Next Article in Journal
On-Orbit Calibration and Wet Tropospheric Correction of HY-2C Correction Microwave Radiometer
Previous Article in Journal
Condition Rating of Bridge Decks with Fuzzy Sets Modeling for SF-GPR Surveys
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 15
Issue 14
10.3390/rs15143632
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Satellite-Based Approach for Quantifying Terrestrial Water Cycle Intensity

by
Fabian J. Zowam
,
Adam M. Milewski
* and
David F. Richards IV
Water Resources & Remote Sensing Laboratory (WRRS), Department of Geology, University of Georgia, 210 Field Street, 306 Geography-Geology Building, Athens, GA 30602, USA
*
Author to whom correspondence should be addressed.
Remote Sens. 2023, 15(14), 3632; https://doi.org/10.3390/rs15143632
Submission received: 12 May 2023 / Revised: 16 July 2023 / Accepted: 18 July 2023 / Published: 21 July 2023
Browse Figures
Versions Notes

Abstract

:
The terrestrial water cycle intensity (WCI) is a widely used tool to quantify the impact of climate change on the distribution of global water resources. In this study, a satellite-based WCI was tested by comparing the parameter-elevation regressions on independent slopes model (PRISM) precipitation estimates with those of the Global Precipitation Measurement (GPM) satellite mission across the contiguous United States (CONUS), based on an existing Köppen–Geiger climate classification for the CONUS. Both precipitation products were not statistically different across all climate classes. Consequently, satellite-based WCI changes between two multiannual periods (2001 to 2009 and 2010 to 2019) were calculated at a 0.1-degree spatial resolution using the GPM and a validated global evapotranspiration dataset. This study showed that: (1) The water cycle is speeding up in many parts of the CONUS, particularly the West, driven by recent increases in both precipitation and evapotranspiration through much of the region. (2) The El Niño-Southern Oscillation (ENSO) may be influencing the WCI of the CONUS by driving precipitation in the west, southeast, and parts of the north, and dryness in the northeast regions. The hydrological impacts of these results cannot be generalized. However, flood and drought risks, water availability and quality issues remain key primary concerns.

Graphical Abstract

1. Introduction

The terrestrial water cycle is the continuous circulation and redistribution of water between the earth’s land surface and atmosphere. The intensity of this circulation over any spatial-temporal scale is expressed as the sum of precipitation and evapotranspiration [1]. Globally, precipitation and evapotranspiration rates are increasing as mean air temperatures continue to rise due to global warming [2]. The period from 2012 to 2021 was the warmest decade recorded since 1901, where the warmest years were 2016 and 2020 [3]. In the United States, the last four decades, prior to 2021, have seen faster warming than the global average [3]. It is expected that these increased temperatures will strengthen the atmosphere’s evaporative demand, causing an increase in evapotranspiration rates due to water availability. This was evident during two fifteen-year periods spanning from 1984 to 2015, where global evapotranspiration increased by at least 3.57 cubic miles/year as the areas covered by water also increased [4]. Increases in global evapotranspiration rates will increase the likelihood and frequencies of droughts in many parts of the world just as flood risks increase with increasing precipitation intensities. While floods and droughts are both undesirable and extreme hydrological events, droughts can particularly disrupt the natural recharge of both surface and groundwater systems and potentially threaten global water security. Thus, the terrestrial water cycle intensity (WCI) is a valuable tool for monitoring and assessing changes in specific components of the water cycle in response to a changing climate.
By using climate variables from the parameter-elevation regressions on independent slopes model (PRISM) to calculate WCI changes for the contiguous United States (CONUS) between the averages of 1945 to 1974 and 1985 to 2014 [1], two possible considerations became apparent. (1) The temporal spread of their study did not exclusively address current climate change anomalies. In the CONUS, the last two decades prior to 2021 have shown most of the warmest years in a 120-year record [3]. (2) PRISM uses point measurements of climatic observations from a wide range of monitoring networks to develop spatially continuous climate datasets. Such methodology will be less effective in data-sparse regions of the world. In fact, most of the earth’s surface lacks adequate in situ precipitation measuring stations [5,6,7]. In developing countries, the situation is not improving [8,9,10]. Cases of inadequacy have been well reported over the continents of Africa [10,11,12,13,14,15] and South America [12,16,17,18,19]. To put this into perspective, the average rain gauge density in the United States is 1.3 gauges per 1000 sq. km [20], but Nigeria, one of the largest developing African countries, has only 87 operational stations and is 970 short of achieving a density of 1 gauge per 874 sq. km [21]. In many regions of Africa and South America, the rain gauge density is as low as 1 gauge per 100,000 sq. km [22]. Of course, understanding WCI changes in these regions where hydrometeorological data may be limited is as useful as it is over data-rich areas. Therefore, addressing both issues is to evaluate the feasibility of using satellite-sourced variables to understand the implications of recent temperature anomalies for the WCI of the CONUS.
The tropical rainfall measuring mission (TRMM) was the first satellite devoted to measuring precipitation from space [23]. It was launched in 1997 and estimated tropical rainfall for about 17 years [24,25,26]. Since the evolution of TRMM and its successor, the GPM mission, several researchers have evaluated the ability of various satellite-derived precipitation products to capture the patterns and intensity of rainfall. In the Middle East, between 1998 and 2013, three of these products namely: the TRMM Multi-Satellite Precipitation Analysis 3B42 (TRMM-3B42) product, the Climate Prediction Center MORPHing technique (CMORPH), and the precipitation estimation from remotely sensed information using artificial neural networks (PERSIANN), failed to effectively replicate severe daily rainfall events [27]. Similar results were reported for a mountainous region in eastern Italy between 2003 and 2010 [28]. In Africa, the TRMM-3B42 was able to detect spatial and seasonal rainfall patterns, and reasonably estimate high-intensity events over the Blue Nile basin in Sudan between 2001 and 2016 [29]. Over the Three Gorges Reservoir area of China, between 2001 and 2016, all three products showed varying strengths in terms of rainfall amount, extreme precipitation, and rainy-day detection ability [30]. In most of these cases, TRMM proved to be a reliable source for continuous measurements in space and time at the monthly and annual scales. Its successor, the GPM mission, made use of the most advanced instruments in space [31] and improved spatial resolution (0.25° to 0.1°), revisit times (3 h to 30 min), and latitudinal coverage (quasi-global) compared to TRMM [26]. Despite these advancements, the integrated multi-satellite retrievals for the GPM (IMERG) algorithm, which estimates precipitation from GPM constellation retrievals, incorporates early TRMM estimates in its latest version to produce a consistent, long-term precipitation record [31,32]. Otherwise, systematic biases exist between both missions, underscoring the need to make appropriate considerations when utilizing their respective products. In east-central China, the final IMERG product outperformed the early IMERG, late IMERG, near-real-time, and post-processing TRMM products, providing the most accurate estimation of daily and monthly rainfall [33]. Over Singapore, the GPM IMERG performed better than two TRMM products (3B42 and 3B42RT) in detecting precipitation, capturing variabilities, and providing more accurate daily estimates [24]. Over a mountainous region in southwest China, the GPM IMERG and 3B42 products were evaluated against observed rain gauge data from a dense gauge network, where the IMERG product showed improved capabilities to capture rainfall variability and detect medium and high-intensity events but tended to overestimate the high-intensity events [34]. In the CONUS, the GPM IMERG performed better than the 3B42 product in capturing precipitation intensity variations, reducing missed-precipitation bias for winter and summer precipitation, reducing false-precipitation bias for summer precipitation, and showing better consistency in capturing spatial distribution patterns at monthly time scales, based on comparisons with ground-based, gauge-corrected radar systems [35]. Globally, IMERG enhances precipitation detection capacity, outperforms other satellite products in its ability to capture spatiotemporal variability of extreme events, and is one of the best alternatives to ground-based measurements [36].
On the other hand, satellite remote sensing of ET generally has lesser efficiencies than precipitation, with significant disparities among various observatory satellites. However, in a recent study, Elnashar et al. (2021) [37] ranked the performance of 12 global (satellite) ET datasets after validation against flux eddy covariance ET from 645 sites based on six metrics. The authors synthesized the best-performing products into a single, much-improved ensemble global ET dataset across all land cover types, from 1982 to 2019. Their methodological approach provided some insight for this study.
The main objectives of this study are twofold. (1) To develop a framework to test the ability of a remote sensing WCI by first evaluating the reliability of satellite products to estimate precipitation over the CONUS. Monthly PRISM and GPM IMERG precipitation estimates between January 2001 and December 2019 will be compared for different climates based on three metrics and a statistical test on their difference. (2) Upon successful completion of the first objective, calculate WCI changes between two separate multiannual periods (2001–2009 and 2010–2019), using the GPM precipitation product and the Elnashar et al. (2021) validated ET datasets (Figure 1).
Shifts in the water cycle are among the most consequential effects of climate change [38]. There are also naturally occurring climate variability signals that potentially modify regional climates and alter WCI patterns. The quasi-periodic fluctuation in equatorial Pacific Ocean temperatures, otherwise known as the El Niño—Southern Oscillation (ENSO) is the most prominent year-to-year climate variation affecting underlying weather and climate patterns on Earth [39]. Sea surface temperatures (SSTs) act as natural indicators of these events, where above-average temperatures may indicate a warm (El Niño) phase and below-average temperatures, a cold (La Niña) phase. Various impacts of El Niño and La Niña on the climatology of the CONUS have been well reported—some of which include: the correlation between summer rainfall in the central CONUS from 1950 to 1990 and El Niño events [40], correlations between the 1988/1993 North America summer droughts/floods and La Niña/El Niño signals in those respective years [41], the influence of El Niño on winter precipitation in California from 1901 to 2010 [42], and the attribution of the 2012 drought in the south and south-central CONUS to the recurring 2010–2011 and 2011–2012 La Niña episodes [43]. Therefore, in addition to calculating satellite-based WCIs, this study also aims to quantify the influence of these events on WCI changes across the CONUS for the selected period of study.

2. Materials and Methods

The final run of the IMERG fusion, the most suitable version for research purposes [44], was used in this study. It was accessed and downloaded from the NASA data portal (https://gpm.nasa.gov/data/directory, accessed on 11 May 2021). A gridded global precipitation dataset with 0.1-degree spatial (~11 km) and monthly temporal scale was downloaded in Tiff format for January 2001 to December 2019.
A gridded PRISM precipitation raster dataset for the CONUS with 4 km (~0.04 degree) cell size and monthly temporal scale was also downloaded. This dataset, acquired in Bil formats from the Oregon State PRISM data portal (https://prism.oregonstate.edu/, accessed on 11 November 2022) for January 2001 to December 2019, was resampled to the spatial resolution of the GPM rasters.
The [37] validated global ET research product was accessed from the Harvard Dataverse research data repository (https://doi.org/10.7910/DVN/ZGOUED, accessed on 3 July 2021). A gridded monthly ET dataset with 0.01-degree (~1 km) spatial resolution was downloaded in Tiff format for the same period covered in both precipitation datasets. The downloaded rasters were upscaled to the spatial resolution of the precipitation datasets.
The gridded GPM and PRISM products were compared for various climates based on the Beck et al. (2018) [45] present-day Köppen–Geiger climate classification (Figure 2). The first letters in the classification scheme (A, B, C, D) represent the main climate types: tropical, arid, temperate, and cold [45]. The second (W, S, f, s, w, m) and third letters (h, k, a, b, c,) indicate moisture and air temperature characteristics, respectively [46]. The Csa climate class, for example, represents a temperate, mild climate with more precipitation in the winter than in summer where the summer months are hot, and is identical to the Csb except that it has hotter summers (Figure 2).
For each of the 17 different climate classes over the CONUS, 17 (0.1-degree) cells, each representing a climate class, were randomly selected (Figure 2). Monthly precipitation estimates were then extracted from both precipitation datasets, and the procedure was repeated twice to obtain a total of three sample runs and 51 (0.1-degree) cells—three cells representing each climate class.
A paired t-test of the difference in means between the GPM and PRISM precipitation estimates was performed on each set of 17 randomly selected points, where a null hypothesis of zero was tested at a significance level of 0.05 (5%). First, a differencing technique to eliminate spatial and time correlations was applied to the datasets. Differencing both datasets reduced the number of observations by one.
The datasets were further compared based on three model evaluation metrics: coefficient of determination (R2), Nash–Sutcliffe efficiency (NSE), and Kling–Gupta efficiency (KGE). The NSE compared each observed (GPM) value with its corresponding simulated (PRISM) value, normalizing their difference by the variance of the observed time series. It ranges from negative infinity to one, with higher values indicating a well-varied observed time series that aligns closely with the simulated values. Like the NSE, the KGE ranges from negative infinity to one but with more straightforward relationships between correlation, variability, and bias [47].
Following successful evaluations at the 51 randomly selected points, monthly WCI was calculated for each 0.1-degree cell across the CONUS by summing up the GPM IMERG precipitation and corresponding evapotranspiration values from Elnashar et al. (2021). The total annual WCI for each year was obtained by adding these monthly WCI values, year by year, and an annual average was calculated by averaging the cell values over the period of study (2001 to 2019). In addition, the WCI difference between the annual averages of 2001 to 2009 and 2010 to 2019 was calculated to show temporal trends and spatial patterns of change between both multiannual periods.
El Niño (or La Niña) events in the Niño 3.4 region of the tropical Pacific Ocean have been defined differently in various studies, such as occurring when the December-January-February (DJF) SST anomaly exceeds +/−0.5 °C [48,49], or when five consecutive three-month running means of SST anomalies exceed +/−0.5 °C between July and June [50]. Although definitions may vary, such events generally occur when anomalies exceed +/−0.5 C for several months [51]. In this study, a straightforward approach was used to analyze SST anomalies in the tropical Pacific Ocean. We simply calculated annual averages of the three-month running means of SST anomalies obtained from [52]. The anomalies were originally derived by calculating SST departures from 30-year average baseline temperature conditions [52]. Applying the threshold of +/−0.5 °C, years with significant warm and cold phases were identified. These years were excluded from the dataset and WCI calculations were repeated. However, years that exceeded both El Niño and La Niña thresholds were considered neutral, indicating a lack of clear dominance of either El Niño or La Niña conditions, and therefore were not removed from the datasets.

3. Results

This section consists of two parts to address the study’s main objectives:

3.1. Validation

The t-tests yielded identical results across all three sample runs, although only the results from the first run are presented (Table 1). P-values were greater than our chosen significance level of 0.05, so we would accept the null hypothesis that both precipitation products are not statistically different. The confidence interval (CI), which represents a range of acceptable null hypotheses, also included a zero at every sample point. Thus, we could still not rule out a zero difference in means between both datasets and would, therefore, accept the null hypothesis that they are not statistically different [53]. Also, (Cohen’s d) effect size calculations showed negligible differences between the GPM and PRISM precipitation datasets. Because of how the tests were set up, negative values imply that the mean of the GPM dataset was lesser than that of the PRISM dataset.
Further evaluations based on the three performance metrics showed that the Dfc (cold, no dry season, cold summer) climate class, having the lowest R2 on all three sample runs, the lowest NSE for the second and third runs, and the lowest KGE value on the first run, demonstrated relatively weak agreement compared to other classes (Figure 3). Representing the climate of the Rockies (Figure 4), it is still unclear whether the weak agreements were due to terrain complexities, climatic factors, both, or other factors. Therefore, further investigation may be necessary to specifically determine the underlying causes of the relatively poor performance of the GPM satellite product over the Dfc climate.
Based on recommended R2 and NSE standards for monthly periods and local scales [54], agreement across all Köppen–Geiger climates was at least satisfactory on average. A ranking of climate classes by their relative agreements between both gridded precipitation products showed that while the Dfc was weakest, the Bwk (arid, desert, cold) class showed the strongest agreement (Figure 5)

3.2. WCI Analyses

The average annual WCI between 2001 and 2019 varied across the CONUS. The west generally showed lower intensities, whereas the highest values were seen in the southeastern CONUS region. However, in the westernmost border around the western Washington region, values of over 4000 mm/yr were recorded (Figure 6a). Similar patterns were seen for both precipitation and evapotranspiration averages. Two essential ingredients for precipitation are moisture and lift. Thus, with the Gulf of Mexico as a potential supplier of moisture, combined with the existence of several mountain ranges in the region, it is no surprise that there is that much rain in the southeast region. Likewise, the active weather caused by the low-pressure system of the Aleutian Islands explains the excessive amounts of rain along the western boundary (Figure 6b). To illustrate the importance of available moisture for evapotranspiration, the areas with high precipitation (Figure 6b) appeared to also show relatively high evapotranspiration rates (Figure 6c).
To calculate the differences, the averages of the first period (2001–2009) were subtracted from the second (2010–2019). Following the initial calculations, El Niño, and La Niña years (Figure 7) were then removed from the datasets based on average annual SST anomalies and a threshold of +/−0.5 °C. The El Niño years 2002, 2006, and 2015, and La Niña years 2007, 2008, and 2011 (Figure 7) were also identified in a different, independent study that compared average November-December-January (NDJ) and December-January-February (DJF) SST anomalies in the Niño 3.4 region with predefined threshold anomalies [55].
The west CONUS showed notable increases in WCI between both periods (Figure 8a), mostly due to increases in both precipitation and evapotranspiration in the second period (Figure 8b,c). However, our adjusted results (Figure 8d–f) show that during the period of study, ENSO impacted the WCI for the CONUS primarily by bringing more water to the southeast, west, and parts of the north and less water to the northeast regions.

4. Discussion

Satellite precipitation retrievals generally experience difficulties due to complex terrains and climate [5,56,57,58,59,60,61]. The GPM, in particular, is less accurate in mountainous regions than in plains [36,62,63,64,65,66]. Specific to the CONUS, snowfall underestimation of the GPM has been recorded in the western mountainous regions [36]. Likewise, we detected an underestimation in the Dfc climate over the Rockies. For each sample point in the region, the NEXRAD-TDWR radar coverage [67] showed a beam height of less than 3000 feet above ground level, indicating good ground-based radar accuracy for the PRISM input. In addition, a local hill shade DEM for each point showed complex mountainous terrain that can affect GPM accuracy and a high elevation where snowfall and seasonal snow cover is expected. The research suggests that the complex terrain and high amounts of snow in the winter played a part in the underestimation of precipitation by the GPM for the Dfc climate region of the CONUS. However, micro-scale studies at finer spatial resolutions that account for more variables, such as local radar propagation characteristics and GPM flyovers to PRISM updates, are necessary to state this conclusively.
The WCI results presented In this study reinforce the intuition that with precipitation and evapotranspiration as indicators, the WCI over any area might increase through one of three scenarios: (1) Increases in both precipitation and evapotranspiration, as seen in parts of the west CONUS. A few studies validate this [68,69], where the warming-induced intensification of the water cycle over the Midwest resulted in increased precipitation and evaporation rates. (2) A dominating influence of precipitation, notably in the Appalachian region of the east CONUS. This scenario may carry flood risks, as flood occurrences have consistently been reported in the south and central Appalachian regions [70,71,72]. (3) A dominating influence of evapotranspiration, showing examples in parts of the west and southeast regions. Increases in drought risks in the West have been attributed to a combination of atmospheric, environmental, and consumptive water demands exceeding supply [73,74]. In general, scenarios 1 and 3 may directly impact regional water availability. Groundwater levels, for example, declined in the Midwest where the intensification of the water cycle increased precipitation and evaporation rates [68].
About 53% of the CONUS showed a positive WCI change, which implies that the water cycle is currently speeding up in and around more than half of the country. Of this, 58.2% showed increases in both average precipitation and evapotranspiration (scenario 1). An increase in precipitation but decrease in evapotranspiration (scenario 2) constituted 21.2%, while the reverse (scenario 3) made up 19.1%. Without ENSO years, 40.4% of the CONUS showed an intensified water cycle where 55.6%, 18.5%, and 24.5% represented the three scenarios, respectively, thereby providing supporting evidence that ENSO may be influencing the WCI of the CONUS. In both cases, however, the dominant representation of a faster water cycle for the periods of investigation is characterized by a simultaneous increase in both precipitation and evapotranspiration, where the increase in evapotranspiration is presumably the result of higher air temperatures and an increase in moisture availability. In a general sense, the impacts of ENSO on the WCI of the CONUS were primarily driven by changes in precipitation (Figure 8g–i).
A faster water cycle presents a variety of hydrological consequences. Extreme precipitation and flood events become more likely, drought risks intensify, and the availability and quality of surface and groundwater resources can be impacted. For example, in coastal areas where aquifers are susceptible to saltwater intrusion, decreasing groundwater recharge rates may lower the water table, allow the infiltration of saltwater, and consequently increase the severity of water quality issues. Unfortunately, these effects cannot be generalized for the CONUS. Therefore, further work at the local, watershed, and regional scales is required to categorically express representative hydrological impacts for the various spatial (and temporal) scales of interest.

5. Conclusions

This study addressed two potential considerations within the framework of the Huntington et al. (2018) [1] study. For the same region, and with similar hydroclimatic variables, we evaluated the feasibility of a remote sensing WCI for a different, more recent period, allowing us to relate WCI changes to current climate change using satellite-based datasets. The period of investigation (2001–2019) was also determined by the availability of complete-year data for the GPM IMERG precipitation (2001–present) and evapotranspiration (1982–2019) datasets. To address the limitations of satellite-based precipitation estimates, particularly in complex terrains, we incorporated comparisons with ground-based PRISM estimates to provide validation for the satellite-based dataset and additional context on its limitations. We showed that the GPM IMERG precipitation product was not statistically different from ground-based gridded PRISM estimates across all Köppen–Geiger climates within the CONUS despite inherent challenges over complex mountainous terrains.
Using the framework introduced by Huntington et al. (2018), we present the first satellite-based quantification of the terrestrial water cycle intensity. Comparing the results of both studies, the spatial patterns of average WCI changes presented in this study contrast those of the framework study. As we’ve shown, between the averages of 2001 to 2009 and 2010 to 2019, faster WCIs were seen around the west CONUS, contrary to the Huntington et al. (2018) study that showed a vulnerability of the east CONUS between the averages of 1945 to 1974 and 1985 to 2014. This not only substantiates the influence of current temperature anomalies on the spatial patterns of regional water cycle intensities for the CONUS but the susceptibility of the global water cycle to climate variabilities. In addition, we attempted to more accurately associate climate change with shifts in WCIs by eliminating years that showed strong occurrences of El Niño and La Niña events from our calculations. Our recommendation for future studies is to address and quantify the impacts of other relevant cyclic patterns, particularly those not directly influenced by ENSO (El Niño and La Niña) events.
Satellite missions for hydroclimatic variables different from those utilized in this study have also enhanced various other studies relating to the water cycle. Examples include GRACE (Gravity Recovery and Climate Experiment) [75] and its follow-on mission GRACE-FO [38] for water storage anomalies, SMOS (soil moisture and ocean salinity) [76,77] and SMAP (soil moisture active passive) [78] for soil moisture, and the Water Cycle Observation Mission (WCOM) for various components of the water cycle [79]. While WCIs generally inform the availability and distribution of water resources, remote sensing provides additional benefits including the availability and consistency of measurements to calculate the WCIs for wider spatial coverages and hydrologically inaccessible and data-sparse areas. A satellite-based WCI approach can be applied to any geographical region, covering spatial and temporal scales for which reliable satellite-based estimates are available.
In particular, the ability to calculate the WCI for any point on the globe and be able to determine if WCI changes are primarily driven by changes in precipitation, evaporation, or both, can allow regional water management agencies to make better-informed decisions on the storage and efficient distribution of water supplies. For example, negative WCI changes where precipitation is the decreasing variable and evapotranspiration is constant suggests that more water may need to be imported to the region or stored than previously in order to support the same amount of evapotranspiration. Conversely, positive WCI changes driven by evapotranspiration indicate the vital role of limited soil moisture in reducing the vapor pressure deficit in the atmosphere. Possible remedies may include supplementary water supply systems such as irrigation schemes to sustain agricultural practices, or importation to secure water for households and industries. Leveraging insights from WCI changes and harnessing the benefits of remote sensing for such calculations will help mitigate and overcome various water resources and associated socio-economic issues. It is our hope that this knowledge contributes to further work in refining water budget calculations to plan for an uncertain climate future.

Author Contributions

A.M.M. was responsible for the project administration, supervision, review, and editing and helped with the GPM data acquisition. F.J.Z. was responsible for the conceptualization, methodology, formal analysis, and original draft preparation. D.F.R.IV helped with funding acquisition and writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the Department of Geology, University of Georgia, through the Miriam Watts-Wheeler research fund.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Acknowledgments

The authors wish to thank Max Appelbaum for his assistance in the implementation of this work.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Huntington, T.G.; Weiskel, P.K.; Wolock, D.M.; McCabe, G.J. A new indicator framework for quantifying the intensity of the terrestrial water cycle. J. Hydrol. 2018, 559, 361–372. [Google Scholar] [CrossRef]
  2. Douville, H.; Raghavan, K.; Renwick, J.; Allan, R.P.; Arias, P.A.; Barlow, M.; Cerezo-Mota, R.; Cherchi, A.; Gan, T.Y.; Gergis, J.; et al. Water Cycle Changes. In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to 45 the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; Masson-Delmotte, V., Zhai, P., Pirani, A., Connors, S.L., Péan, C., Berger, S., Caud, N., Chen, Y., Goldfarb, L., Gomis, M.I., et al., Eds.; Cambridge University Press: Cambridge, UK, 2021; pp. 1055–1210, (in press). [Google Scholar]
  3. United States Environmental Protection Agency. Climate Change Indicators: Weather and Climate. Available online: https://www.epa.gov/climate-indicators/climate-change-indicators-us-and-global-temperature (accessed on 18 November 2022).
  4. Zhan, S.; Song, C.; Wang, J.; Sheng, Y.; Quan, J. A Global Assessment of Terrestrial Evapotranspiration Increase Due to Surface Water Area Change. Earth’s Futur. 2019, 7, 266–282. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  5. Milewski, A.; Seyoum, W.M.; Elkadiri, R.; Durham, M. Multi-Scale Hydrologic Sensitivity to Climatic and Anthropogenic Changes in Northern Morocco. Geosciences 2019, 10, 13. [Google Scholar] [CrossRef] [Green Version]
  6. Milewski, A.; Elkadiri, R.; Durham, M. Assessment and Comparison of TMPA Satellite Precipitation Products in Varying Climatic and Topographic Regimes in Morocco. Remote Sens. 2015, 7, 5697–5717. [Google Scholar] [CrossRef] [Green Version]
  7. Kidd, C.; Huffman, G. Global precipitation measurement. Meteorol. Appl. 2011, 18, 334–353. [Google Scholar] [CrossRef]
  8. Nkiaka, E.; Nawaz, N.R.; Lovett, J.C. Evaluating global reanalysis precipitation datasets with rain gauge measurements in the Sudano-Sahel region: Case study of the Logone catchment, Lake Chad Basin. Meteorol. Appl. 2017, 24, 9–18. [Google Scholar] [CrossRef]
  9. Buytaert, W.; Friesen, J.; Liebe, J.; Ludwig, R. Assessment and Management of Water Resources in Developing, Semi-arid and Arid Regions. Water Resour. Manag. 2012, 26, 841–844. [Google Scholar] [CrossRef] [Green Version]
  10. Hughes, D.A. Comparison of satellite rainfall data with observations from gauging station networks. J. Hydrol. 2006, 327, 399–410. [Google Scholar] [CrossRef]
  11. Abdelmoneim, H.; Soliman, M.R.; Moghazy, H.M. Evaluation of TRMM 3B42V7 and CHIRPS Satellite Precipitation Products as an Input for Hydrological Model over Eastern Nile Basin. Earth Syst. Environ. 2020, 4, 685–698. [Google Scholar] [CrossRef]
  12. Massari, C.; Brocca, L.; Pellarin, T.; Abramowitz, G.; Filippucci, P.; Ciabatta, L.; Maggioni, V.; Kerr, Y.; Prieto, D.F. A daily 25 km short-latency rainfall product for data-scarce regions based on the integration of the Global Precipitation Measurement mission rainfall and multiple-satellite soil moisture products. Hydrol. Earth Syst. Sci. 2020, 24, 2687–2710. [Google Scholar] [CrossRef]
  13. Satgé, F.; Defrance, D.; Sultan, B.; Bonnet, M.-P.; Seyler, F.; Rouché, N.; Pierron, F.; Paturel, J.-E. Evaluation of 23 gridded precipitation datasets across West Africa. J. Hydrol. 2020, 581, 124412. [Google Scholar] [CrossRef] [Green Version]
  14. Dieulin, C.; Mahé, G.; Paturel, J.E.; Ejjiyar, S.; Tramblay, Y.; Rouché, N.; El Mansouri, B. A new 60-year 1940/1999 monthly-gridded rainfall data set for Africa. Water 2019, 11, 387. [Google Scholar] [CrossRef] [Green Version]
  15. Crétat, J.; Vizy, E.K.; Cook, K.H. How well are daily intense rainfall events captured by current climate models over Africa? Clim. Dyn. 2014, 42, 2691–2711. [Google Scholar] [CrossRef]
  16. Trejo, F.J.P.; Barbosa, H.Á.; Peñaloza-Murillo, M.A.; Moreno, M.A.; Farias, A. Intercomparison of improved satellite rainfall estimation with CHIRPS gridded product and rain gauge data over Venezuela. Atmósfera 2016, 29, 323–342. [Google Scholar] [CrossRef] [Green Version]
  17. Salio, P.; Hobouchian, M.P.; Skabar, Y.G.; Vila, D. Evaluation of high-resolution satellite precipitation estimates over southern South America using a dense rain gauge network. Atmos. Res. 2015, 163, 146–161. [Google Scholar] [CrossRef]
  18. Rozante, J.R.; Moreira, D.S.; De Goncalves, L.G.; Vila, D.A. Combining TRMM and surface observations of precipitation: Technique and validation over South America. Am. Meteorol. Soc. 2010, 25, 885–894. [Google Scholar] [CrossRef] [Green Version]
  19. de Angelis, C.F.; Mcgregor, G.R.; Kidd, C. A 3 year climatology of rainfall characteristics over tropical and subtropical South America based on tropical rainfall measuring mission precipitation radar data. Int. J. Clim. 2004, 24, 385–399. [Google Scholar] [CrossRef]
  20. Dzotsi, K.A.; Matyas, C.J.; Jones, J.W.; Baigorria, G.; Hoogenboom, G. Understanding high resolution space-time variability of rainfall in southwest Georgia, United States. Int. J. Climatol. 2014, 34, 3188–3203. [Google Scholar] [CrossRef]
  21. Ngene, B.; Agunwamba, J.; Nwachukwu, B.; Okoro, B. The Challenges to Nigerian Raingauge Network Improvement. Res. J. Environ. Earth Sci. 2015, 7, 68–74. [Google Scholar] [CrossRef]
  22. Brocca, L.; Massari, C.; Pellarin, T.; Filippucci, P.; Ciabatta, L.; Camici, S.; Kerr, Y.H.; Fernández-Prieto, D. River flow prediction in data scarce regions: Soil moisture integrated satellite rainfall products outperform rain gauge observations in West Africa. Sci. Rep. 2020, 10, 12517. [Google Scholar] [CrossRef]
  23. Kidd, C.; Levizzani, V. Status of satellite precipitation retrievals. Hydrol. Earth Syst. Sci. 2011, 15, 1109–1116. [Google Scholar] [CrossRef] [Green Version]
  24. Tan, M.L.; Duan, Z. Assessment of GPM and TRMM Precipitation Products over Singapore. Remote Sens. 2017, 9, 720. [Google Scholar] [CrossRef] [Green Version]
  25. Beria, H.; Nanda, T.; Bisht, D.S.; Chatterjee, C. Does the GPM mission improve the systematic error component in satellite rainfall estimates over TRMM? An evaluation at a pan-India scale. Hydrol. Earth Syst. Sci. 2017, 21, 6117–6134. [Google Scholar] [CrossRef] [Green Version]
  26. Chen, F.; Li, X. Evaluation of IMERG and TRMM 3B43 Monthly Precipitation Products over Mainland China. Remote Sens. 2016, 8, 472. [Google Scholar] [CrossRef] [Green Version]
  27. El Kenawy, A.M.; McCabe, M.F.; Lopez-Moreno, J.I.; Hathal, Y.; Robaa, S.M.; Al Budeiri, A.L.; Jadoon, K.Z.; Abouelmagd, A.; Eddenjal, A.; Domínguez-Castro, F.; et al. Spatial assessment of the performance of multiple high-resolution satellite-based precipitation data sets over the Middle East. Int. J. Clim. 2019, 39, 2522–2543. [Google Scholar] [CrossRef]
  28. Mei, Y.; Anagnostou, E.N.; Nikolopoulos, E.I.; Borga, M. Error Analysis of Satellite Precipitation Products in Mountainous Basins. J. Hydrometeorol. 2014, 15, 1778–1793. [Google Scholar] [CrossRef]
  29. Elhamid, A.M.; Eltahan, A.M.; Mohamed, L.M.; Hamouda, I.A. Assessment of the two satellite-based precipitation products TRMM and RFE rainfall records using ground based measurements. Alex. Eng. J. 2020, 59, 1049–1058. [Google Scholar] [CrossRef]
  30. Zhang, T.; Yang, Y.; Dong, Z.; Gui, S. A Multiscale Assessment of Three Satellite Precipitation Products (TRMM, CMORPH, and PERSIANN) in the Three Gorges Reservoir Area in China. Adv. Meteorol. 2021, 2021, 1–27. [Google Scholar] [CrossRef]
  31. Skofronick-Jackson, G.; Petersen, W.A.; Berg, W.; Kidd, C.; Stocker, E.F.; Kirschbaum, D.B.; Kakar, R.; Braun, S.A.; Huffman, G.J.; Iguchi, T.; et al. The Global Precipitation Measurement (GPM) Mission for Science and Society. Bull. Am. Meteorol. Soc. 2017, 98, 1679–1695. [Google Scholar] [CrossRef]
  32. Huffman, G.J.; Bolvin, D.T.; Nelkin, E.J.; Tan, J. Integrated Multi-Satellite Retrievals for GPM (IMERG) technical documentation. Nasa/Gsfc Code 2015, 612, 2019. [Google Scholar]
  33. Yang, X.; Lu, Y.; Tan, M.L.; Li, X.; Wang, G.; He, R. Nine-Year Systematic Evaluation of the GPM and TRMM Precipitation Products in the Shuaishui River Basin in East-Central China. Remote Sens. 2020, 12, 1042. [Google Scholar] [CrossRef] [Green Version]
  34. He, Z.; Yang, L.; Tian, F.; Ni, G.; Hou, A.; Lu, H. Intercomparisons of Rainfall Estimates from TRMM and GPM Multisatellite Products over the Upper Mekong River Basin. J. Hydrometeorol. 2017, 18, 413–430. [Google Scholar] [CrossRef] [Green Version]
  35. Wang, J.; Petersen, W.A.; Wolff, D.B. Validation of Satellite-Based Precipitation Products from TRMM to GPM. Remote Sens. 2021, 13, 1745. [Google Scholar] [CrossRef]
  36. Pradhan, R.K.; Markonis, Y.; Godoy, M.R.V.; Villalba-Pradas, A.; Andreadis, K.M.; Nikolopoulos, E.I.; Papalexiou, S.M.; Rahim, A.; Tapiador, F.J.; Hanel, M. Review of GPM IMERG performance: A global perspective. Remote Sens. Environ. 2022, 268, 112754. [Google Scholar] [CrossRef]
  37. Elnashar, A.; Wang, L.; Wu, B.; Zhu, W.; Zeng, H. Synthesis of global actual evapotranspiration from 1982 to 2019. Earth Syst. Sci. Data 2021, 13, 447–480. [Google Scholar] [CrossRef]
  38. Rodell, M.; Reager, J.T. Water cycle science enabled by the GRACE and GRACE-FO satellite missions. Nat. Water 2023, 1, 47–59. [Google Scholar] [CrossRef]
  39. McPhaden, M.J.; Zebiak, S.E.; Glantz, M.H. ENSO as an Integrating Concept in Earth Science. Science 2006, 314, 1740–1745. [Google Scholar] [CrossRef] [Green Version]
  40. Wang, Z.; Chang, C.P.; Wang, B. Impacts of El Niño and La Niña on the US climate during northern summer. J. Clim. 2007, 20, 2165–2177. [Google Scholar] [CrossRef] [Green Version]
  41. Trenberth, K.E.; Guillemot, C.J. Physical Processes Involved in the 1988 Drought and 1993 Floods in North America. J. Clim. 1996, 9, 1288–1298. [Google Scholar] [CrossRef]
  42. Jong, B.T.; Ting, M.; Seager, R. El Niño’s impact on California precipitation: Seasonality, regionality, and El Niño intensity. Environ. Res. Lett. 2016, 11, 054021. [Google Scholar] [CrossRef]
  43. Rippey, B.R. The US drought of 2012. Weather. Clim. Extrem. 2015, 10, 57–64. [Google Scholar] [CrossRef] [Green Version]
  44. Huffman, G.J.; Bolvin, D.T.; Braithwaite, D.; Hsu, K.; Joyce, R.; Xie, P.; Yoo, S.H. NASA global precipitation measurement (GPM) integrated multi-satellite retrievals for GPM (IMERG). Algorithm Theor. Basis Doc. (ATBD) 2015, 4, 26. [Google Scholar]
  45. Beck, H.E.; Zimmermann, N.E.; McVicar, T.R.; Vergopolan, N.; Berg, A.; Wood, E.F. Present and future Köppen-Geiger climate classification maps at 1-km resolution. Sci. Data 2018, 5, 180214. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  46. Kottek, M.; Grieser, J.; Beck, C.; Rudolf, B.; Rubel, F. World map of the Köppen-Geiger climate classification updated. Meteorol. Z. 2006, 15, 259–263. [Google Scholar] [CrossRef] [PubMed]
  47. Gupta, H.V.; Kling, H.; Yilmaz, K.K.; Martinez, G.F. Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modelling. J. Hydrol. 2009, 377, 80–91. [Google Scholar] [CrossRef] [Green Version]
  48. Wang, Y.; Hu, K.; Huang, G.; Tao, W. Asymmetric impacts of El Niño and La Niña on the Pacific–North American teleconnection pattern: The role of subtropical jet stream. Environ. Res. Lett. 2021, 16, 114040. [Google Scholar] [CrossRef]
  49. Jia, X.; Ge, J.; Wang, S. Diverse impacts of ENSO on wintertime rainfall over the Maritime Continent. Int. J. Clim. 2016, 36, 3384–3397. [Google Scholar] [CrossRef] [Green Version]
  50. Lam, H.C.Y.; Haines, A.; McGregor, G.; Chan, E.Y.Y.; Hajat, S. Time-Series Study of Associations between Rates of People Affected by Disasters and the El Niño Southern Oscillation (ENSO) Cycle. Int. J. Environ. Res. Public Health 2019, 16, 3146. [Google Scholar] [CrossRef] [Green Version]
  51. Met Office El Niño Southern Oscillation (ENSO) Region Sea Surface Temperature Forecasts. Available online: https://www.metoffice.gov.uk/research/climate/seasonal-to-decadal/gpc-outlooks/el-nino-la-nina (accessed on 30 June 2023).
  52. Trenberth, K. The Climate Data Guide: Nino SST Indices (Nino 1 + 2, 3, 3.4, 4; ONI and TNI). Available online: https://climatedataguide.ucar.edu/climate-data/nino-sst-indices-nino-12-3-34-4-oni-and-tni (accessed on 12 July 2022).
  53. Holland, S. Data Analysis in the Geosciences. Available online: http://strata.uga.edu/8370/lecturenotes/pvaluesConfidenceIntervals.html (accessed on 7 September 2021).
  54. Moriasi, D.N.; Gitau, M.W.; Pai, N.; Daggupati, P. Hydrologic and water quality models: Performance measures and evaluation criteria. Trans. ASABE 2015, 58, 1763–1785. [Google Scholar]
  55. Santoso, A.; Mcphaden, M.J.; Cai, W. The defining characteristics of ENSO extremes and the strong 2015/2016 El Niño. Rev. Geophys. 2017, 55, 1079–1129. [Google Scholar] [CrossRef]
  56. Gonzalez, R.L. Consistency in the AMSR-E Snow Products: Groundwork for a Coupled Snowfall and SWE Algorithm. Ph.D. Dissertation, Colorado State University, Fort Collins, CO, USA, 2019. [Google Scholar]
  57. Behrangi, A.; Bormann, K.J.; Painter, T.H. Using the Airborne Snow Observatory to Assess Remotely Sensed Snowfall Products in the California Sierra Nevada. Water Resour. Res. 2018, 54, 7331–7346. [Google Scholar] [CrossRef] [Green Version]
  58. Tang, G.; Ma, Y.; Long, D.; Zhong, L.; Hong, Y. Evaluation of GPM Day-1 IMERG and TMPA Version-7 legacy products over Mainland China at multiple spatiotemporal scales. J. Hydrol. 2016, 533, 152–167. [Google Scholar] [CrossRef]
  59. Seyyedi, H.; Anagnostou, E.N.; Beighley, E.; McCollum, J. Satellite-driven downscaling of global reanalysis precipitation products for hydrological applications. Hydrol. Earth Syst. Sci. 2014, 18, 5077–5091. [Google Scholar] [CrossRef] [Green Version]
  60. AghaKouchak, A.; Behrangi, A.; Sorooshian, S.; Hsu, K.; Amitai, E. Evaluation of satellite-retrieved extreme precipitation rates across the central United States. J. Geophys. Res. Atmos. 2011, 116. [Google Scholar] [CrossRef]
  61. Turk, F.; Miller, S. Toward improved characterization of remotely sensed precipitation regimes with MODIS/AMSR-E blended data techniques. IEEE Trans. Geosci. Remote Sens. 2005, 43, 1059–1069. [Google Scholar] [CrossRef]
  62. Moazami, S.; Najafi, M. A comprehensive evaluation of GPM-IMERG V06 and MRMS with hourly ground-based precipitation observations across Canada. J. Hydrol. 2021, 594, 125929. [Google Scholar] [CrossRef]
  63. Yang, M.; Liu, G.; Chen, T.; Chen, Y.; Xia, C. Evaluation of GPM IMERG precipitation products with the point rain gauge records over Sichuan, China. Atmos. Res. 2020, 246, 105101. [Google Scholar] [CrossRef]
  64. Lu, X.; Tang, G.; Wang, X.; Liu, Y.; Jia, L.; Xie, G.; Li, S.; Zhang, Y. Correcting GPM IMERG precipitation data over the Tianshan Mountains in China. J. Hydrol. 2019, 575, 1239–1252. [Google Scholar] [CrossRef]
  65. Wu, Y.; Zhang, Z.; Huang, Y.; Jin, Q.; Chen, X.; Chang, J. Evaluation of the GPM IMERG v5 and TRMM 3B42 v7 Precipitation Products in the Yangtze River Basin, China. Water 2019, 11, 1459. [Google Scholar] [CrossRef] [Green Version]
  66. Sungmin, O.; Kirstetter, P. Evaluation of diurnal variation of GPM IMERG-derived summer precipitation over the contiguous US using MRMS data. Q. J. R. Meteorol. Soc. 2018, 144, 270–281. [Google Scholar] [CrossRef] [Green Version]
  67. Center, R.O. NEXRAD and TDWR Radar Locations. Available online: https://www.roc.noaa.gov/WSR88D/Maps.aspx (accessed on 28 November 2022).
  68. Yeh, P.J.F.; Wu, C.H. Recent Acceleration of the Terrestrial Hydrologic Cycle in the U.S. Midwest. J. Geophys. Res. Atmos. 2018, 123, 2993–3008. [Google Scholar] [CrossRef]
  69. Baker, J.M.; Griffis, T.J.; Ochsner, T.E. Coupling landscape water storage and supplemental irrigation to increase productivity and improve environmental stewardship in the U.S. Midwest. Water Resour. Res. 2012, 48. [Google Scholar] [CrossRef]
  70. Miller, D.; Forsythe, J.; Kusselson, S.; Straka III, W.; Yin, J.; Zhan, X.; Ferraro, R. A Study of Two Impactful Heavy Rainfall Events in the Southern Appalachian Mountains during Early 2020, Part I; Societal Impacts, Synoptic Overview, and Historical Context. Remote Sens. 2021, 13, 2452. [Google Scholar] [CrossRef]
  71. Wix, J.M. Constructing a Flash Flood Climatology of the Appalachian Mountains: Comparison of Eastern Kentucky and West Virginia Summertime Rain Events. Bachelor’s Thesis, Western Kentucky University, Bowling Green, KY, USA, 2010. [Google Scholar]
  72. Gaffin, D.M.; Hotz, D.G. A precipitation and flood climatology with synoptic features of heavy rainfall across the southern Appalachian Mountains. Natl. Weather. Dig. 2000, 24, 3–16. [Google Scholar]
  73. Martin, J.T.; Pederson, G.T.; Woodhouse, C.A.; Cook, E.R.; McCabe, G.J.; Anchukaitis, K.J.; Wise, E.K.; Erger, P.J.; Dolan, L.; McGuire, M.; et al. Increased drought severity tracks warming in the United States’ largest river basin. Proc. Natl. Acad. Sci. USA 2020, 117, 11328–11336. [Google Scholar] [CrossRef] [PubMed]
  74. US Bureau of Reclamation. Colorado River Basin Water Supply and Demand Study: Executive Summary. Available online: https://www.usbr.gov/watersmart/bsp/docs/finalreport/ColoradoRiver/CRBS_Executive_Summary_FINAL.pdf (accessed on 18 November 2022).
  75. Eicker, A.; Forootan, E.; Springer, A.; Longuevergne, L.; Kusche, J. Does GRACE see the terrestrial water cycle “intensifying”? J. Geophys. Res. Atmos. 2016, 121, 733–745. [Google Scholar] [CrossRef] [Green Version]
  76. Durack, P. Ocean Salinity and the Global Water Cycle. Oceanography 2015, 28, 20–31. [Google Scholar] [CrossRef] [Green Version]
  77. Reul, N.; Fournier, S.; Boutin, J.; Hernandez, O.; Maes, C.; Chapron, B.; Alory, G.; Quilfen, Y.; Tenerelli, J.; Morisset, S.; et al. Sea Surface Salinity Observations from Space with the SMOS Satellite: A New Means to Monitor the Marine Branch of the Water Cycle. Surv. Geophys. 2014, 35, 681–722. [Google Scholar] [CrossRef] [Green Version]
  78. Entekhabi, D.; Njoku, E.G.; O’Neill, P.E.; Kellogg, K.H.; Crow, W.T.; Edelstein, W.N.; Entin, J.K.; Goodman, S.D.; Jackson, T.J.; Johnson, J.; et al. The Soil Moisture Active Passive (SMAP) Mission. Proc. IEEE 2010, 98, 704–716. [Google Scholar] [CrossRef]
  79. Shi, J.; Dong, X.; Zhao, T.; Du, Y.; Liu, H.; Wang, Z.; Zhu, D.; Ji, D.; Xiong, C.; Jiang, L. The water cycle observation mission (WCOM): Overview. IEEE Int. Geosci. Remote Sens. Symp. 2016, 3430–3433. [Google Scholar] [CrossRef]
Remotesensing 15 03632 g001 550
Figure 1. Flow chart showing the conceptual framework, data acquisition, and processing steps towards the main objectives of the study. The blue boxes represent the precipitation datasets, and the red box represents the evapotranspiration dataset used in the study.
Figure 1. Flow chart showing the conceptual framework, data acquisition, and processing steps towards the main objectives of the study. The blue boxes represent the precipitation datasets, and the red box represents the evapotranspiration dataset used in the study.
Remotesensing 15 03632 g001
Remotesensing 15 03632 g002 550
Figure 2. Present day (1980–2016) Köppen–Geiger climate classification map by [45] showing 17 classes across the CONUS and the 17 randomly selected sample points for the first sample run.
Figure 2. Present day (1980–2016) Köppen–Geiger climate classification map by [45] showing 17 classes across the CONUS and the 17 randomly selected sample points for the first sample run.
Remotesensing 15 03632 g002
Remotesensing 15 03632 g003 550
Figure 3. Comparison of GPM and PRISM precipitation datasets based on R2, NSE, and KGE. The numbers (1), (2), and (3) represent the first, second, and third sample runs. The Dfc climate is represented as climate class 17.
Figure 3. Comparison of GPM and PRISM precipitation datasets based on R2, NSE, and KGE. The numbers (1), (2), and (3) represent the first, second, and third sample runs. The Dfc climate is represented as climate class 17.
Remotesensing 15 03632 g003
Remotesensing 15 03632 g004 550
Figure 4. The Dfc climate region where GPM and PRISM showed the weakest agreement.
Figure 4. The Dfc climate region where GPM and PRISM showed the weakest agreement.
Remotesensing 15 03632 g004
Remotesensing 15 03632 g005 550
Figure 5. Ranking of the 17 climate classes based on relative agreements between the GPM and PRISM gridded precipitation products, according to R2, NSE, and KGE values.
Figure 5. Ranking of the 17 climate classes based on relative agreements between the GPM and PRISM gridded precipitation products, according to R2, NSE, and KGE values.
Remotesensing 15 03632 g005
Remotesensing 15 03632 g006 550
Figure 6. Average annual (a) water cycle intensity, (b) precipitation, and (c) evapotranspiration from 2001 to 2019.
Figure 6. Average annual (a) water cycle intensity, (b) precipitation, and (c) evapotranspiration from 2001 to 2019.
Remotesensing 15 03632 g006
Remotesensing 15 03632 g007 550
Figure 7. El Niño and La Niña years over the period of study. Average annual SST anomalies were compared with a threshold anomaly of +/−0.5 °C to identify El Niño and La Niña years, respectively. Years such as 2009, 2010, and 2018 where both positive and negative anomalies exceeded their respective El Niño and La Niña threshold anomalies were considered neutral, indicating a lack of clear dominance of either condition.
Figure 7. El Niño and La Niña years over the period of study. Average annual SST anomalies were compared with a threshold anomaly of +/−0.5 °C to identify El Niño and La Niña years, respectively. Years such as 2009, 2010, and 2018 where both positive and negative anomalies exceeded their respective El Niño and La Niña threshold anomalies were considered neutral, indicating a lack of clear dominance of either condition.
Remotesensing 15 03632 g007
Remotesensing 15 03632 g008 550
Figure 8. The difference in (a) water cycle intensity, (b) precipitation, and (c) evapotranspiration between the annual averages of 2001–2009 and 2010–2019, and their respective differences without ENSO years (d–f). The cumulative density function (CDF) plots show the cumulative probabilities of change values for water cycle intensity (g), precipitation (h), and evapotranspiration (i) for both scenarios.
Figure 8. The difference in (a) water cycle intensity, (b) precipitation, and (c) evapotranspiration between the annual averages of 2001–2009 and 2010–2019, and their respective differences without ENSO years (d–f). The cumulative density function (CDF) plots show the cumulative probabilities of change values for water cycle intensity (g), precipitation (h), and evapotranspiration (i) for both scenarios.
Remotesensing 15 03632 g008
Table
Table 1. Results of the t-test for the first run (first 17 points for 17 different climate classes). P-values and CIs were generated from the paired t-tests. The CI is a range of acceptable null hypotheses defined by lower and upper confidence limits [53]. Narrower CIs indicate smaller uncertainties in our estimates, and vice versa. The effect size represents the magnitude of the difference between GPM and PRISM datasets at each sample point.
Table 1. Results of the t-test for the first run (first 17 points for 17 different climate classes). P-values and CIs were generated from the paired t-tests. The CI is a range of acceptable null hypotheses defined by lower and upper confidence limits [53]. Narrower CIs indicate smaller uncertainties in our estimates, and vice versa. The effect size represents the magnitude of the difference between GPM and PRISM datasets at each sample point.
IDClass: Descriptionp-ValueCIEffect Size
1Am: Tropical Monsson 0.98−7.96–8.130.0006
2Aw: Tropical Savannah0.95−6.84–7.240.0017
3BWh: Arid, desert, hot0.98−2.01–1.96−0.0006
4BWk: Arid, desert, cold0.90−0.98–1.120.0030
5BSh: Arid, steppe, hot0.99−5.35–5.27−0.0004
6BSk: Arid, steppe, cold0.99−3.14–3.09−0.0004
7Csa: Temperate, dry summer, hot summer0.99−8.44–8.580.0003
8Csb: Temperate, dry summer, warm summer0.93−8.27–9.080.0022
9Cfa: Temperate, no dry season, hot summer0.99−6.05–6.00−0.0002
10Cfb: Temperate, no dry season, warm summer0.82−7.80–9.860.0045
11Dsb: Cold, dry summer, warm summer0.98−3.82–3.930.0007
12Dsc: Cold, dry summer, cold summer0.98−3.40–3.30−0.0010
13Dwa: Cold, dry winter, hot summer0.99−3.50–3.45−0.0004
14Dwb: Cold, dry winter, warm summer0.99−2.79–2.830.0003
15Dfa: Cold, no dry season, hot summer1.00−6.06–6.070.0000
16Dfb: Cold, no dry season, warm summer0.93−5.28–5.760.0024
17Dfc: Cold, no dry season, cold summer0.89−2.24–1.94−0.0046
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zowam, F.J.; Milewski, A.M.; Richards IV, D.F. A Satellite-Based Approach for Quantifying Terrestrial Water Cycle Intensity. Remote Sens. 2023, 15, 3632. https://doi.org/10.3390/rs15143632

AMA Style

Zowam FJ, Milewski AM, Richards IV DF. A Satellite-Based Approach for Quantifying Terrestrial Water Cycle Intensity. Remote Sensing. 2023; 15(14):3632. https://doi.org/10.3390/rs15143632

Chicago/Turabian Style

Zowam, Fabian J., Adam M. Milewski, and David F. Richards IV. 2023. "A Satellite-Based Approach for Quantifying Terrestrial Water Cycle Intensity" Remote Sensing 15, no. 14: 3632. https://doi.org/10.3390/rs15143632

APA Style

Zowam, F. J., Milewski, A. M., & Richards IV, D. F. (2023). A Satellite-Based Approach for Quantifying Terrestrial Water Cycle Intensity. Remote Sensing, 15(14), 3632. https://doi.org/10.3390/rs15143632

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop