Accurate water-balance measurements in the seasonal,
snow-dominated Sierra Nevada are important for forest and downstream water
management. However, few sites in the southern Sierra offer detailed records
of the spatial and temporal patterns of snowpack and soil-water storage and
the fluxes affecting them, i.e., precipitation as rain and snow, snowmelt,
evapotranspiration, and runoff. To explore these stores and fluxes we
instrumented the Wolverton basin (2180–2750 m) in Sequoia National Park
with distributed, continuous sensors. This 2006–2016 record of snow depth,
soil moisture and soil temperature, and meteorological data quantifies the
hydrologic inputs and storage in a mostly undeveloped catchment. Clustered
sensors record lateral differences with regards to aspect and canopy cover at
approximately 2250 and 2625 m in elevation, where two meteorological
stations are installed. Meteorological stations record air temperature,
relative humidity, radiation, precipitation, wind speed and direction, and
snow depth. Data are available at hourly intervals by water year
(1 October–30 September) in non-proprietary formats from online data
repositories (
The western slope of the Sierra Nevada, California, has an elevation-driven climate gradient where wintertime snowpack accumulation provides for a large part of California's annual water needs. Small temperature increases will raise the snowline elevation and alter snowmelt, and subsequent water use by vegetation, streamflow, and forest dryness (Goulden and Bales, 2014). The implications of such a shift affect forest management, downstream water management, and ultimately, regional water policy. This research program in the Wolverton basin (1720–3500 m elevation) of Sequoia National Park is part of a larger effort to quantify the two main near-surface stores of water, snowpack, and soil-water storage, across the current Sierra Nevada rain–snow transition.
As part of this research, we used distributed snow and soil sensors, as well as remotely sensed data, to characterize snow accumulation, snowmelt, and other components of the distributed water balance. The data set covers late 2006 through the end of water year (WY) 2016. This period includes the drought years of 2012–2015 as well as the high snow year in WY 2011.
Kirchner et al. (2014) found that snowmelt at the upper elevations occurs later in the season and at faster rates compared to lower elevations. Historically, the Wolverton basin has been in the seasonal snow zone of the southern Sierra Nevada, characterized by winter snow accumulation and spring ablation. At the lower elevations, winter snow cover is ephemeral in some years, with 70 %–95 % of the snow-covered season having slow melt rates (Kirchner et al., 2014). At upper elevations, melt occurs during less than 65 % of the season. Based on a 2010 set of lidar flights (Anderson et al., 2012), snow depth near the date of accumulation is greatest at an elevation of 3300 m, which is 200 m below the highest ridge. Between 1850 and 3300 m, snow depth increases at a rate of approximately 15 cm per 100 m; at higher elevations, snow depth decreases at a rate of 48 cm per 100 m (Kirchner et al., 2014). Elevation is one of the most important factors for determining snow depth across the catchment, though snow depth is also mediated by canopy penetration fraction northness and vegetation height (Zheng et al., 2016; Tennant et al., 2017). The difference between open canopy and under-canopy areas increases through the rain–snow transition up to 25–45 cm at high elevations (Fig. 1).
Meteorological factors, as well as the interference of canopy coverage, impact snow accumulation and melt at these sites. Sub-canopy direct-beam irradiance and a sky-view factor explain the most variation in snow ablation rates, especially at finer timescales (Musselman et al., 2012). A time-varying canopy parameter in snow modeling reduces errors by 7 days in the simulated snow disappearance date, and errors in the timing of soil-water fluxes by 11 days, on average, compared to a bulk parameterization of radiation transfer through the canopy (Kirchner et al., 2014).
Studies of this basin indicate that it is sensitive to a changing climate. The timing and rate of snowmelt indicate that this elevation range is sensitive to seasonal meteorology, especially where upper elevations may begin to experience snow melt during more of the snow-covered season (Kirchner et al., 2014). At both the upper and lower sites, peak soil moisture precedes the average date of snow disappearance, meaning that soil moisture declines even while snowmelt is infiltrating into the soil system (Harpold et al., 2015). With peak snow depth around 3300 m (Kirchner et al., 2014), such changes to the hydrological system could have major implications for snowpack water storage and runoff.
Snow-depth changes over elevation observed from the lidar data in
the Wolverton basin
Wolverton catchment in the Marble Fork of the Kaweah River, Sequoia
and Kings Canyon National Parks. Distributed instruments are clustered around
the
The Wolverton basin is a montane, forested catchment northeast of Giant
Forest in Sequoia National Park (Fig. 2). In the 5.4 km
Most of the catchment is undeveloped forest with no history of thinning, though there was a small downhill ski area in the basin for about 50 years until 1994. There is some recreational infrastructure at the base of the catchment, including trailhead parking lots, a water treatment plant, and recreational buildings. In addition, there are several recreational hiking trails around the catchment.
During the WY 2007–2016 period, mean annual precipitation was 728 mm at the
nearby cooperative Lodgepole temperature and weather station operated by
Sequoia and Kings Canyon National Parks. The mean annual temperature was
6.0
We installed and instrumented two meteorological stations (met) in 2008. The Wolverton meteorological station is at 2206 m, on a 7 m steel triangular-frame tower. The Panther meteorological station, at 2618 m, is a 6.1 m steel pole with cross-arms near the apex. Sensor instruments were installed for air temperature, relative humidity, wind speed, wind direction, net radiation, solar radiation, snow depth, and soil moisture and soil temperature. Sensor manufacturers and installation heights are listed in Table 1. Data are recorded at 60 min intervals on a on Campbell Scientific CR1000 data logger. Data acquisition programs are located on the UC Merced-SNRI digital library with the data.
Meteorological station instrumentation.
The distributed sensor nodes were installed in four clusters at the upper and
lower elevations in roughly north and south aspects (Fig. 2). The
lower-elevation sites, 1 and 2, are south of Wolverton meteorological station
and face north and southeast, respectively, at elevations of around 2226 and
2260 m. The upper-elevation sites, 3 and 4, are near the Panther
meteorological station and face southeast and north, respectively, at
elevations of around 2596 and 2636 m. Soil moisture and temperature, snow
depth, and meteorological data (air temperature, relative humidity, and solar
radiation) are measured at these clusters across the network at hourly
intervals. Snow depth is stratified by canopy coverage, with 15 measurements
at open canopy locations, six at the canopy drip edge, and five under the tree canopy
(canopy classification was verified on-site). Distance to snow and soil surface
and air temperature are measured with Judd Communications ultrasonic depth
sensors, using analog control. Total solar radiation is measured with the
LI-COR Environmental LI-200R Pyranometer, like the met stations. Soil
volumetric water content (VWC) and soil temperature are measured using
the Campbell Scientific CS 616 Water Content Reflectometer and 107 temperature
probes at depths of 10, 30, and 60 cm below the mineral soil surface. The
standard calibration for the CS-616 sensors was used. The fact that the sensors were
not individually calibrated to soils at each site is one of the flaws in the
original experimental design, but the standard calibration does have an
accuracy of
Data are recorded at 60 min scan intervals, with control and storage on Campbell Scientific CR1000 data logger, using an AM16/32B multiplexer. Program for data acquisition are located on the UC Merced-SNRI digital library with the data.
The 24 sensor nodes highlight the spatial variability of soil moisture and snow depth through water years 2010 (average), 2011 (record-setting wet year), and 2012 (extremely dry). Soil moisture can be higher at the lower elevation than the higher elevation, and peak soil moisture may predate the end of snowmelt.
In water years 2007–2016, snowpack was deeper and more persistent at the
higher-elevation sites (2640 m; Fig. 3a). Warmer temperatures at the
lower-elevation sites (2245 m; Fig. 3b) result in earlier soil wet-up,
higher mid-winter soil moisture storage, and earlier peak soil moisture
storage in wet and dry years. Long data gaps at Site 2 in 2013 and at Site 1
in 2016 are due to battery and logger issues. Comparing average (WY 2010),
wet (WY 2011), and dry (WY 2012) years, snowpack is persistent at both the
upper- and lower-elevation sites (Fig. 4). However, in WY 2014 and 2015,
snowpack receded multiple times during each winter. Soil moisture peaks
earlier at the two lower-elevation clusters. The Wolverton met station
experienced a wider swing in temperatures than the Panther met station. Mean
annual temperature was 6.3
Some of the sensors have inherent possible biases. The precipitation sensors are unshielded and unheated tipping buckets. Precipitation measurements in each data set will underreport precipitation through undercatch, freezing, and coverage by snow. Considering this bias, the precipitation records shown for comparison with the other meteorological data (in Fig. 3a and b). That meteorological station has a weighing gauge with a long-term record, though the sensor is also unshielded.
Data are available at
hourly intervals by water year (WY; 1 October–30 September), from WY 2007
through WY 2016. Data are available through online data repositories.
Distributed snow depth, air temperature, and soil moisture and temperature
are available through
A 10-year meteorological and hydrologic data record is presented for a catchment in Sequoia National Park, in the southern Sierra Nevada. Distributed snow depth and soil temperature and moisture combined with two meteorological stations provide a means of testing and evaluating hydrologic processes in a productive montane forest. The Wolverton basin research site serves as a southern comparison point with installations in the Kings River (SSCZO; 2008–present; Bales et al., 2018), Merced River (MRB; 2006–present; Roche, 2017), Stanislaus River (2013–present, Pickard, 2015), American River (2014–present; Zhang et al., 2016, 2017a, b), and Feather River (2016–present; Avanzi et al., 2018) basins. Studies compiled to date indicate that hydrological variables, including snow depth, the timing and rate of snow melt, and soil moisture, are susceptible to changing climate patterns like warmer temperature or increased vegetation water demands in this part of the Sierra Nevada. We invite others to use this data for their own studies of the basin or as a comparison point to further the discussion.
RCB, MHC, and PBK designed the sensor networks. PBK, EMS, XM, and ZZ installed and maintained the sensor networks and processed the sensor-network data. RCB, EMS, and ZZ prepared the paper with contributions from all authors.
The authors declare that they have no conflict of interest.
This article is part of the special issue “Hydrometeorological data from mountain and alpine research catchments”. It is not associated with a conference.
We thank the Sequoia and Kings Canyon National Parks research and permitting staff and the staff and research team at the Southern Sierra Critical Zone Observatory. This research was funded in part by NSF EAR-0725097, EAR-1239521, and EAR-1331939 for the Southern Sierra Critical Zone Observatory, and the University of California, Merced. Edited by: Danny Marks Reviewed by: Edward Bair and one anonymous referee