://doi.org/10.1080/07038992.2016.1171132. Wang, S., 2019, Freezing temperature controls winter water discharge for cold region watershed. Water Resources Research, 55, 10,479-10,493. doi: https://doi.org/10.1029/2019WR026030. Wang, S., J. Huang, D. Yang, G. Pavlic, J. Li, 2015, Long-term water budget imbalances and error sources for cold region drainage basins. Hydrological Processes, 29, 2125–2136, doi: https://doi.org/10.1002/hyp.10343. Wang, S., J. Huang, J. Li, A. Rivera, D.W. McKenney, and J. Sheffield, 2014, Assessment of water budget for sixteen large drainage basins in Canada. Journal of Hydrology, 512: 1-15, doi: https://doi.org/10.1016/j.jhydrol.2014.02.058. Wang, S., Li, J., and Russell, H. A. J., 2021: A novel method for cold-region streamflow hydrograph separation using GRACE satellite observations. Hydrol. Earth Syst. Sci., 25, 2649–2662, https://doi.org/10.5194/hess-25-2649-2021. Wang, S., Zhou, F., Russell, H. A. J., 2017, Estimating snow mass and peak river flows for the Mackenzie River basin using GRACE satellite observations. Remote Sensing, 9, 256, doi: https://doi.org/10.3390/rs9030256. Xiong, Jinghua & Guo, Shenglian & , Abhishek & Chen, Jie & Yin, Jiabo. (2022). Global evaluation of the “dry gets drier, and wet gets wetter” paradigm from a terrestrial water storage change perspective. Hydrology and Earth System Sciences. 26. 6457-6476. 10.5194/hess-26-6457-2022. Yu, Q., S. Wang, H. He, K. Yang, L. Ma, J. Li, 2021, Reconstructing GRACE-like TWS Anomalies for the Canadian Landmass Using Deep Learning and Land Surface Model. International Journal of Applied Earth Observation and Geoinformation, 102, 102404, DOI: https://doi.org/10.1016/j.jag.2021.102404. Zhong, D., S. Wang, and J. Li, 2021, Spatiotemporal Downscaling of GRACE Total Water Storage Using Land Surface Model Outputs. Remote Sensing, 13, 900, https://doi.org/10.3390/rs13050900. Additional Declarations No competing interests reported. Supplementary Files SuppInfo.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6253460","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":445149743,"identity":"dc9581f1-0a11-48a4-9a52-6ce09ef0cd96","order_by":0,"name":"Shusen Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABMElEQVRIiWNgGAWjYBAC+RkQOrGBgYHxAVFaDG4gtDAbEKdFAqGFTYI4LdI9xh8+7mBI7J/dfq26cI+dvDkD77MHP2ruMMi7H2D88ANTi/ycM2aSM88w5M64c6bs9oxnyYY7G9jNDXuOPWMwPJPALNmDxZobOWbMvG0MuQ03ctJu8xw4wLjh/jM2aQa2wwyGMxjYGHiwajH+/BeoZT5QSzFQi/2GA2xALf8gWhj/YNViIM3YxlC/4Ub6MWaglkSwFsa2wwzyEgxszFhsMbiRVibZ2yZRvPFGDrM0z4HkZJAWyd6+wzwGPInN0jJYvD8jefOHn202ifNupD/8zHPAzhakReLHt8Ny8u2HD358gzOsQTHCgxqTPAYHGBtwaoAA9gdoDiCkYRSMglEwCkYKAABX22pjaApCHwAAAABJRU5ErkJggg==","orcid":"","institution":"Canada Centre for Remote Sensing","correspondingAuthor":true,"prefix":"","firstName":"Shusen","middleName":"","lastName":"Wang","suffix":""},{"id":445149744,"identity":"a1871ba0-fff7-4d48-8099-772f24600e65","order_by":1,"name":"Fuqun Zhou","email":"","orcid":"","institution":"Canada Centre for Remote Sensing","correspondingAuthor":false,"prefix":"","firstName":"Fuqun","middleName":"","lastName":"Zhou","suffix":""},{"id":445149745,"identity":"614946ed-a491-462d-9924-674c3e1a2076","order_by":2,"name":"Farzam Fatolazadeh","email":"","orcid":"","institution":"Université de Sherbrooke","correspondingAuthor":false,"prefix":"","firstName":"Farzam","middleName":"","lastName":"Fatolazadeh","suffix":""},{"id":445149746,"identity":"6a8e9f38-346e-4423-a013-6c4ca1ccc174","order_by":3,"name":"Hazen Russell","email":"","orcid":"","institution":"Geological Survey of Canada","correspondingAuthor":false,"prefix":"","firstName":"Hazen","middleName":"","lastName":"Russell","suffix":""},{"id":445149747,"identity":"875b1954-2933-414d-9009-e1ca740c42a9","order_by":4,"name":"Melissa Bunn","email":"","orcid":"","institution":"Geological Survey of Canada","correspondingAuthor":false,"prefix":"","firstName":"Melissa","middleName":"","lastName":"Bunn","suffix":""}],"badges":[],"createdAt":"2025-03-18 13:08:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6253460/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6253460/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":81031308,"identity":"bae1770d-b8ba-4e27-a1e5-d22952a4213b","added_by":"auto","created_at":"2025-04-21 11:22:27","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":66460,"visible":true,"origin":"","legend":"\u003cp\u003eTerrestrial water storage (TWS) monthly time-series for Canada’s landmass (SH: based on TWS products by Spherical Harmonics solution. Mascon: based on TWS products by Mascon solution).\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6253460/v1/e7a0b7e30bcf6ddc9e6ec5bc.png"},{"id":81032285,"identity":"0c39ac37-2b32-4011-abe9-a2cc9ebcf40a","added_by":"auto","created_at":"2025-04-21 11:30:27","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":358029,"visible":true,"origin":"","legend":"\u003cp\u003eTerrestrial water storage (TWS) trends over April 2002–March 2024 (mm year\u003csup\u003e-1\u003c/sup\u003e).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6253460/v1/5bb8962f4bd07407400f3e28.png"},{"id":81032287,"identity":"f7935c18-dcd9-4582-bf00-47ff49913afb","added_by":"auto","created_at":"2025-04-21 11:30:27","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":67024,"visible":true,"origin":"","legend":"\u003cp\u003ePacific Coastal Drainage Region terrestrial water storage (TWS) monthly time-series (See Fig. S5 in the Supplementary Information for location. SH: based on TWS products by Spherical Harmonics solution. Mascon: based on TWS products by Mascon solution.)\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6253460/v1/3c465c4ff744ab55f015ffb3.png"},{"id":81031309,"identity":"10a7287f-929e-444c-96b6-c1d937193520","added_by":"auto","created_at":"2025-04-21 11:22:27","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":73797,"visible":true,"origin":"","legend":"\u003cp\u003eSt. Lawrence Drainage Region terrestrial water storage (TWS) monthly time-series (See Fig. S5 in Supplementary Information for location. SH: based on TWS products by Spherical Harmonics solution. Mascon: based on TWS products by Mascon solution).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6253460/v1/4804ae1fa4c981082bafff23.png"},{"id":81031310,"identity":"bfc826f1-f974-435d-9829-5aba204dc22c","added_by":"auto","created_at":"2025-04-21 11:22:27","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":72862,"visible":true,"origin":"","legend":"\u003cp\u003eSouth Saskatchewan Drainage Region terrestrial water storage (TWS) monthly time-series (See Fig. S5 in Supplementary Information for location. SH: based on TWS products by Spherical Harmonics solution. Mascon: based on TWS products by Mascon solution).\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6253460/v1/51b54665b5ffd28ad01774b3.png"},{"id":81032863,"identity":"0a2ef06d-c741-46aa-8437-4e73ea602daf","added_by":"auto","created_at":"2025-04-21 11:38:27","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":360111,"visible":true,"origin":"","legend":"\u003cp\u003eChanges of terrestrial water storage trends (Difference of Sen’s slope between 2002–2024 and 2002–2014, in mm year\u003csup\u003e-1\u003c/sup\u003e).\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-6253460/v1/9a29caa798593ff46ff0b6d3.png"},{"id":83250390,"identity":"7f1dfd04-9996-4c85-83b4-0a39557833c4","added_by":"auto","created_at":"2025-05-21 19:31:37","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1456592,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6253460/v1/4027c06a-c3a3-4266-96dd-20bd00d62ee3.pdf"},{"id":81031315,"identity":"e48fe791-7650-4845-97a5-1fab144b59e2","added_by":"auto","created_at":"2025-04-21 11:22:27","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1070975,"visible":true,"origin":"","legend":"","description":"","filename":"SuppInfo.docx","url":"https://assets-eu.researchsquare.com/files/rs-6253460/v1/503996a9af2c5c75fc836cc1.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Accelerated water loss over Canada’s landmass in 2002–2024","fulltext":[{"header":"Introduction","content":"\u003cp\u003eTerrestrial Water Storage (TWS) is the sum of the water stored in terrestrial ecosystems, including surface water in rivers and lakes, soil water held in soil zones, groundwater stored in aquifers, and snow and ice covering land surface. Given its importance to the global water cycle and sensitivity to climate change impacts, TWS was added as one of 54 Essential Climate Variables (ECVs) by the Global Climate Observing System (GCOS) in 2022. Information on the status and trends of TWS plays a crucial role not only in understanding the global climate and water cycle but also in\u0026nbsp;better-informed\u0026nbsp;water resource management and policy development at various scales.\u003c/p\u003e\n\u003cp\u003eThe Gravity Recovery and Climate Experiment (GRACE) satellites and their successor, GRACE Follow-on (GRACE-FO), have been providing direct measurements of TWS changes since 2002 (Landerer et al., 2020). These measurements have been used to characterize the dynamics of TWS at various scales, including changes in its components such as groundwater storage, glacier loss and snow melt, and basin discharge (e.g., Famiglietti 2014; Rodell et al., 2018; Ciracì et al., 2020; Duvvuri et al., 2023; Li and Rodell, 2024).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCanada has a vast landmass of nearly ten million square kilometers with a diverse range of cold region hydrological processes, including glaciers distributed over the Pacific Cordillera and Arctic Cordillera, permafrost in the north, a humid climate in the east, and a semi-arid climate in the Prairies. Frozen soil covers almost the entire country during winter. Snow accumulation and snowmelt-driven surface runoff are important components of the water cycle. The region is experiencing the fastest rates of climate change, which has highly impacted its water conditions. GRACE observations have significantly contributed to our understanding of cold-region hydrological processes in Canada, including estimating snow mass and snowmelt-driven river flows (Wang and Russell 2016; Wang et al., 2017), modelling groundwater discharge and freezing temperature control on aquifer conductivity (Wang 2019; Wang et al., 2021), quantifying groundwater storage change (Li and Wang, 2022), and assessing water budget closures (Wang et al., 2014; Wang et al., 2015). At a national scale, Wang and Li (2016) characterized Canada’s TWS climatology by retrieving a suite of parameters. However, this study was based on a relatively short GRACE record of 13 years (2002–2014),\u0026nbsp;leaving open questions about which trends in TWS are persistent and indicative of longer-term changes, and which are driven by short-term climate variability.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHere we present the TWS changes for Canada’s landmass using the latest GRACE and GRACE-FO TWS products over a 22-year period from April 2002 to March 2024, and\u0026nbsp;revisit issues relating to long-term TWS changes.\u0026nbsp;A total of six GRACE/GRACE-FO TWS products from both Spherical Harmonics and Mascon solutions were used in this study as detailed in the Methods section. Unless noted, the results represent the assemble means of the six products. We also evaluated the differences among these six TWS products over Canada’s landmass, with the results provided in the Supplementary Information.\u0026nbsp;\u003c/p\u003e"},{"header":"Results and Discussion","content":"\u003cdiv id=\"Sec2\" class=\"Section2\"\u003e\n \u003ch2\u003eTerrestrial water storage (TWS) trends in 2002\u0026ndash;2024\u003c/h2\u003e\n \u003cp\u003eThe TWS for Canada\u0026rsquo;s entire landmass trended down almost continuously over 2002\u0026ndash;2024 at an average rate of 11.1 mm year\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). Over the 22-year study period, Canada\u0026rsquo;s landmass lost 2430 km\u003csup\u003e3\u003c/sup\u003e of freshwater, equivalent to 1.5 times the total water volume of Lake Ontario and accounting for 6.9 mm or 8.5% of the global sea level rise observed over the 22 years (Hamlington et al., 2024). As shown below, the loss of water was driven mainly by melting of glaciers and ice sheets, as well as the loss of ground ice and subsurface water due to permafrost degradation. Notably, the decreasing rate of TWS appeared to accelerate after 2020, averaging more than 26 mm year\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e for the last four years. This accelerated water loss in Canada is associated with the recent development of severe drought conditions in south-central Canada.\u003c/p\u003e\n \u003cp\u003eThe TWS trends over Canada\u0026rsquo;s landmass showed large variations geographically. Figure \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e shows the Sen\u0026rsquo;s slope values calculated for each grid. Sen\u0026rsquo;s slope is a nonparametric estimate of the slope for a time-series dataset. A negative Sen\u0026rsquo;s slope value indicates an overall downward trend, signifying water loss over the entire 22-year study period, and vice versa.\u003c/p\u003e\n \u003cp\u003eThe most pronounced feature in the results is the large negative trends observed over the western Pacific Cordillera and the Arctic Cordillera, with Sen\u0026rsquo;s slope values in some areas exceeding \u0026minus;\u0026thinsp;200 mm year\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e. Relatively, the decreasing trend in the western Pacific Cordillera, particularly in its northern part, was more pronounced than that in the Arctic Cordillera. Figure \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e shows the monthly TWS time series data for the Pacific Coastal Drainage Region, which covers an area of over 320\u0026times;10\u003csup\u003e3\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e and largely represents the western Pacific Cordillera (see Fig. S5 in the Supplementary Information for location). The average downward trend for this drainage region reached about \u0026minus;\u0026thinsp;60 mm year\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e. The data also revealed that the TWS loss accelerated in the latter part of the study period, starting in 2012, when the rate of loss exceeded \u0026minus;\u0026thinsp;85 mm year\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e. Analyses of precipitation and water yield data (Li and Wang, 2021) over the Pacific Cordillera and the Arctic Cordillera regions indicate that these factors could not explain the downward TWS trends. Both regions have widespread distributions of glaciers, ice sheets, and permanent snow cover, all of which have experienced significant climate warming. Our results suggest that the TWS decreasing trends were mainly driven by the significant water loss from the melting of ice and snow cover, followed by decreases in sub-surface water storage once the ice or snow cover disappeared, as evidenced by numerous in situ glacier and ice mass balance studies (e.g., The GlaMBIE Team, 2025; Burgess et al., 2024).\u003c/p\u003e\n \u003cp\u003eAnother region that experienced significant large-scale decreasing trends in TWS is the central part of the landmass around latitude 60\u0026deg;N, including the far north of Quebec. The water loss rates in some parts of this region reached over \u0026minus;\u0026thinsp;10 mm year\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e. The TWS downward trends are not explained by corresponding precipitation and water yield data (Li and Wang, 2021), which showed slightly positive trends during our study period. This region lies in the transition zone from permafrost to seasonal frozen soil, where significant soil thaw and permafrost degradation have occurred over the past decades due to climate change (Thibault and Payette, 2009; Pelletier et al., 2019). Several studies have revealed that permafrost thaw has resulted in changes of water cycle and energy balance over this region (Kurylyk et al., 2016; G\u0026ouml;ckede; et al., 2019). Permafrost thaw has activated soil drainage and aquifer discharge processes (Sergeant et al., 2021; Li and Wang, 2022). Aquifer conductivity and groundwater discharge can significantly increase with climate warming (Wang, 2019). Moreover, increased soil water drainage leads to drier soil surface, resulting in more surface energy being partitioned to sensible heat flux rather than latent heat flux. This change in surface energy balance leads to the increase in land surface temperature, which positively feeds back and further accelerates permafrost thaw processes. All these processes would contribute to the water loss over the region, as reported by ground-based studies (Bouchard et al., 2013).\u003c/p\u003e\n \u003cp\u003eRegions around Hudson Bay are experiencing significant post glacial rebound. Some areas of the permafrost transition zone mentioned earlier are located over regions with the most dramatic isostatic rebound. The Glacial Isostatic Adjustment (GIA) in this region plays an important role in retrieving TWS trends from GRACE observations (Peltier et al., 2015, 2017). Due to the uncertainties in GIA models, the trends observed in the region likely have larger uncertainty compared to other regions.\u003c/p\u003e\n \u003cp\u003eRegions exhibiting positive TWS trends at a large scale were mainly distributed in Eastern Canada, stretching from southern Ontario northeastward to Newfoundland. The magnitudes of these trends over some regions exceeded 10 mm year\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e. The positive trends over this region were found to be associated with increasing precipitation and water yield, indicating that Eastern Canada experienced a net gain in TWS due to higher precipitation during the study period. Figure \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e shows the time series of monthly TWS variations for the St. Lawrence Drainage Region (see Fig. S5 in the Supplementary Information for location). It shows that the TWS rate of increase over the region was fairly consistent over the 22 years of 2002\u0026ndash;2024.\u003c/p\u003e\n \u003cp\u003eThe TWS trends over other regions of the landmass were relatively small. This includes the south-central part of the country, notably the Canadian Prairies. The detailed interannual variations of TWS over this region will be discussed in the following section.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n \u003ch2\u003eShift of trends\u003c/h2\u003e\n \u003cp\u003eThe Prairie region over south-central Canada had overall small trends from 2002\u0026ndash;2024, as shown in Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e. However, this region was reported to have significant positive trends in an earlier study using the GRACE data for 2002\u0026ndash;2014 (Wang and Li, 2016). This suggests that there was a change or a shift in trends over different periods (sub-trends) within the 22-year timeframe. Figure \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e shows the monthly TWS time series for the South Saskatchewan Drainage Region, located in the centre of the Prairies (see Fig. S5 in the Supplementary Information for location). It is worth noting that the Prairies have a semi-arid climate, with annual total precipitation typically below 400 mm, while potential evaporation can exceed 1000 mm year\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (Li and Wang, 2020). The region experienced one of its most severe droughts in 2000\u0026thinsp;\u0026minus;\u0026thinsp;2003 since instrumental record was available. It resulted in large deficits in groundwater, surface water, and soil water (Hanesiak et al. 2011). Consequently, TWS was at a very low level at the beginning of the GRACE mission. During the subsequent 10 years following this drought event, the region saw substantial recovery as precipitation returned to normal levels, leading to a positive TWS trend. However, in the past 12 years, a new severe drought has emerged, causing a consistent decline in TWS. This prolonged dry period has completely reversed the water gains from the previous decade. By the end of our study period in March 2024, TWS in the region had dropped more than 50 mm below its April 2002 level. The small overall trend obtained for the full 22-year study period reflects the opposing effects of the initial positive sub-trend and the later negative sub-trend, which largely canceled each other out.\u003c/p\u003e\n \u003cp\u003eAnother notable feature of TWS over the Prairies is its large interannual variability. The region went through another short drought period in the seasons around 2009 which led to a substantial drop in TWS even during the first 10 years when it had an overall positive sub-trend. Similarly, the region received above-normal precipitation around 2017, which led to a significant increase in TWS, despite the region being in a period of overall decline. These findings highlight the importance of accounting for short-term sub-trends and extreme climate events when assessing long-term TWS trends in arid and semi-arid regions.\u003c/p\u003e\n \u003cp\u003eTo evaluate the possible shifts in TWS trends for the entire landmass in the study period, we calculated Sen\u0026rsquo;s slope for each grid over the period of 2002\u0026ndash;2014. We then compared it with that from 2002\u0026ndash;2024 to identify changes over time. Figure \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e shows the difference in Sen\u0026rsquo;s slope between these periods. A negative difference indicates that the region became \u0026ldquo;drier\u0026rdquo; in the later years, either due to accelerated water loss or reduced water gain. Conversely, a positive difference suggests the region became \u0026ldquo;wetter\u0026rdquo;, which could result from either a slowdown in water loss or an increase in water accumulation.\u003c/p\u003e\n \u003cp\u003eThe results reveal two distinct TWS change patterns across Canada. Southwestern Canada experienced a negative trend shift, while the rest of the country showed a positive trend shift (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e). It suggests that water loss in the southwest accelerated in the later years of the study period. Specifically, the results indicate (\u003cspan class=\"CitationRef\"\u003e1\u003c/span\u003e) an acceleration of glacier melt over the Pacific Cordillera region, and (\u003cspan class=\"CitationRef\"\u003e2\u003c/span\u003e) a reversal of positive trends over the Prairie region, where early water gains transitioned into water losses in recent years. In contrast, for the rest of the country, the results indicate that (\u003cspan class=\"CitationRef\"\u003e1\u003c/span\u003e) glacier melt in the Arctic Cordillera, snow cover reduction in the Arctic, and surface/subsurface water discharge in the permafrost transition zone have slowed, and (\u003cspan class=\"CitationRef\"\u003e2\u003c/span\u003e) water gain rates in eastern Canada were enhanced in the later years compared with the early years of the study period due to heightened precipitation.\u003c/p\u003e\n \u003cp\u003eIn summary, the TWS for Canada\u0026rsquo;s entire landmass continuously trended down over 2002\u0026ndash;2024, with an average loss rate of 11.1 mm year\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e. This water loss rate appeared to have accelerated in recent years, coinciding with the recent drought development in south-central Canada. Compared to the global TWS decline rate reported by Li and Rodell (2024), Canada\u0026rsquo;s landmass water loss rate over the 22-year period was approximately 10% higher. Moreover, while Li and Rodell (2024) noted a global slowdown in TWS loss since 2020, Canada\u0026rsquo;s landmass TWS loss has instead accelerated. Geographically, there are three major regions that experienced significant large-scale water loss. They include the western Pacific Cordillera, the Arctic Cordillera, and the permafrost transition area in central Canada. Drivers of water loss in these regions involve consistent ice sheets melt and permafrost degradation. In contrast, eastern Canada experienced significant water gains, primarily driven by increased trend in precipitation. The Canadian Prairies experienced a trend reversal, shifting from an increasing sub-trend in the early years to a decreasing sub-trend in the later years, resulting in an overall near-zero trend for 2002\u0026ndash;2024. Our results show that, for the ice-free regions in Canada, the dry areas (i.e., south-central Canada) have become drier, and the wet areas (i.e., eastern Canada) have become wetter, aligning with the paradigm that has been used to summarize the expected trends of the global hydrological cycle under climate change (Xiong et al., 2022). For ice-covered regions, water loss rates have accelerated over the Pacific Cordillera but have slowed over the Arctic Cordillera and permafrost zones in northern Canada.\u003c/p\u003e\n \u003cp\u003eThe 22-year record of TWS data from the GRACE and GRACE-FO satellite missions provides an unprecedented opportunity to understand the cold region dynamics of TWS for Canada. As the GRACE-FO observations continue and the GRACE-C continuity mission is expected to be launched in 2028, the TWS data records will be further extended. This will enable more robust characterization of the TWS climatology, particularly in assessing long-term trends, decadal variations, and extreme events. With the impact of accelerated water fluxes in a warmer climate, future TWS changes are likely to become even more dynamic. Improved TWS information will be increasingly critical for enhancing water resources management and informed decision-making.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Methods","content":"\u003ch3\u003eGRACE and GRACE-FO TWS datasets\u003c/h3\u003e\u003cp\u003eThe GRACE and Follow-On (GRACE-FO) TWS datasets used in this study include a total of six products from four research institutions: the Center for Space Research at the University of Texas (CSR), GeoForschungsZentrum (GFZ), Jet Propulsion Laboratory (JPL), and NASA Goddard Space Flight Center (GSFC). Three products are derived from Spherical Harmonics (SH) solutions, while the other three are based on Mass Concentration Blocks (Mascon) solutions. These datasets were obtained from the respective data provider’s websites, as listed in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, and represent the latest available versions at the time of downloading (June 2024). These datasets consist of Level 3 gridded land fields of time-series GRACE and GRACE-FO TWS, provided at a monthly temporal resolution, covering a 22-year period from April-2002 to March-2024. Although previous studies have applied gap-filling techniques and downscaled TWS data to higher spatial resolutions for Canada (He et al., 2021; Yu et al., 2021; Zhong et al., 2021; Fatolazadeh et al., 2022), our analysis in this study relies solely on the original GRACE and GRACE-FO TWS data. One consideration is that all these gap-filling and downscaling studies used data from additional sources (e.g., land surface models) in addition to GRACE and GRACE-FO observations, which may introduce potential uncertainties in the data products. Another consideration is that this study primarily focuses on large scale and trend analyses, where the impact of data gaps and coarse spatial resolution on our results is minimized.\u003c/p\u003e\u003cp\u003eAll the downloaded GRACE TWS datasets were subset for Canada’s landmass, resampled at a 5-km grid cell, and re-projected into the Lambert Conformal Conic (LCC) projection. Scale factors were applied to the corresponding datasets (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://grace.jpl.nasa.gov/data/get-data/monthly-mass-grids-land/\u003c/span\u003e\u003cspan address=\"https://grace.jpl.nasa.gov/data/get-data/monthly-mass-grids-land/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). For grid cells where the scale factor had a null value (mainly in the Arctic Coast-Islands for the SH datasets), the scale factor was set to 1.0. All reported results are anomalies calculated relative to the baseline (average) of the entire TWS time series dataset (April 2002 – March 2024). Note that the baseline for the original datasets is the average of TWS over the period of January 2004 – December 2009.\u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eData information for the GRACE and GRACE-FO TWS products used in this study.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"8\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSolutions\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eSpherical Harmonics (SH)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003eMass Concentration Blocks (Mascon)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eProducer\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCSR\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eJPL\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGFZ\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCSR\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eJPL\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eGSFC\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVersion\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRL06v4.0\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRL06v4.0\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRL06v4.0\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRL06v2.0\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eRL06v3.0\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eRL06v2.0\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGrid size\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.0°×1.0°\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.0°×1.0°\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.0°×1.0°\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.25°×0.25°\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.5°×0.5°\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.5°×0.5°\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"8\" nameend=\"c8\" namest=\"c1\"\u003e \u003cp\u003eGRACE and GRACE-FO Datasets download Website (accessed in 2024/06)\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCSR-SH\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e \u003cp\u003eGRACE:\u003c/p\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://podaac.jpl.nasa.gov/dataset/TELLUS_GRAC_L3_CSR_RL06_LND_v04\u003c/span\u003e\u003cspan address=\"https://podaac.jpl.nasa.gov/dataset/TELLUS_GRAC_L3_CSR_RL06_LND_v04\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003eGRACE-FO:\u003c/p\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://podaac.jpl.nasa.gov/dataset/TELLUS_GRFO_L3_CSR_RL06.1_LND_v04\u003c/span\u003e\u003cspan address=\"https://podaac.jpl.nasa.gov/dataset/TELLUS_GRFO_L3_CSR_RL06.1_LND_v04\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eJPL-SH\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e \u003cp\u003eGRACE:\u003c/p\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://podaac.jpl.nasa.gov/dataset/TELLUS_GRAC_L3_JPL_RL06_LND_v04\u003c/span\u003e\u003cspan address=\"https://podaac.jpl.nasa.gov/dataset/TELLUS_GRAC_L3_JPL_RL06_LND_v04\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003eGRACE-FO:\u003c/p\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://podaac.jpl.nasa.gov/dataset/TELLUS_GRFO_L3_JPL_RL06.1_LND_v04\u003c/span\u003e\u003cspan address=\"https://podaac.jpl.nasa.gov/dataset/TELLUS_GRFO_L3_JPL_RL06.1_LND_v04\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGFZ-SH\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e \u003cp\u003eGRACE:\u003c/p\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://podaac.jpl.nasa.gov/dataset/TELLUS_GRAC_L3_GFZ_RL06_LND_v04\u003c/span\u003e\u003cspan address=\"https://podaac.jpl.nasa.gov/dataset/TELLUS_GRAC_L3_GFZ_RL06_LND_v04\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003eGRACE-FO:\u003c/p\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://podaac.jpl.nasa.gov/dataset/TELLUS_GRFO_L3_GFZ_RL06.1_LND_v04\u003c/span\u003e\u003cspan address=\"https://podaac.jpl.nasa.gov/dataset/TELLUS_GRFO_L3_GFZ_RL06.1_LND_v04\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCSR-Mascon\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e \u003cp\u003eGRACE and GRACE-FO:\u003c/p\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www2.csr.utexas.edu/grace/RL06_mascons.html\u003c/span\u003e\u003cspan address=\"https://www2.csr.utexas.edu/grace/RL06_mascons.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eJPL-Mascon\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e \u003cp\u003eGRACE and GRACE-FO:\u003c/p\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/TELLUS_GRAC-GRFO_MASCON_CRI_GRID_RL06.1_V3/\u003c/span\u003e\u003cspan address=\"https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/TELLUS_GRAC-GRFO_MASCON_CRI_GRID_RL06.1_V3/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGSFC-Mascon\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"6\" nameend=\"c7\" namest=\"c2\"\u003e \u003cp\u003eGRACE and GRACE-FO:\u003c/p\u003e \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://earth.gsfc.nasa.gov/geo/data/grace-mascons\u003c/span\u003e\u003cspan address=\"https://earth.gsfc.nasa.gov/geo/data/grace-mascons\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e\u003ch3\u003eEvaluations of TWS datasets\u003c/h3\u003e\u003cp\u003eThe differences among the six GRACE TWS datasets used in this study were assessed using the Mean Absolute Difference (MAD), calculated as:\u003c/p\u003e\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:MAD(i,\\:j)=\\frac{1}{M}\\sum\\:_{t=1}^{t=M}\\left(\\frac{1}{N}\\sum\\:_{n=1}^{n=N}|{TWS}_{n}\\left(t,\\:i,j\\right)-{TWS}_{mean}\\left(t,\\:i,j\\right)|\\right)$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003cp\u003ewhere TWS\u003csub\u003emean\u003c/sub\u003e is the mean for a TWS dataset group (i.e., SH or Mascon), TWS\u003csub\u003en\u003c/sub\u003e is the TWS for an individual dataset \u003cem\u003en, t\u003c/em\u003e is the time (month), (\u003cem\u003ei, j\u003c/em\u003e) are grid cell coordinates, \u003cem\u003eN\u003c/em\u003e is the total number of TWS products in a group (3 for both SH and Mascon), and \u003cem\u003eM\u003c/em\u003e is the length of the TWS data time series.\u003c/p\u003e\u003cp\u003eThe TWS differences were assessed for three groups of datasets: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) the three TWS datasets by the Spherical Harmonics solution (SH), (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) the three TWS datasets by the Mass Concentration Blocks solution (Mascon), and (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) the product means between the two TWS solutions. Results for the intercomparisons are provided in the Supplementary Information. In summary, the MAD for the three TWS products by SH solution is very small over Canada’s landmass, varying mostly under 15 mm. The difference among the three TWS products by Mascon solution is much larger than that by SH solution. Regions with MAD greater than 30 mm account for almost 33% of the landmass. The highest MAE exceeded 200 mm. High MAD values are found mainly over regions where ice cover is distributed, namely the Pacific Cordillera and the Arctic Cordillera. The differences of TWS between the SH products and the Mascon products are found to be larger than the differences among the products within each of the two solutions. Regions with MAD exceeding 30 mm accounts for more than 43% of the landmass. High MAD values are distributed over regions where the high MAD values for the Mascon products are found.\u003c/p\u003e\u003cp\u003eIn addition to the grid-level assessment, the results were also compared at the scale of the Canadian Drainage Regions and are reported in the Supplementary Information. Canada has a total of 25 Drainage Regions that drain water to the Pacific Ocean, Arctic Ocean, Atlantic Ocean, Hudson Bay, and Gulf of Mexico. The impacts of the differences among the TWS products are found to be small at the Drainage Region scale. The correlation coefficient of the TWS trends between SH and Mascon solutions is 0.94 for the 25 Drainage Regions. At the national scale, the TWS products from SH and Mascon solutions agreed well (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The correlation coefficient between the two TWS time series is greater than 0.99, with negligible bias. The TWS trends generated by the SH and Mascon solutions also have negligible differences at the national scale.\u003c/p\u003e\u003cp\u003eOur evaluation is based on GRACE observations only, and no benchmark data is available for direct comparison. As such, a strict validation of the TWS products is not yet possible. However, it is worth noting that the TWS trends by the Mascon solution were found to have outperformed the SH solution over regions with ice covers when compared to the trends of high-resolution precipitation and water yield data over the study period. This suggests the advantages of Mascon solution in reducing leakage errors and other uncertainties in TWS from post-processing over Spherical Harmonics solution, making it increasingly preferrable for applications, particularly in regional studies. With new observation methods becoming available for improving the quantifications of various TWS components (for example, the Surface Water and Ocean Topography (SWOT) mission for quantifying surface water storage changes), it is expected that water budget closure analyses will advance. This progress could potentially help evaluate the data quality across different GRACE TWS products.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by the Earth Observation for Cumulative Effects (EO4CE) Project of the Natural Resources Canada (NRCan). The EO4CE project is part of NRCan’s Status and Trends Program and is funded as a component of the Government of Canada’s initiative for the assessment and monitoring of cumulative effects.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData used in this study are openly available and can be obtained in the websites listed in Table 1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAdditional information\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis article contains Supplementary Information as attached.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBouchard, F., K. W. Turner, L. A. MacDonald, et al. 2013. Vulnerability of shallow subarctic lakes to evaporate and desiccate when snowmelt runof is low. Geophys. Res. Lett. 40, 6112\u0026ndash;6117, https://doi.org/10.1002/2013GL058635.\u003c/li\u003e\n\u003cli\u003eBurgess, D., G. Wolken, B. Wouters, L. M. Andreassen, C. Florentine, J. Kohler, B. Luks, F. P\u0026aacute;lsson, L. Sass, L. Thomson, and T. Thorsteinsson, 2024: Glaciers and ice caps outside Greenland [in \u0026ldquo;State of the Climate in 2023\u0026rdquo;]. \u003cem\u003eBull. Amer. Meteor. Soc.\u003c/em\u003e, \u003cstrong\u003e105 \u003c/strong\u003e(8), S307\u0026ndash;S310, https://doi.org/10.1175/BAMS-D-24-0101.1.\u003c/li\u003e\n\u003cli\u003eCirac\u0026igrave;, E., Velicogna, I., and Swenson, S. 2020. Continuity of the mass loss of the world\u0026apos;s glaciers and ice caps from the GRACE and GRACE Follow-On missions. \u003cem\u003eGeophysical Research Letters\u003c/em\u003e, \u003cem\u003e47\u003c/em\u003e, e2019GL086926. https://doi.org/10.1029/2019GL086926.\u003c/li\u003e\n\u003cli\u003eDuvvuri, B., and E. Beighley, 2023. Estimating Monthly River Discharges from GRACE/GRACE-FO Terrestrial Water Storage Anomalies. Remote Sens. 15, 4516. https://doi.org/10.3390/rs15184516.\u003c/li\u003e\n\u003cli\u003eFamiglietti, J. S. 2014. The global groundwater crisis. Nature Clim. Change, 4(11), 945\u0026ndash;948. https://doi.org/10.1038/nclimate2425\u003c/li\u003e\n\u003cli\u003eFatolazadeh, F.; Eshagh, M.; Go\u0026iuml;ta, K. and Wang, S. 2022. A New Spatiotemporal Estimator to Downscale GRACE Gravity Models for Terrestrial and Groundwater Storage Variations Estimation. Remote Sens. 14, 5991. https://doi.org/10.3390/rs14235991.\u003c/li\u003e\n\u003cli\u003eG\u0026ouml;ckede, M., Kwon, M.J., Kittler, F., Heimann, M., Zimov, N., Zimov, S., 2019. Negative feedback processes following drainage slow down permafrost degradation. Global Change Biology 25, 3254\u0026ndash;3266. https://doi.org/10.1111/gcb.14744\u003c/li\u003e\n\u003cli\u003eHamlington, B.D., Bellas-Manley, A., Willis, J.K. et al., 2024, The rate of global sea level rise doubled during the past three decades. Commun Earth Environ 5, 601 (2024). https://doi.org/10.1038/s43247-024-01761-5\u003c/li\u003e\n\u003cli\u003eHanesiak, J.M., R.E. Stewart, B.R. Bonsal, et al., 2011, Characterization and Summary of the 1999\u0026ndash;2005 Canadian Prairie Drought. Atmosphere-Ocean, 2011, 1\u0026ndash;32 http://dx.doi.org/10.1080/07055900.2011.626757.\u003c/li\u003e\n\u003cli\u003eHe, H., K. Yang, S. Wang, H.A. Petrosian, M. Liu, J. Li, J. M. Junior, W. Gon\u0026ccedil;alves, L. Wang, and J. Li, 2021. A deep learning approach to spatial downscaling of GRACE terrestrial water storage using EALCO model over Canada. Canadian Journal of Remote Sensing, 47, 657\u0026ndash;675. DOI: https://doi.org/10.1080/07038992.2021.1954498.\u003c/li\u003e\n\u003cli\u003eKurylyk, B. L., M. Hayashi,W. L. Quinton, J. M. McKenzie, andC. I. Voss (2016), Influence of vertical and lateral heat transfer on permafrost thaw, peatland landscape transition, and groundwater flow, Water Resour.Res., 52, 1286\u0026ndash;1305, doi:10.1002/2015WR018057.\u003c/li\u003e\n\u003cli\u003eLanderer, F.W., Flechtner, F., et al., 2020, Extending the global mass change data record: GRACE Follow-On instrument and science data performance, Geophys. Res. Lett., https://doi.org/10.1029/2020GL088306.\u003c/li\u003e\n\u003cli\u003eLi, B., Rodell, M. 2024, Terrestrial water storage in 2023. Nat Rev Earth Environ 5, 247\u0026ndash;249. https://doi.org/10.1038/s43017-024-00545-x.\u003c/li\u003e\n\u003cli\u003eLi, J. and S. Wang, 2022, Seasonal variations and long-term trends of groundwater in Canada. \u003cem\u003eHydrogeology J.\u003c/em\u003e, 30:401\u0026ndash;415, https://doi.org/10.1007/s10040-022-02460-1.\u003c/li\u003e\n\u003cli\u003eLi, Z. \u0026amp; S. Wang, 2021, Water yield variability and response to climate change across Canada, \u003cem\u003eHydrological Sciences Journal,\u003c/em\u003e 66:7, 1169-1184, DOI: https://doi.org/10.1080/02626667.2021.1925122.\u003c/li\u003e\n\u003cli\u003eLi, Z., Wang, S., Li, J., 2020, Spatial variations and long-term trends of potential evaporation in Canada. Scientific Reports, 10, 22089, www.nature.com/articles/s41598-020-78994-9.\u003c/li\u003e\n\u003cli\u003ePelletier, M., Allard, M., Levesque, E., 2019. Ecosystem changes across a gradient of permafrost degradation in subarctic Qu\u0026eacute;bec (Tasiapik Valley, Nunavik, Canada). Arctic Science 5, 1\u0026ndash;26. https://doi.org/10.1139/as-2016-0049\u003c/li\u003e\n\u003cli\u003ePeltier, W. R., Argus, D. F. \u0026amp; Drummond, R. Comment on \u0026ldquo;An Assessment of the ICE-6G_C (VM5a) glacial isostatic adjustment model by Purcell et al. J. Geophys. Res. Solid Earth 122, 2019\u0026ndash;2028 (2017). https://doi.org/10.1002/2016JB013844\u003c/li\u003e\n\u003cli\u003ePeltier, W. R., Argus, D. F. \u0026amp; Drummond, R. Space geodesy constrains ice age terminal deglaciation: the global ICE-6G_C (VM5a) model. J. Geophys. Res. Solid Earth 120, 450\u0026ndash;487 (2015). https://doi.org/10.1002/2014JB011176\u003c/li\u003e\n\u003cli\u003eRodell, M., Famiglietti, J.S., Wiese, D.N. et al. Emerging trends in global freshwater availability. Nature 557, 651\u0026ndash;659 (2018). https://doi.org/10.1038/s41586-018-0123-1.\u003c/li\u003e\n\u003cli\u003eSergeant, F., Therrien, R., Oudin, L., Jost, A., Anctil, F., 2021. Evolution of Arctic rivers recession flow: Global assessment and data-based attribution analysis. Journal of Hydrology 601, 126577. https://doi.org/10.1016/j.jhydrol.2021.126577\u003c/li\u003e\n\u003cli\u003eThe GlaMBIE Team. Community estimate of global glacier mass changes from 2000 to 2023. Nature (2025), https://doi.org/10.1038/s41586-024-08545-z. \u003c/li\u003e\n\u003cli\u003eThibault, S., Payette, S., 2009. Recent permafrost degradation in bogs of the James Bay area, northern Quebec, Canada. Permafrost and Periglacial Processes 20, 383\u0026ndash;389. https://doi.org/10.1002/ppp.660\u003c/li\u003e\n\u003cli\u003eWang, S. and H. Russell, 2016, Forecasting snowmelt-induced flooding using GRACE satellite data: A case study for the Red River watershed. Canadian Journal of Remote Sensing, 42, 203-213, doi: https://doi.org/10.1080/07038992.2016.1171134.\u003c/li\u003e\n\u003cli\u003eWang, S. and J. Li, 2016, Terrestrial water storage climatology for Canada from GRACE satellite observations in 2002-2014. Canadian Journal of Remote Sensing, 42, 190-202, doi: https://doi.org/10.1080/07038992.2016.1171132.\u003c/li\u003e\n\u003cli\u003eWang, S., 2019, Freezing temperature controls winter water discharge for cold region watershed. Water Resources Research, 55, 10,479-10,493. doi: https://doi.org/10.1029/2019WR026030.\u003c/li\u003e\n\u003cli\u003eWang, S., J. Huang, D. Yang, G. Pavlic, J. Li, 2015, Long-term water budget imbalances and error sources for cold region drainage basins. Hydrological Processes, 29, 2125\u0026ndash;2136, doi: https://doi.org/10.1002/hyp.10343.\u003c/li\u003e\n\u003cli\u003eWang, S., J. Huang, J. Li, A. Rivera, D.W. McKenney, and J. Sheffield, 2014, Assessment of water budget for sixteen large drainage basins in Canada. Journal of Hydrology, 512: 1-15, doi: https://doi.org/10.1016/j.jhydrol.2014.02.058.\u003c/li\u003e\n\u003cli\u003eWang, S., Li, J., and Russell, H. A. J., 2021: A novel method for cold-region streamflow hydrograph separation using GRACE satellite observations. \u003cem\u003eHydrol. Earth Syst. Sci.,\u003c/em\u003e 25, 2649\u0026ndash;2662, https://doi.org/10.5194/hess-25-2649-2021.\u003c/li\u003e\n\u003cli\u003eWang, S., Zhou, F., Russell, H. A. J., 2017, Estimating snow mass and peak river flows for the Mackenzie River basin using GRACE satellite observations. Remote Sensing, 9, 256, doi: https://doi.org/10.3390/rs9030256.\u003c/li\u003e\n\u003cli\u003eXiong, Jinghua \u0026amp; Guo, Shenglian \u0026amp; , Abhishek \u0026amp; Chen, Jie \u0026amp; Yin, Jiabo. (2022). Global evaluation of the \u0026ldquo;dry gets drier, and wet gets wetter\u0026rdquo; paradigm from a terrestrial water storage change perspective. Hydrology and Earth System Sciences. 26. 6457-6476. 10.5194/hess-26-6457-2022.\u003c/li\u003e\n\u003cli\u003eYu, Q., S. Wang, H. He, K. Yang, L. Ma, J. Li, 2021, Reconstructing GRACE-like TWS Anomalies for the Canadian Landmass Using Deep Learning and Land Surface Model. International Journal of Applied Earth Observation and Geoinformation, 102, 102404, DOI: https://doi.org/10.1016/j.jag.2021.102404.\u003c/li\u003e\n\u003cli\u003eZhong, D., S. Wang, and J. Li, 2021, Spatiotemporal Downscaling of GRACE Total Water Storage Using Land Surface Model Outputs. Remote Sensing, 13, 900, https://doi.org/10.3390/rs13050900. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6253460/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6253460/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eTerrestrial water storage (TWS) changes significantly influence the global water cycle and the development of better-informed water policies. Here we investigate TWS variations across Canada’s landmass using the GRACE and GRACE-FO satellites observations. We show that Canada’s TWS exhibited an accelerated downward trend in 2002–2024, resulting in a total water loss of 2430 km\u003csup\u003e3\u003c/sup\u003e that corresponds to a global sea-level rise of 6.9 mm. The loss was mainly driven by glacier and snow melt over the Pacific Cordillera and the Arctic Cordillera, as well as permafrost degradation in central Canada. For ice-free regions, the dry areas became drier, and the wet areas became wetter. For ice-covered regions, the water loss rate accelerated over the Pacific Cordillera but slowed down over the Arctic Cordillera and the permafrost regions. As water fluxes accelerate in a warming climate, it is expected that future TWS trends may experience further changes.\u003c/p\u003e","manuscriptTitle":"Accelerated water loss over Canada’s landmass in 2002–2024","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-21 11:22:22","doi":"10.21203/rs.3.rs-6253460/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9eed7173-13ad-4d85-b9a1-53dd922ce535","owner":[],"postedDate":"April 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":47395193,"name":"Earth and environmental sciences/Hydrology"},{"id":47395194,"name":"Earth and environmental sciences/Climate sciences/Climate change"}],"tags":[],"updatedAt":"2025-05-21T19:23:29+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-21 11:22:22","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6253460","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6253460","identity":"rs-6253460","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}