Next Article in Journal
Optimizing Policies and Regulations for Zero Routine Gas Flaring and Net Zero
Previous Article in Journal
Trends in Atmospheric Emissions in Central Asian Countries Since 1990 in the Context of Regional Development
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Climate
Volume 13
Issue 9
10.3390/cli13090177
climate-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
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Snowmelt Streamflow Trends over Colorado (U.S.A.) Mountain Watersheds

by
Steven R. Fassnacht
1,2,* and
Anna K. D. Pfohl
1
1
ESS-Watershed Science, Colorado State University, Fort Collins, CO 80523-1476, USA
2
Cooperative Institute for Research in the Atmosphere, Colorado State University, Fort Collins, CO 80523-1375, USA
*
Author to whom correspondence should be addressed.
Climate 2025, 13(9), 177; https://doi.org/10.3390/cli13090177
Submission received: 11 July 2025 / Revised: 14 August 2025 / Accepted: 27 August 2025 / Published: 28 August 2025
(This article belongs to the Special Issue Impacts of Climate Change on Hydrological Processes)
Browse Figures
Versions Notes

Abstract

Streamflow generated from snowmelt is important, and changing, in snow dominated regions of the world. We used a recently developed technique to estimate the start and end of snowmelt streamflow for 39 gauging stations across Colorado and determined the 40-year trends from 1981 to 2020. Most watersheds showed a trend towards an earlier start (34 watersheds) or end (29 watersheds) of snowmelt streamflow, but the mean of the start and end dates showed mixed trends (earlier in 12 watersheds and later in 20). We determined the correlation between these streamflow snowmelt trends and terrain parameters plus trends in canopy cover, winter precipitation, peak snow water equivalent, and melt-period temperature. There were some significant correlations, primarily for total annual streamflow and the timing and volume of the end of snowmelt streamflow contribution to winter precipitation (decreasing), minimum temperature (warming), and slope (negatively). Higher elevation watersheds tend to be steeper, less snow has been observed at higher elevations, and the snowpack is melting sooner. Snowmelt streamflow trends are partially explained by climate trends and watershed characteristics.

1. Introduction

Mountain snowmelt generates water for streams and rivers and is a major water source for a substantially increasing portion of the Earth’s population [1]. Across the semi-arid western United States, a majority of the precipitation falls as snow ([2,3,4]. The timing of the start of the melt season and when the snowmelt enters streams in the high elevation watersheds is crucial for estimating water availability [5,6] this timing has shifted [7,8,9,10,11] due mostly to climate change [12,13,14,15,16].

1.1. Snowmelt Streamflow Timing Metrics

Peak streamflow date is a simple metric of streamflow timing but neglects the remaining data for a given year [17]. Court [17] introduced the half-flow, also known as the Center of Volume date, which is the day when 50% of total annual streamflow has passed a stream gauging station (tQ50). This tQ50 is used extensively as a streamflow timing metric [18,19,20], especially to assess the impacts of climate change [5,13,15,21,22,23]. Other percentages of annual streamflow passage have been used as proxies for the start of snowmelt contribution, i.e., the date of 20% (tQ20) [15] or 25% [22] of streamflow, and the end of snowmelt, i.e., the date of 75% [22] or 80% (tQ80) [15] of streamflow. To highlight the snowmelt period further, Dudley et al. [23] proposed the center of volume for 50% of the streamflow from January to July (tQDudley). However, all these methods are static based on a specific quantity of the total annual or winter [23] streamflow. tQ20, tQ50, and tQ80 are not appropriate indicators specifically of snowmelt timing [24], and sometimes they are biased due to large precipitation events [20,24]. These metrics can be influenced by inter-annual variability in streamflow volume [24] and do not reflect how a changing climate impacts streamflow timing [5].

1.2. Objectives of the Paper

An analytical approach considering the change in streamflow using a departure from baseflow has previously been proposed to identify the start (tQstart) and end of the snowmelt contribution to streamflow (tQend), i.e., considering the characteristics of the hydrograph [5]. Here, we use that method to quantify the start and end of snowmelt from the hydrograph and to determine if and how snowmelt timing is changing. Since snowmelt streamflow characteristics change for a variety of reasons [25,26], we evaluate these changes considering terrain parameters (e.g., elevation, aspect, slope) that are constant over time and variables that change temporally where a trend could be computed (e.g., canopy characteristics, winter precipitation, peak snow water equivalent).
The objectives of this paper are as follows: (1) to apply a recent method for estimating snowmelt timing and volume for streamflow for the Southern Rocky Mountains of Colorado, (2) conduct a trend analysis of different snowmelt timing and volume variables, (3) determine possible explanations for these trends based on time trends in vegetation, winter precipitation, peak SWE, and melt temperature, as well as terrain parameters. We explored 39 smaller (<900 km2) high-elevation watersheds in Colorado. Colorado is a headwater state that includes the Colorado River and its tributaries (Yampa, Gunnison, Uncompahgre, San Miguel, Dolores, Animas, and San Juan) [20], the North and South Platte Rivers, the Arkansas River, and the Rio Grande [15]. The highest mountain peaks reach over 4400 m in elevation and snowcover persists from October through May [2]. Streamflow in these watersheds is snowmelt-dominated, with 60 to 80% of the annual streamflow coming from snow [3,6,15,16].

2. Methodology

2.1. Snowmelt Streamflow Timing and Volume

The tQstart, or timing (date) of the start of the snowmelt contribution to streamflow, was computed as the increase in streamflow from baseflow by a change in slope of at least 10 mm/day [5] (an example is shown in Figure A1). The tQend, or the date of the end of snowmelt contribution, was computed as the decrease in streamflow back to baseflow as the change in slope of at least 17.5 mm/day [5] (Figure A1). These start and end thresholds of 10 and 17.5 mm/day were optimized for all the watersheds used herein by Pfohl and Fassnacht [5]. The average timing of snowmelt (tQstart–end) was computed as the mean of tQstart and tQend (Figure A1). This was used in lieu of tQ50 or tQDudley. We then determined volumes of streamflow, in particular the total annual runoff volume (Q100), the runoff volume that passed the gauge prior to the start and end of snowmelt (Qstart and Qend, respectively), and the runoff volume in between (Qstart–end) (Figure A1).

2.2. Trend Analysis

The rate of change for the trends were calculated as Theil–Sen’s Slope [27,28]. This non-parametric computation is a median of the slopes computed from all pairs (1981 vs. 1982, 1981 vs. 1983, … 2019 vs. 2020) in the time series. The significance was calculated using the Mann–Kendall Test [29,30]. This is a non-parametric accounting of increasing versus decreasing slopes over all pairs in the time series. We used a p-value less than 0.05 as significant and less than 0.10 as moderately significant.

2.3. Correlation and Regression

We computed the cross-correlation between the snowmelt streamflow timing and volume trends amongst the 39 watersheds using the Pearson correlation coefficient. We used individual and multi-variate regressions to explain the trends in snowmelt streamflow timing and streamflow volume. We considered various terrain parameters, in particular the mean incoming winter solar radiation across the watershed, the mean elevation, the mean slope, and the location (latitude and longitude) [31]. We used variables that changed over time and computed their trend. We also computed the cross-correlation between the terrain parameters and trend variables.
To address changes within the watershed from land use, beetle-kill, or wildfires, we used Normalized Difference Vegetation Index (NDVI) data from the U.S. Geological Survey [32]. These data start in July of 1989 but would still capture major changes in vegetation because major beetle-kill [33] and fires [34] did not occur in the Southern Rocky Mountains until the late 1990s and early 2000s.
Previous studies that examined trends in timing of streamflow snowmelt have primarily relied on climactic indices to explain their observations [13,15,16,23]. We used precipitation data from the Precipitation-Elevation Regressions on Independent Slopes Model (PRISM) dataset [35] to evaluate winter precipitation (October through March), starting in 1982.
We used snow water equivalent (SWE) data from the high elevation Snow Telemetry (SNOTEL) stations closest to each streamflow gauge to identify annual peak SWE. The nearest SNOTEL station was on average 17 km away from the streamflow gauging station. The closest SNOTEL station was less than one kilometer away (Joe Wright), 13 were withing 10 km, and 31 within 20 km. The Conejos River (54 km) and Rock Creek (152 km) were not close to SNOTEL stations (Figure 1).
Temperature has had a strong correlation to trends in the specified percentage of streamflow that has passed [15,23]. Temperature sensors were installed at the SNOTEL station starting in the 1980s, so the time series is not complete. Further, there is a significant inhomogeneity in the middle (from 1998 to 2006) of the SNOTEL temperature time series [36]. Two approaches have been used to correct this inhomogeneity [37,38], including SWE modeling to identify the optimal correction approach [38] for the longer-term SNOTEL across the state of Colorado. Here, the average temperature for the melt months (March through May) was used from the Ma et al. [38] time series correction. SNOTEL stations also report the daily maximum and minimum temperature time series, and these data were considered.
We calculated the correlation between the trends in snowmelt streamflow timing and streamflow volume and individual terrain parameters and trend variables using the Pearson correlation coefficient. We evaluated multi-variate linear regressions using all parameters (constant over time) and variable trends. The most highly correlated parameters/variables from the individual correlation were then considered to reduce the number of independent variables in the regressions.

3. Study Domain

We examined 40 years of streamflow (1981 through 2020) for 39 United States Geological Survey (USGS) gauging stations across the Southern Rocky Mountains of Colorado, each with at least 30 years of record (Figure 1; Table A1). Streamflow data were obtained from the National Water Information System [39]. All were headwater streams gauged at an elevation higher than 2000 m above sea level (Figure 1; Table A1). The mean basin elevation varied from 2494 to 3644 m.a.s.l. (Figure 2a), with the mean April clear sky solar radiation loading ranging from 1407 to 1760 Wh/m2 (Figure 2b). The basins had a mean slope from 17 to 26° (Figure 2c) and ranged in size from 4 to 878 km2 (Figure 2d). The stations are summarized in Pfohl and Fassnacht [5], and Table A1. The SWE data were obtained from the Natural Resources Conservation Service [40]. The temperature data were adjusted/taken from Ma et al. [38].

4. Results

The canopy density, as NDVI, increased for 37 of the 39 watersheds (Figure 2e), significantly at six watersheds (moderately significant at one). Both winter precipitation (Figure 2f) and the adjacent SNOTEL peak SWE (Figure 2g) were decreasing for all but one watershed. Seven of the decreasing trends in winter precipitation were significant. Only two were significant for peak SWE. Melt temperatures increased for all SNOTEL stations, with most (34) being significant (Figure 2h). Maximum temperatures increased less than the mean melt temperatures (22 trended cooler), while minimum temperatures increased more (36 significantly) (Figure A2). Temperature trends were highly correlated (Appendix C; Figure A2).
The snowmelt characteristics of streamflow have changed across most of the watersheds over the study period (Figure 3). Most (34) see a trend of an earlier start of the snowmelt streamflow, while only three are later (Figure 3a), with 38% of the trends being significant (and five being moderately significant). Twenty-nine watersheds see an earlier end of the snowmelt contribution and seven are later (Figure 3b), with about 40% being significant (9 watersheds) or moderately significant (7 watersheds). The change in timing of the peak, denoted tQstart–end, is mixed, being earlier at 12 watersheds and later at 20 (1 significantly in each direction; Figure 3c). For 27 watersheds, both tQstart and tQend trends were earlier while for only one watershed both became later. Earlier trends were observed for all three metrics in nine watersheds.
Trends for the volume of streamflow that has passed the gauge were more mixed (Figure 3d–g), i.e., both increasing and decreasing. Total annual streamflow (Q100) increased in 23 (2 significantly and 1 moderately significant) watersheds while it decreased at the (16) others (1 significantly and 3 moderately significant) (Figure 3d). Qstart changed by the smallest amount (Figure 3e). Trends for Qend (Figure 3f) and Qstart–end (Figure 3g) were similar (16 with more and 23 with less), with 35 having the same sign (15 watersheds where both increased streamflow and 20 where both decreased). The trend was in the same direction for Qstart and Qend at 20 watersheds (7 less, 13 more), and for 18 watersheds for all four metrics (6 less, 12 more). Trends were in the same direction and significant for three watersheds: Joe Wright Creek (more streamflow), Vasquez Creek (more streamflow), and Conejos River (less streamflow). The trends in Q100, Qend and Qstart–end illustrated a latitudinal pattern with most stations north of 39.7o (Figure 1) increasing in streamflow volume and most south decreasing (Figure 3d,f,g).
Except for Qstart, snowmelt streamflow trends are more correlated to winter precipitation or peak SWE than NDVI (Figure 4). Winter precipitation is more correlated with NDVI (R = 0.43) than with peak SWE (R = 0.29) (Table A2). Snowmelt streamflow trends are weakly correlated to mean solar radiation and elevation. Mean basin slope is significantly correlated (negatively) to trends in tQend, Q100, Qend and Qstart–end. Latitude is positively correlated to all streamflow characteristics (3 significantly) while longitude is less correlated (except tQend and tQstart–end) than latitude.
A linear multi-variate regression identified some significant correlations between trends in streamflow timing and volume metrics with terrain parameters (Table 1), but limited with the trends, i.e., NDVI, winter precipitation, peak SWE, or melt temperature. Using all parameters and variables, the strongest correlations were for Qend and Qstart–end including all variables (R2 of 0.52 and 0.46, respectively) (Table 1). Reducing the number of independent variables made the regressions for tQend, Q100, Qend, and Qstart–end significant (Table 1). The variance explained decreased, as shown by R2, but the individual regression variables were more consistently significant. Slope was negatively correlated, and latitude was positively correlated with snowmelt streamflow timing and volume trends. Trends in the start of snowmelt streamflow, i.e., tQstart and Qstart, as well as tQstart–end, were poorly explained by the regression, not significant, and R2 was mostly less than 0.1 when a subset of the parameters/variables was used (Table 1).
Maximum temperature was correlated to Qend (moderately significant) and Qstart–end (Figure A3). In addition to Qend and Qstart–end, minimum temperature was also correlated with Q100 (moderately significant) and tQstart (Figure A3). Since minimum temperature showed significant correlation with three of the snowmelt metrics, it was used in lieu of the mean melt period temperature in the multi-variate regression (Table A4), using variables/parameters that were individually correlated (Figure 4). In these regressions, the minimum temperature was always significant in the regression for tQstart. Reducing the number of regression variables, winter precipitation was moderately significant for tQend and Q100, and significant for Qstart, Qend and Qstart–end, when including the minimum temperature (Table A4).

5. Discussion

Snowmelt-driven streamflow is occurring earlier for most basins across the study domain (tQstart in Figure 2e and the tQend in Figure 2f), as has also been seen using time-constant streamflow metrics, i.e., tQ20, and tQ80 [7,8,9,10,11,13,15,16,21,22]. However, the trends in the tQstart–end or mean of start and end timing were mixed (Figure 3, Table 1), reflecting tQstart and tQend trends (Table A3) and their differences (Figure 3a versus Figure 3b). The time-constant metrics that are meant to represent the middle of the snowmelt streamflow peak, i.e., tQ50 [17] or tQDudley [23], are getting earlier [15]. These time-constant streamflow metrics are relative to the water year, while tQstart–end is relative to the characteristics of the hydrograph [5], as recommended by Whitfield [24].
The change in slope for identifying the start and end of melt (here 10 and 17.5 mm/day) can influence the estimation of the timing metrics, possibly for other climate regions. For high-elevation Colorado streamflow gauges, the values used herein were shown to be acceptable minima [5]. These should be assessed for other climates.
The trends in snowmelt streamflow can be partially explained by temporal trends or watershed parameters. In the best case, only about 50% of the variance is explained for only Qend and Qstart–end using all variables/parameters (Table 1 and Table A4). The other regressions with all variables/parameters were not significant and the R2 value was less than 0.4, yet reducing the number of regression variables made more regressions significant (tQend and Q100) while further reducing R2 (Table 1). Winter precipitation combined with minimum temperature and basin area yielded weak (R2 < 0.3) but two moderate and two significant correlations (Table A4). Winter precipitation is decreasing for almost all watersheds (Figure 2f), with a larger decline in the south [41], as these southern stations have seen a large decline in snowfall since about 2000 [34,42]. Thus, latitude is a partial surrogate for winter precipitation (R = 0.59 in Table A2), including for Qend and Qstart–end (Table 1).
Steeper sloped watersheds are seeing a decreasing trend in total streamflow (Q100) and the end of snowmelt streamflow (tQend, Qend), and Qstart–end (Figure A4), while more gently sloped watersheds are seeing an increasing trend (Figure 4 and Table 1), i.e., gentler sloped watersheds are melting out later [43]. Higher elevation watersheds tend to be steeper (R2 of 0.42 without the Roaring Fork) (Figure 2a,c; Table A2). There is some correlation between elevation and peak SWE and thus the amount of snow available to melt (Table A2), with SWE decreasing at higher elevation SNOTEL stations and increasing at lower elevation stations, at least in northern Colorado [44]. Less snow means sooner melt [14]. Mean basin slope only explains some changes, since slopes usually vary substantially across mountain watersheds and the mean slope may not represent watershed processes well [45]. Aspect plays an important role in snowmelt [46,47], and the combination of slope and aspect must be considered over the watershed [48]. Steeper slopes are also correlated with a larger difference in the trend of the start and in the trend of the end of snowmelt streamflow (Figure 3a versus Figure 3b), highlighting that even as an average across a watershed, slope is relevant.
This correlation is also seen between NDVI and winter precipitation (R = 0.43 in Table A2) and thus NDVI and latitude (R = 0.40). Peak SWE trends were correlated with elevation (Table A2) [46], but here (Figure 2g) were less correlated with winter precipitation (R = 0.29 in Table A2). Peak SWE was extracted from SNOTEL station data [40], and these may not be representative of the watershed [49]. These are mostly small watersheds (Figure 2d), so current SWE products [50] may not have the necessary resolution to assess changes.
There is a lot of spatial variability across each watershed that dictates snowmelt and snowmelt changes, and these are not captured using basin-average values (Figure 2). More insight is necessary to understand changes in processes that may dictate variations in streamflow timing. This would include the interplay between slope and aspect [46,47]. Snowpack modeling would be useful [51] to assess spatial variability at relevant scales [52].
Temperature increases are a major indication and result of climate change [12,13,14,15,16]. Here, minimum temperature trends over the melt months (March to May) were more relevant than mean temperature (Figure A3) to changes in snowmelt-driven streamflow (Table 1 versus Table A4). The minimum temperatures are warming more than mean temperatures (Figure A2) [38]. More than half of the watersheds are seeing cooling maximum temperatures (Figure A2). Colorado has a continental climate where snowmelt is driven by solar radiation; daily cold temperatures, i.e., minimum temperature, are indicative of the timing of snowmelt, and thus snowmelt streamflow. Spatial temperature data, instead of the point SNOTEL station (Figure 2h) may be more informative of trends across each watershed.
There are some spatial patterns in the changes in snowmelt-driven streamflow, specifically latitude, and to a lesser degree longitude (Figure 3 and Figure 4). Others [15] have used the Regional Kendall test [53] to evaluate trends and their significance across an area; due to the limited spatial patterns observed here (Figure 3), it is recommended to use the Mann–Kendall test [29,30] and Theil–Sen slope [27,28] on individual stations. Using the Regional Kendall test can produce trends that are smaller in magnitude than observed trends at individual sites [54].
The approach used herein [5] identifies the start and end of the snowmelt contribution for snowmelt dominated systems as an improvement to the traditional statistics approaches, such as tQ20, tQ50, and tQ80 [15,17]. This information may be helpful for water forecasters and managers making decisions about water storage and reservoirs for the future [55], especially if the timing of peak streamflow is incorporated [56,57]. It does not specifically identify baseflow, although it has been used for that purpose [58]. Baseflow separation techniques could be used to identify when direct or non-baseflow starts to contribute to streamflow. This could be applied to a snowmelt dominated system to determine when snowmelt streamflow started and ended. There are analytical approaches [59] using only streamflow data. Snowmelt is often separated from baseflow using isotopes [60]. However, such measurements are labor- and cost-intensive. Specific conductance is measured as an in situ water quality variable in a few locations and has been used with streamflow to separate baseflow from non-baseflow [61]. There are now some time series long enough to examine trends.
The temperature data have been adjusted [38] to address the inconsistency in the SNOTEL temperature time series for the western U.S. [36,37]. These data are used for many of the spatial products (e.g., PRISM [35]) so the SNOTEL inconsistency needs to be examined further. The SNOTEL time series has been consistent for the past 20 or more years [40], so future investigations should focus on the last two decades. This time period would ensure that a complete record of NDVI data [32] are used.
Change in land cover type and the nature of the canopy will influence snowmelt and thus streamflow timing [62,63]. Here we used NDVI [37] to assess changes in canopy due to the length of the time series (back to 1989) and the 30-m resolution from Landsat data. This time period includes the late 1990s and early 2000s forest changes from beetle-kill [33] and fires [34]. Further, the resolution is fine enough to identify such forest changes, as compared to the longer time series’ AVHRR-based 8-km NDVI [64]. However, NDVI has a limited capacity to capture forest structural changes [65,66], which influences snowmelt [47,67]. Other datasets may be more useful than NDVI, such as OpenET [68].

6. Conclusions

Applying a recently developed method to estimate the timing and volume of snowmelt streamflow, we found that the onset and end of snowmelt-driven streamflow occurred earlier in almost all of the watersheds. The total annual streamflow increased at a majority of the watersheds, as did the volume before the onset of snowmelt and the volume at the end of snowmelt. These trends were most correlated with winter precipitation, minimum temperature during the melt month, and slope (negatively). There was correlation with peak SWE for total runoff volume and the volume at the end of snowmelt; these two variables are highly correlated. Due to climatic differences across the domain, in particular drying trends in southern Colorado, winter precipitation was correlated with latitude. Multi-variate regressions illustrated the more highly correlated variables.

Author Contributions

Conceptualization, S.R.F. and A.K.D.P.; methodology, S.R.F. and A.K.D.P.; software, S.R.F. and A.K.D.P.; formal analysis, S.R.F. and A.K.D.P.; investigation, S.R.F. and A.K.D.P.; resources, S.R.F.; writing—original draft preparation, S.R.F. and A.K.D.P.; writing—review and editing, S.R.F. and A.K.D.P.; visualization, S.R.F. and A.K.D.P.; supervision, S.R.F.; funding acquisition, S.R.F. All authors have read and agreed to the published version of the manuscript.

Funding

This research was partially funded by The Leona M. and Harry B. Helmsley Charitable Trust for the Vertically Integrated Project Program (PI, Georgia Institute of Technology).

Data Availability Statement

The data used in this paper are available online. The streamflow data are from [30] the U.S. Geological Survey National Water Dashboard <https://dashboard.waterdata.usgs.gov/> (last accessed on 13 November 2024). The DEM and NDVI data are from [37] the EarthExplorer <https://earthexplorer.usgs.gov/> (last accessed on 13 November 2024). The winter precipitation data are from [35] the PRISM Climate Group <https://prism.oregonstate.edu/> (last accessed on 13 November 2024).

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A. Sample Hydrograph

The appendix presents a sample daily hydrograph, focusing on the spring and summer time period (April through September in Figure A1) to demonstrate the timing of the start (tQstart) and end (tQend) of the snowmelt contribution to streamflow, as per the method of Pfohl and Fassnacht, 2023 [5]. The average of start and end (tQstart–end) of the snowmelt contribution to streamflow is also shown as “start end.” The traditional metrics (start of snowmelt streamflow = tQ20, center of mass or tQ50, end of snowmelt streamflow = tQ80) [15] are also presented (Figure A1).
Climate 13 00177 g0a1
Figure A1. Sample daily (blue) and cumulative (brown) hydrograph for the Michigan River gauging stations in northern Colorado for April through September 1993, illustrating the timing of the start (tQstart as a dashed vertical line with single dot in blue) and end (tQend as a dashed vertical line with double dot in orange) of the snowmelt contribution to streamflow, and the timing of the average between tQstart and tQend (tQstart–end as a dotted vertical line in blue). The previous timing metrics [15] are presented as black lines using the same symbology as the above: start or tQ20 as a dashed single dot line, center of mass or tQ50 as a dotted line, and end or tQ80 as a dashed double dot line. The cumulative runoff is the sum of the daily streamflow divided by the area of the basin to yield a depth of water.
Figure A1. Sample daily (blue) and cumulative (brown) hydrograph for the Michigan River gauging stations in northern Colorado for April through September 1993, illustrating the timing of the start (tQstart as a dashed vertical line with single dot in blue) and end (tQend as a dashed vertical line with double dot in orange) of the snowmelt contribution to streamflow, and the timing of the average between tQstart and tQend (tQstart–end as a dotted vertical line in blue). The previous timing metrics [15] are presented as black lines using the same symbology as the above: start or tQ20 as a dashed single dot line, center of mass or tQ50 as a dotted line, and end or tQ80 as a dashed double dot line. The cumulative runoff is the sum of the daily streamflow divided by the area of the basin to yield a depth of water.
Climate 13 00177 g0a1

Appendix B. Station Summary

This appendix presents the location and areas for each study watershed (Table A1).
Table A1. Name, USGS station number, latitude, longitude and gauge elevation, and basin area for the 39 gauges presented in Figure 1.
Table A1. Name, USGS station number, latitude, longitude and gauge elevation, and basin area for the 39 gauges presented in Figure 1.
NameNumberLatitude (°)Longitude (°)Gauge Elevation (m)Basin Area (km2)
Joe Wright Creek0674609540.540−105.88330458
Michigan River0661480040.496−105.86531674
Colorado River0901050040.326−105.8572667165
Cabin Creek0903210039.986−105.745291413
Ranch Creek0903200039.950−105.766264052
Vasquez Creek0902500039.920−105.785267372
St. Louis Creek0902650039.910−105.878273785
Fraser River0902200039.846−105.752290227
S Fork of Williams0903590039.801−106.026272871
Darling Creek0903580039.797−106.026272523
Piney River0905950039.796−106.5742217219
Williams Fork0903550039.779−105.928298742
Bobtail Creek0903490039.760−105.906317915
East Meadow Creek0905880039.732−106.42728829
Dickson Creek0905861039.704−106.45728189
Freeman Creek0905870039.698−106.44628458
Red Sandstone Creek0906640039.683−106.401280819
Booth Creek0906620039.648−106.323253716
Middle Creek0906630039.646−106.382249915
Pitkin Creek0906615039.644−106.303259814
Bighorn Creek0906610039.640−106.293262912
Gore Creek0906550039.626−106.278262138
Black Gore Creek0906600039.596−106.265278932
Keystone Gulch0904770039.594−105.973285024
Tenmile Creek0905010039.575−106.1112774239
Turkey Creek0906340039.523−106.337271861
Wearyman Creek0906320039.522−106.324282925
Eagle River0906300039.508−106.3672638182
Blue River0904660039.456−106.0322749319
Homestake Creek0906400039.406−106.433280492
Missouri Creek0906390039.390−106.470304217
Crystal River0908160039.233−107.2282105433
Halfmoon Creek0708300039.172−106.389299661
Roaring Fork River0907330039.141−106.7742475196
Rock Creek0710594538.707−104.847200018
Lake Fork0912450038.299−107.2302386878
Uncompahgre River0914620038.184−107.7462096386
Vallecito Creek0935290037.478−107.5442410188
Conejos River0824500037.300−105.7473007104

Appendix C. Daily Maximum, Mean, and Minimum Temperature Time Series

The trends in the melt temperature derived from the mean daily time series are well correlated with the maximum (R = 0.75) and minimum (R = 0.80) daily temperatures (Figure A2). The correlation between the timing and volume trends across the 39 watersheds and the three temperature time series trends is presented in Figure A3.
Climate 13 00177 g0a2
Figure A2. Comparison of station trends in maximum and minimum daily temperature versus mean temperature for the melt period (March through May). Moderate significance and significance for maximum and minimum daily temperature trends are shown by dashed and solid symbols, respectively.
Figure A2. Comparison of station trends in maximum and minimum daily temperature versus mean temperature for the melt period (March through May). Moderate significance and significance for maximum and minimum daily temperature trends are shown by dashed and solid symbols, respectively.
Climate 13 00177 g0a2
Climate 13 00177 g0a3
Figure A3. Correlation between snowmelt timing and volume streamflow trends and maximum, mean, and minimum temperature from the Pearson correlation coefficient.
Figure A3. Correlation between snowmelt timing and volume streamflow trends and maximum, mean, and minimum temperature from the Pearson correlation coefficient.
Climate 13 00177 g0a3

Appendix D. Cross-Correlation of Trends, Parameters, and Variables

This appendix presents the cross-correlation between the variables/parameters used in the regression (Table A2) and between the timing and volume trends across the 39 watersheds (Table A3). The cross-correlation is represented by the Pearson correlation coefficient (R).
Table A2. Independent cross-correlation between time trend variables (NDVI, winter precipitation, peak SWE, melt-period temperature) and parameters (basin mean solar radiation, basin mean elevation, basin mean slope, area, latitude, longitude) from the Pearson correlation coefficient.
Table A2. Independent cross-correlation between time trend variables (NDVI, winter precipitation, peak SWE, melt-period temperature) and parameters (basin mean solar radiation, basin mean elevation, basin mean slope, area, latitude, longitude) from the Pearson correlation coefficient.
Winter PPeak SWEMelt TempSolar Rad.ElevationSlopeAreaLatitudeLongitude
NDVI0.43−0.130.170.04−0.06−0.05−0.060.400.13
Winter P 0.290.060.310.17−0.14−0.300.590.46
Peak SWE −0.24−0.100.420.09−0.040.250.02
Melt Temp 0.130.040.0100.280.02
Solar Rad. 0.14−0.01−0.450.330.36
Elevation 0.44−0.030.190.20
Slope 0.13−0.01−0.26
Area −0.46−0.57
Latitude 0.45
Table A3. Dependent cross-correlation between snowmelt timing and volume streamflow trends from the Pearson correlation coefficient.
Table A3. Dependent cross-correlation between snowmelt timing and volume streamflow trends from the Pearson correlation coefficient.
tQendtQstart–endQ100QstartQendQstart–end
tQstart0.12−0.560.290.340.370.37
tQend 0.660.200.010.370.34
tQstart–end −0.06−0.21−0.005−0.03
Q100 0.120.890.82
Qstart 0.08−0.09
Qend 0.94
Table A4. Linear multi-variate regression results, as per Table 1 for using minimum temperature for (a) regression with all variables, (b) regression with winter precipitation, minimum temperature, elevation and area, and (c) regression with winter precipitation, minimum temperature and area. The moderately significant correlations (p < 0.1) are italicized and denoted with a plus sign (+); the significant correlations (p < 0.05) are in bold and denoted with an asterisk (*).
Table A4. Linear multi-variate regression results, as per Table 1 for using minimum temperature for (a) regression with all variables, (b) regression with winter precipitation, minimum temperature, elevation and area, and (c) regression with winter precipitation, minimum temperature and area. The moderately significant correlations (p < 0.1) are italicized and denoted with a plus sign (+); the significant correlations (p < 0.05) are in bold and denoted with an asterisk (*).
VariableR2Sign. FInterceptNDVIWinter PPeak SWEMin. TempSolar Rad.Elev.SlopeAreaLat.Long.
(a) Regression with all variables/parameters
tQstart0.240.56627.1−0.2160.260−0.012−4.61 * −0.0010.002−0.1340.000−0.0980.247
tQend0.330.22922.6−2.490.767−0.0100.83−0.015 * 0.000−0.222−0.0010.5880.168
tQstart–end0.160.86154.7−1.940.1100.0144.28−0.010−0.002−0.0620.0000.4840.474
Q1000.380.119−1794−11.92.980.482−31.5−0.0520.014−5.29 * 0.03923.3−9.82
Qstart0.370.148−1215.501.05−0.024−7.330.0090.006−0.2570.012 +−0.724−1.18
Qend0.530.008 * −1029−26.347.790.397−27.2−0.068−0.016−3.17 *−0.00420.8 +−4.34
Qstart–end0.480.025 * −748−27.274.570.379−36.3−0.055−0.023−3.13−0.04921.8 +−1.39
(b) Regression with winter precipitation, minimum temperature, elevation and area
tQstart0.170.156−1.40 0.276 −4.37 * 0.001 0.000
tQend0.130.3048.16 0.705 + −0.444 −0.003 0.000
tQstart–end0.060.6736.15 0.138 2.97 −0.002 0.001
Q1000.150.242103 10.8 + −66.5 −0.015 0.027
Qstart0.280.022 * −0.263 1.25 * −6.16 0.002 0.013 *
Qend0.310.011 * 131 13.0 * −56.7 + −0.026 −0.015
Qstart–end0.310.011 * 155 10.4 * −64.0 + −0.032 −0.069 +
(c) Regression with winter precipitation, minimum temperature and area
tQstart0.170.086 +0.500 0.291 −4.46 * 0.000
tQend0.090.345−0.857 0.631 + −0.039 0.000
tQstart–end0.040.704−1.47 0.075 3.31 0.001
Q1000.140.14653.4 * 10.4 + −64.3 0.027
Qstart0.270.010 * 5.53 * 1.30 * −6.42 0.013 *
Qend0.290.007 * 42.4 * 12.3 * −52.8 −0.016
Qstart–end0.190.054 +0.383 * 0.064 * 0.046 0.000

Appendix E. Slope Versus Correlated Snowmelt Trends

This appendix presents the correlation between the mean basin slope and timing and volume trends (Figure A4), specifically for (a) tQend, (b) Q100, (c) Qend, and (d) Qstart–end. The correlation is represented by the coefficient of determination (R2) and includes the improved correlation with the removal of one watershed.
Climate 13 00177 g0a4
Figure A4. Correlation between slope and snowmelt timing and volume streamflow trends, specifically (a) tQend, (b) Q100, (c) Qend, and (d) Qstart–end.
Figure A4. Correlation between slope and snowmelt timing and volume streamflow trends, specifically (a) tQend, (b) Q100, (c) Qend, and (d) Qstart–end.
Climate 13 00177 g0a4

References

  1. Viviroli, D.; Kummu, M.; Meybeck, M.; Kallio, M.; Wada, Y. Increasing dependence of lowland populations on mountain water resources. Nat. Sustain. 2020, 3, 917–928. [Google Scholar] [CrossRef]
  2. Serreze, M.C.; Clark, M.P.; Armstrong, R.L. Characteristics of the western United States snowpack from snowpack telemetry (SNOTEL) data. Water Resour. Res. 1999, 35, 2145–2160. [Google Scholar] [CrossRef]
  3. Doesken, N.J.; Pielke, R.A.; Bliss, O.A.P. Climate of Colorado; Climatography of the United States No. 60; Colorado Climate Center, Atmospheric Science Department, Colorado State University: Fort Collins, CO, USA, 2003; Available online: https://hdl.handle.net/10217/236290 (accessed on 14 August 2025).
  4. Ikeda, K.; Rasmussen, R.; Liu, C.; Newman, A.; Chen, F.; Barlage, M.; Gutmann, E.; Dudhia, J.; Dai, A.; Luce, C.; et al. Snowfall and snowpack in the Western U.S. as captured by convection permitting climate simulations: Current climate and pseudo global warming future climate. Clim. Dyn. 2021, 57, 2191–2215. [Google Scholar] [CrossRef]
  5. Pfohl, A.K.D.; Fassnacht, S.R. Evaluating Methods of Streamflow Timing to Approximate Snowmelt Contribution in High-Elevation Mountain Watersheds. Hydrology 2023, 10, 75. [Google Scholar] [CrossRef]
  6. Hammond, J.C.; Kampf, S.K. Subannual streamflow responses to rainfall and snowmelt inputs in snow-dominated watersheds of the western United States. Water Resour. Res. 2020, 56, e2019WR026132. [Google Scholar] [CrossRef]
  7. Cayan, D.R.; Kammerdiener, S.A.; Dettinger, M.D.; Caprio, J.M.; Peterson, D.H. Changes in the onset of spring in the western United States. Bull. Am. Meteorol. Soc. 2001, 82, 399–415. [Google Scholar] [CrossRef]
  8. Viviroli, D.; Archer, D.R.; Buytaert, W.; Fowler, H.J.; Greenwood, G.B.; Hamlet, A.F.; Huang, Y.; Koboltschnig, G.; Litaor, M.I.; López-Moreno, J.I.; et al. Climate change and mountain water resources: Overview and recommendations for research, management and policy. Hydrol. Earth Syst. Sci. 2011, 15, 471–504. [Google Scholar] [CrossRef]
  9. Schlaepfer, D.R.; Lauenforth, W.K.; Bradford, J.B. Consequences of declining snow accumulation for water balance of mid-latitude dry regions. Glob. Change Biol. 2012, 18, 1988–1997. [Google Scholar] [CrossRef]
  10. Barnhart, T.B.; Tague, C.L.; Molotch, N.P. The counteracting effects of snowmelt rate and timing on runoff. Water Resour. Res. 2020, 56, e2019WR026634. [Google Scholar] [CrossRef]
  11. Burn, D.H.; Whitfield, P.H. Shifting cold regions streamflow regimes in North America affect flood frequency analysis. Hydrol. Sci. J. 2024, 70, 51–70. [Google Scholar] [CrossRef]
  12. Leung, L.R.; Qian, Y.; Bian, X.; Washington, W.M.; Han, J.; Roads, J.O. Mid-century ensemble regional climate change scenarios for the western United States. Clim. Change 2004, 62, 75–113. [Google Scholar] [CrossRef]
  13. Stewart, I.T.; Cayan, D.R.; Dettinger, M.D. Changes toward earlier streamflow timing across Western North America. J. Clim. 2005, 18, 1136–1155. [Google Scholar] [CrossRef]
  14. Stewart, I.T. Changes in snowpack and snowmelt runoff for key mountain regions. Hydrol. Process. 2009, 23, 78–94. [Google Scholar] [CrossRef]
  15. Clow, D.W. Changes in timing of snowmelt and streamflow in Colorado: A response to recent warming. J. Clim. 2010, 23, 2293–2306. [Google Scholar] [CrossRef]
  16. Harpold, A.; Brooks, P.; Rajagopal, S.; Heidbuchel, I.; Jardine, A.; Stielstra, C. Changes in snowpack accumulation and ablation in the intermountain west. Water Resour. Res. 2012, 48, W11501. [Google Scholar] [CrossRef]
  17. Court, A. Measures of Streamflow Timing. J. Geophys. Res. 1962, 67, 4335–4339. [Google Scholar] [CrossRef]
  18. Johnson, F.A. Comments on paper by Arnold Court, ‘Measures of streamflow timing’. J. Geophys. Res. 1964, 69, 3525–3527. [Google Scholar] [CrossRef]
  19. Satterlund, D.R.; Eschner, A.R. Land use, snow, and streamflow regimen in central New York. Water Resour. Res. 1965, 1, 397–405. [Google Scholar] [CrossRef]
  20. Fassnacht, S.R. Upper versus lower Colorado River sub-basin streamflow: Characteristics, runoff estimation, and model simulation. Hydrol. Process. 2006, 20, 2187–2205. [Google Scholar] [CrossRef]
  21. Stewart, I.T.; Cayan, D.R.; Dettinger, M.D. Changes in snowmelt runoff timing in western North America under a ‘Business as Usual’ climate change scenario. Clim. Change 2024, 62, 217–232. [Google Scholar] [CrossRef]
  22. Rauscher, S.A.; Pal, J.S.; Diffenbaugh, N.S.; Benedetti, M.M. Future changes in snowmelt-driven runoff timing over the western US. Geophys. Res. Lett. 2008, 35, L16703. [Google Scholar] [CrossRef]
  23. Dudley, R.W.; Hodgkins, G.A.; McHale, M.R.; Kolian, M.J.; Renard, B. Trends in snowmelt-related streamflow timing in the conterminous United States. J. Hydrol. 2017, 547, 208–221. [Google Scholar] [CrossRef]
  24. Whitfield, P.H. Is ‘Centre of Volume’ a robust indicator of changes in snowmelt timing? Hydrol. Process. 2013, 27, 2691–2698. [Google Scholar] [CrossRef]
  25. Al Sawaf, M.B.; Kawanisi, K. Assessment of mountain river streamflow patterns and flood events using information and complexity measures. J. Hydrol. 2020, 590, 125508. [Google Scholar] [CrossRef]
  26. Gordon, B.L.; Brooks, P.D.; Krogh, S.A.; Boisrame, G.F.S.; Carroll, R.W.H.; McNamara, J.P.; Harpold, A.A. Why does snowmelt-driven streamflow response to warming vary? A data-driven review and predictive framework. Environ. Res. Lett. 2022, 17, 053004. [Google Scholar] [CrossRef]
  27. Theil, H. A rank-invariant method of linear and polynomial regression analysis. Proc. R. Neth. Acad. Sci. 1950, 53, 386–392. [Google Scholar]
  28. Sen, P.K. Estimates of the Regression Coefficient Based on Kendall’s Tau. Am. Stat. Assoc. J. 1968, 63, 1379–1389. [Google Scholar] [CrossRef]
  29. Mann, H.B. Nonparametric Tests Against Trends. Econometrica 1945, 13, 245–259. [Google Scholar] [CrossRef]
  30. Kendall, M.; Gibbons, J.D. Rank Correlation Methods, 5th ed.; Edward Arnold: London, UK, 1990. [Google Scholar]
  31. Meromy, L.; Molotch, N.P.; Link, T.E.; Fassnacht, S.R.; Rice, R. Subgrid variability of snow water equivalent at operational snow stations in the western USA. Hydrol. Process. 2013, 27, 2382–2400. [Google Scholar] [CrossRef]
  32. U.S. Geological Survey. EarthExplorer. U.S. Department of Interior. Available online: https://earthexplorer.usgs.gov/ (accessed on 13 November 2024).
  33. Wehner, C.E.; Stednick, J.D. Effects of mountain pine beetle-killed forests on source water contributions to streamflow in headwater streams of the Colorado Rocky Mountains. Front. Earth Sci. 2017, 11, 496–504. [Google Scholar] [CrossRef]
  34. Kampf, S.K.; McGrath, D.; Sears, M.G.; Fassnacht, S.R.; Kiewiet, L.; Hammond, J.C. Increasing wildfire impacts on snowpack in the western US. Proc. Natl. Acad. Sci. USA 2022, 119, e2200333119. [Google Scholar] [CrossRef] [PubMed]
  35. PRISM Climate Group. PRISM Climate Data. Northwest Alliance for Computational Science & Engineering; Oregon State University, 2024. Available online: https://prism.oregonstate.edu/ (accessed on 13 November 2024).
  36. Julander, R.P.; Curtis, J.; Beard, A. The SNOTEL temperature dataset. Mt. Views Newsl. 2007, 1, 4–7. Available online: https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=cdb460f0a71e023a4e9be7ac16e2abbe8679416a#page=8 (accessed on 14 August 2025).
  37. Oyler, J.W.; Dobrowski, S.Z.; Ballantyne, A.P.; Klene, A.E.; Running, S.W. Artificial amplification of warming trends across the mountains of the western United States. Geophys. Res. Lett. 2015, 42, 153–161. [Google Scholar] [CrossRef]
  38. Ma, C.; Fassnacht, S.R.; Kampf, S.K. How temperature sensor change affects warming trends and modeling: An evaluation across the state of Colorado. Water Resour. Res. 2019, 55, 9748–9764. [Google Scholar] [CrossRef]
  39. United States Geological Survey. National Water Information System: Web Interface. United States Department of the Interior USGS Water Data for USA. Available online: http://waterdata.usgs.gov/nwis/ (accessed on 12 November 2024).
  40. NRCS. Natural Resources Conservation Service. U.S. Department of Agriculture. Available online: https://www.nrcs.usda.gov/ (accessed on 27 November 2024).
  41. Von Thaden, B.C. Spatial Accumulation Patterns of Snow Water Equivalent in the Southern Rocky Mountains. Master’s Thesis, Colorado State University, Fort Collins, CO, USA, 2016. Available online: http://hdl.handle.net/10217/173515 (accessed on 14 August 2025).
  42. Whitfield, P.H.; Shook, K.R. Changes to rainfall, snowfall, and runoff events during the autumn–winter transition in the Rocky Mountains of North America. Can. Water Resour. J./Rev. Can. Ressour. Hydr. 2020, 45, 28–42. [Google Scholar] [CrossRef]
  43. Kilmister, I.F.; Campbell, P.; Dee, M. Till the End—Chapter 9. In Bad Magic; UDR GmbH: Husum, Germany, 2015. [Google Scholar]
  44. Fassnacht, S.R.; Patterson, G.G.; Venable, N.B.H.; Cherry, M.L.; Pfohl, A.K.D.; Sanow, J.E.; Tedesche, M.E. How Do We Define Climate Change? Considering the Temporal Resolution of Niveo-Meteorological Data. Hydrology 2020, 7, 38. [Google Scholar] [CrossRef]
  45. Winstral, A.; Elder, K.; Davis, R.E. Spatial Snow Modeling of Wind-Redistributed Snow Using Terrain-Based Parameters. J. Hydrometeorol. 2002, 3, 524–538. [Google Scholar] [CrossRef]
  46. Meiman, J.R. Snow accumulation related to elevation, aspect, and forest canopy. In Proceedings of the Snow Hydrology Workshop Seminar, Fredericton, NB, Canada, 28–29 February 1968; pp. 35–47. [Google Scholar]
  47. Reis, W.; McGrath, D.; Elder, K.; Kampf, S.; Rey, D. Quantifying aspect-dependent snowpack response to high-elevation wildfire in the Southern Rocky Mountains. Water Resour. Res. 2024, 60, e2023WR036539. [Google Scholar] [CrossRef]
  48. López-Moreno, J.I.; Revuelto, J.; Gilaberte, M.; Morán-Tejeda, E.; Pons, M.; Jover, E.; Esteban, P.; García, C.; Pomeroy, J.W. The Effect of Slope Aspect on the Response of Snowpack to Climate Warming in the Pyrenees. Theor. Appl. Clim. 2014, 117, 207–219. [Google Scholar] [CrossRef]
  49. Daly, S.F.; Davis, R.; Ochs, E.; Pangburn, T. An approach to spatially distributed snow modelling of the Sacramento and San Joaquin basins, California. Hydrol. Process. 2000, 14, 3257–3271. [Google Scholar] [CrossRef]
  50. Dawson, N.; Broxton, P.; Zeng, X. Evaluation of Remotely Sensed Snow Water Equivalent and Snow Cover Extent over the Contiguous United States. J. Hydrometeor. 2018, 19, 1777–1791. [Google Scholar] [CrossRef]
  51. Hammond, J.C.; Sexstone, G.A.; Putman, A.L.; Barnhart, T.B.; Rey, D.M.; Driscoll, J.M.; Liston, G.E.; Rasmussen, K.L.; McGrath, D.; Fassnacht, S.R.; et al. High Resolution SnowModel Simulations Reveal Future Elevation-Dependent Snow Loss and Earlier, Flashier Surface Water Input for the Upper Colorado River Basin. Earth’s Future 2023, 11, e2022EF003092. [Google Scholar] [CrossRef]
  52. Barnhart, T.B.; Putman, A.L.; Heldmyer, A.J.; Rey, D.M.; Hammond, J.C.; Driscoll, J.M.; Sexstone, G.A. Evaluating distributed snow model resolution and meteorology parameterizations against streamflow observations: Finer is not always better. Water Resour. Res. 2024, 60, e2023WR035982. [Google Scholar] [CrossRef]
  53. Helsel, D.R.; Frans, L.M. Regional Kendall test for trend. Environ. Sci. Technol. 2006, 40, 4066–4070. [Google Scholar] [CrossRef] [PubMed]
  54. Fassnacht, S.R.; Cherry, M.L.; Venable, N.B.H.; Saavedra, F. Snow and albedo climate change impacts across the United States Northern Great Plains. Cryosphere 2016, 10, 329–339. [Google Scholar] [CrossRef]
  55. Gómez-Landesa, E.; Rango, A. Operational snowmelt runoff forecasting in the Spanish Pyrenees using the snowmelt runoff model. Hydrol. Process. 2002, 16, 1583–1591. [Google Scholar] [CrossRef]
  56. Painter, T.H.; Skiles, S.M.; Deems, J.S.; Brandt, W.T.; Dozier, J. Variation in Rising Limb of Colorado River Snowmelt Runoff Hydrograph Controlled by Dust Radiative Forcing in Snow. Geophys. Res. Lett. 2018, 45, 797–808. [Google Scholar] [CrossRef]
  57. Doskocil, L.G.; Fassnacht, S.R.; Barnard, D.M.; Pfohl, A.K.D.; Derry, J.E.; Sanford, W.E. Twin-Peaks Streamflow Timing: Can We Use Forest and Alpine Snow Melt-Out Response to Estimate? Water 2025, 17, 2017. [Google Scholar] [CrossRef]
  58. Flynn, H.; Fassnacht, S.R.; MacDonald, M.S.; Pfohl, A.K.D. Baseflow from Snow and Rain in Mountain Watersheds. Water 2024, 16, 1665. [Google Scholar] [CrossRef]
  59. Eckhardt, K. How to construct recursive digital filters for baseflow separation. Hydrol. Process. 2005, 19, 507–515. [Google Scholar] [CrossRef]
  60. Taylor, S.; Feng, X.; Williams, M.; McNamara, J. How isotopic fractionation of snowmelt affects hydrograph separation. Hydrol. Process. 2002, 16, 3683–3690. [Google Scholar] [CrossRef]
  61. Miller, M.P.; Susong, D.D.; Shope, C.L.; Heilweil, V.M.; Stolp, B.J. Continuous estimation of baseflow in snowmelt-dominated streams and rivers in the Upper Colorado River Basin: A chemical hydrograph separation approach. Water Resour. Res. 2014, 50, 6986–6999. [Google Scholar] [CrossRef]
  62. Bates, C.G.; Henry, A.J. Second Phase of Streamflow Experiment at Wagon Wheel Gap, Colo. Mon. Weather. Rev. 1928, 56, 79–80. [Google Scholar] [CrossRef]
  63. Wilm, H.G.; Dunford, E.G. Effect of Timber Cutting on Water Available for Stream Streamflow from a Lodgepole Pine Forest; Department of Agriculture: Washington, DC, USA, 1948; Report No. 968. [Google Scholar]
  64. Pinzon, J.; Tucker, C. A Non-Stationary 1981–2012 AVHRR NDVI3g Time Series. Remote Sens. 2014, 6, 6929–6960. [Google Scholar] [CrossRef]
  65. Gamon, J.A.; Field, C.B.; Goulden, M.L.; Griffin, K.L.; Hartley, A.E.; Joel, G.; Penuelas, J.; Valentini, R. Relationships Between NDVI, Canopy Structure, and Photosynthesis in Three Californian Vegetation Types. Ecol. Appl. 1995, 5, 28–41. [Google Scholar] [CrossRef]
  66. Ren, Z.; Zheng, H.; He, X.; Zhang, D.; Yu, X.; Shen, G. Spatial estimation of urban forest structures with Landsat TM data and field measurements. Urban For. Urban Green. 2015, 14, 336–344. [Google Scholar] [CrossRef]
  67. McGrath, D.; Zeller, L.; Bonnell, R.; Reis, W.; Kampf, S.; Williams, K.; Okal, M.; Olsen-Mikitowicz, A.; Bump, E.; Sears, M.; et al. Declines in peak snow water equivalent and elevated snowmelt rates following the 2020 Cameron Peak wildfire in Northern Colorado. Geophys. Res. Lett. 2023, 50, e2022GL101294. [Google Scholar] [CrossRef]
  68. OpenET. OPEN ET—Filling the Biggest Data Gap in Water Management. 2025. Available online: https://etdata.org/ (accessed on 27 June 2025).
Climate 13 00177 g001
Figure 1. Distribution of gauging stations across Colorado in the Southern Rocky Mountains.
Figure 1. Distribution of gauging stations across Colorado in the Southern Rocky Mountains.
Climate 13 00177 g001
Climate 13 00177 g002
Figure 2. Study watershed parameters of (a) mean basin elevation, (b) mid-April clear sky solar radiation input, (c) slope, and (d) basin area (logarithmic scale). The time trend in the (e) basin-wide NDVI, (f) basin-wide winter precipitation, (g) peak SWE from the nearest SNOTEL station, and (h) average temperature from the nearest SNOTEL station. The watersheds are ordered from top to bottom by decreasing latitude with horizontal dashed lines separating the basins by proximity, as per Figure 1. Bars with solid outlines are statistically significant trends at p < 0.05 and dashed bars are moderately significant at p < 0.1 (in (e–h)). For a small decreasing trend (in (e–h)) an em dash (–) was added to the left of the bar, for a small increasing trend a plus sign (+) was added to the right of the bar, and for no trend a zero (0) was added to the right of the bar.
Figure 2. Study watershed parameters of (a) mean basin elevation, (b) mid-April clear sky solar radiation input, (c) slope, and (d) basin area (logarithmic scale). The time trend in the (e) basin-wide NDVI, (f) basin-wide winter precipitation, (g) peak SWE from the nearest SNOTEL station, and (h) average temperature from the nearest SNOTEL station. The watersheds are ordered from top to bottom by decreasing latitude with horizontal dashed lines separating the basins by proximity, as per Figure 1. Bars with solid outlines are statistically significant trends at p < 0.05 and dashed bars are moderately significant at p < 0.1 (in (e–h)). For a small decreasing trend (in (e–h)) an em dash (–) was added to the left of the bar, for a small increasing trend a plus sign (+) was added to the right of the bar, and for no trend a zero (0) was added to the right of the bar.
Climate 13 00177 g002
Climate 13 00177 g003
Figure 3. Trends in the streamflow timing (a) tQstart, (b) tQend, (c) tQstart–end, and volume (d) Q100, (e) Qstart, (f) Qend, and (g) Qstart–end across the 39 study basins. Bars with solid outlines are statistically significant trends at p < 0.05 and dashed bars are moderately significant at p < 0.1. For a small decreasing trend an em dash (–) was added to the left of the bar, for a small increasing trend a plus sign (+) was added to the right of the bar, and for no trend a zero (0) was added to the right of the bar.
Figure 3. Trends in the streamflow timing (a) tQstart, (b) tQend, (c) tQstart–end, and volume (d) Q100, (e) Qstart, (f) Qend, and (g) Qstart–end across the 39 study basins. Bars with solid outlines are statistically significant trends at p < 0.05 and dashed bars are moderately significant at p < 0.1. For a small decreasing trend an em dash (–) was added to the left of the bar, for a small increasing trend a plus sign (+) was added to the right of the bar, and for no trend a zero (0) was added to the right of the bar.
Climate 13 00177 g003
Climate 13 00177 g004
Figure 4. Pearson correlation coefficient between streamflow timing (tQstart, tQend, tQstart–end) or volume (Q100, Qstart, Qend, Qstart–end) trends and changes in temporal trends in NDVI, winter precipitation, peak SWE, melt temperature, or watershed terrain parameters (mean radiation, mean elevation, mean slope, basin area, latitude, longitude) across the 39 study basins.
Figure 4. Pearson correlation coefficient between streamflow timing (tQstart, tQend, tQstart–end) or volume (Q100, Qstart, Qend, Qstart–end) trends and changes in temporal trends in NDVI, winter precipitation, peak SWE, melt temperature, or watershed terrain parameters (mean radiation, mean elevation, mean slope, basin area, latitude, longitude) across the 39 study basins.
Climate 13 00177 g004
Table 1. Linear multi-variate regression results for (a) regression with all variables (NDVI, winter precipitation, peak SWE, melt temperature) and parameters (basin mean solar radiation, basin mean elevation, basin mean slope, area, latitude, longitude), (b) regression with winter precipitation, slope and latitude, and (c) regression with solar radiation, slope and latitude, and (d) regression slope and latitude as the independent variables to estimate timing (tQstart, tQend, tQstart–end) and streamflow volumes (Q100, Qstart, Qend, Qstart–end). The coefficient of determination (R2) and the statistical significance are presented with the regression coefficients. The moderately significant correlations (p < 0.1) are italicized and denoted with a plus sign (+); the significant correlations (p < 0.05) are in bold and denoted with an asterisk (*).
Table 1. Linear multi-variate regression results for (a) regression with all variables (NDVI, winter precipitation, peak SWE, melt temperature) and parameters (basin mean solar radiation, basin mean elevation, basin mean slope, area, latitude, longitude), (b) regression with winter precipitation, slope and latitude, and (c) regression with solar radiation, slope and latitude, and (d) regression slope and latitude as the independent variables to estimate timing (tQstart, tQend, tQstart–end) and streamflow volumes (Q100, Qstart, Qend, Qstart–end). The coefficient of determination (R2) and the statistical significance are presented with the regression coefficients. The moderately significant correlations (p < 0.1) are italicized and denoted with a plus sign (+); the significant correlations (p < 0.05) are in bold and denoted with an asterisk (*).
VariableR2Sign. FInterceptNDVIWinter PPeak SWEMelt TempSolar Rad.Elev.SlopeAreaLat.Long.
(a) Regression with all variables/parameters
tQstart0.160.8648.60.1830.177−0.007−2.86−0.0010.002−0.1380.0000.0140.503
tQend0.360.16−32.2−3.060.805 +−0.0073.40−0.015 *0.000−0.235 +−0.0010.674−0.323
tQstart–end0.150.897.45−2.580.1990.0124.20−0.010−0.003−0.0660.0000.437−0.002
Q1000.370.14−1870−11.32.510.532−6.94−0.0600.016−5.38 *0.03624.5−10.0
Qstart0.320.26−1035.990.92−0.015−3.660.0070.006−0.2670.012 +−0.513−0.908
Qend0.520.01 *−1390−28.77.530.46210.9−0.079−0.014−3.33 *−0.01022.5 *−7.12
Qstart–end0.460.03 *−1170−29.94.190.46211.0−0.068−0.020−3.32 +−0.05723.9 +−4.56
(b) Regression with winter precipitation, slope and latitude
tQstart0.0710.45−5.9 0.23 −0.096 0.15
tQend0.210.043 *−0.03 0.49 −0.25 * 0.086
tQstart–end0.210.804.23 0.006 −0.13 −0.046
Q1000.230.024 *−632 2.64 −4.17 * 18.2
Qstart0.080.3953.2 1.12 −0.03 −1.25
Qend0.400.0004 * −686 + 6.24 −3.25 * 19.0 *
Qstart–end0.360.001 *−962 * 3.16 −3.74 * 26.1 *
(c) Regression with solar radiation, slope and latitude
tQstart0.060.55 −0.001 −0.11 0.44
tQend0.250.02 * −0.011 + −0.27 * 1.1
tQstart–end0.070.45 −0.008 −0.13 0.28
Q1000.270.01 * −0.12 −4.3 * 25.9 *
Qstart0.0060.97 −0.001 −0.09 0.22
Qend0.390.001 * −0.083 −3.5 * 29.6
Qstart–end0.360.001 * −0.044 −3.9 * 31.5 *
(d) Regression with slope and latitude
tQstart0.0570.35−16.9 −0.107 0.42
tQend0.170.035 *−23.7 −0.271 * 0.68
tQstart–end0.0280.603.94 −0.13 −0.039
Q1000.230.009 *−758 + −4.29 * 21.3 *
Qstart0.0060.90−4.69 −0.86 0.20
Qend0.370.0003 * −984 * −3.54 * 26.4 *
Qstart–end0.350.0004 *−1113 * −3.89 * 29.8 *
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

Fassnacht, S.R.; Pfohl, A.K.D. Snowmelt Streamflow Trends over Colorado (U.S.A.) Mountain Watersheds. Climate 2025, 13, 177. https://doi.org/10.3390/cli13090177

AMA Style

Fassnacht SR, Pfohl AKD. Snowmelt Streamflow Trends over Colorado (U.S.A.) Mountain Watersheds. Climate. 2025; 13(9):177. https://doi.org/10.3390/cli13090177

Chicago/Turabian Style

Fassnacht, Steven R., and Anna K. D. Pfohl. 2025. "Snowmelt Streamflow Trends over Colorado (U.S.A.) Mountain Watersheds" Climate 13, no. 9: 177. https://doi.org/10.3390/cli13090177

APA Style

Fassnacht, S. R., & Pfohl, A. K. D. (2025). Snowmelt Streamflow Trends over Colorado (U.S.A.) Mountain Watersheds. Climate, 13(9), 177. https://doi.org/10.3390/cli13090177

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