Upper Missouri River Basin Soil Moisture and Snowpack Dashboard

UMRB Soil Moisture & Snowpack Dashboard

UMRB Monitoring Network Data: View and download time series graphs of soil moisture and snow data, as well as photos of the mesonet stations.

View soil moisture and snowpack data from the UMRB state mesonets:

• Montana Mesonet
• Nebraska Mesonet
• Mesonet at South Dakota State University
• North Dakota Agricultural Weather Network (NDAWN)
• Wyoming Mesonet

USACE Upper Missouri River Basin Soil Moisture and Plains Snow Monitoring Build-Out

Current Status of the Build-Out

Upper Missouri River Basin Soil Moisture and Plains Snow Data Value Study

Weekly changes in soil moisture conditions provide one of the most straightforward measures of evolving or trending hydrological conditions. To represent a variety of short- to long-term conditions, weekly changes in volumetric water content (VWC) are represented in units of m3/m3 for topsoil (2-inch and 4-inch depth observations), subsoil (8-inch and 20-inch depth observations), and column (2-, 4-, 8-, and 20-inch depth observations). Differences are computed as the most recent Tuesday at 12Z minus Tuesday at 12Z X weeks ago.

Computing Change Maps with Frozen Soils

The type of soil moisture sensors deployed across the Upper Missouri River Basin and at most monitoring networks are not well suited to quantify ice water content, which complicates efforts to derive soil moisture change around the winter season. However, ice water content is generally unaffected by hydrological processes (i.e., gravitation drainage, lateral or downhill flow, or evapotranspiration) such that it can be well approximated by liquid water content immediately following pre-freeze. A general consensus based on feedback from soil scientists and leading experts in this field has identified that soil temperatures at or above 2°C have near negligible ice water content. Given the stasis of ice water content, it is reasonable to compute a soil moisture change when one or both ends of the change period are in a frozen condition by substituting the most recent unfrozen (soil temperature > 2°C) soil moisture observation. Stations with shifted dates due to frozen soil conditions are denoted with a triangle on the soil moisture change maps.

Soil Moisture Change Plots

To compute soil moisture change, valid observations of both soil moisture and temperature are necessary. Differences are to be computed at each depth individually as current Tuesday at 12Z (SMcurrent) minus starting Tuesday at 12Z X-weeks ago (SMx-weeks). For instance, a 4-week soil moisture change is SMcurrent minus SM4-weeks (Fig 1a). Soil moisture change at each depth is then averaged over the respective layer's (i.e., top, subsoil, column) depths.

To derive a soil moisture change for a period, set i to the current Tuesday at 12Z representing SMcurrent.

• If soil temperature at week i (STi) is missing then set SMcurrent to missing.
• If soil moisture at week i (SMi) is missing and STi ≥ 2°C, then set SMcurrent to missing.
• If STi > 2°C then set SMcurrent to SMi.
• If STi is less than < 2°C then set i to the previous week (i-1) and repeat this process until one finds an observation week in the past where STi > 2°C. Then, set SMcurrent to SMi

This same process will be used to define SMx-weeks with i initially set to the respective Tuesday at 12Z X-weeks ago.

In this way, changes in soil moisture that temporally include frozen soils represent the liquid water content change over that period (Fig. 1). For a 4-week change map, SMcurrent and SM4-week will represent soil moisture conditions based on soil temperature for Fig. 1. 

• When STcurrent ≥ 2°C and STx-weeks ≥ 2°C (outside of frozen period), SMcurrent and SM4-week represented the soil moisture conditions for the current Tuesday and Tuesday 4 weeks ago at 12Z (Fig. 1a). SMChange = SMcurrent – SM4-weeks
• When STcurrent < 2°C and ST4-weeks ≥ 2°C (going into a frozen period) SMcurrent will be set to the soil moisture value on the Tuesday at 12Z denoted by L (SML) as noted in (Fig. 1b).
• When STcurrent < 2°C and ST4-weeks < 2°C (contained in frozen period) SMcurrent and SMx-weeks are both set to the Tuesday at 12Z denoted by L` (SML`), which represents no change (Fig. 1c).
• When STcurrent ≥ 2°C and STx-weeks < 2°C (emerging from frozen period), SM4-weeks is set to the Tuesday at 12Z denoted by L` (SML`) as shown in Fig. 1d.

Fractional available water (or FAW) is a measure of the amount of moisture in the soil available to plants. Since this measure is sensitive to soil type (e.g., density, texture), FAW is a useful metric that accounts for differences in soil properties across diverse regions and a useful metric for agricultural drought. FAW values are typically represented as a fraction; however, for this product we are displaying them as a percentage ranging from 0% (limited available water) to 100% (plenty of available water). It is possible to exceed 100% during heavy rains and dip below 0% during droughts, but values above 100% and below 0% were capped at the respective 100% and 0% levels. This was done since values below 0% do not necessarily change the interpretation that water available to plants is limited; this is also the case for values above 100%. 

FAW is computed as a ratio of the current available water to plants (AvailableWater) divided by soil water capacity (WaterCapacity; see eq. 1). These quantities are based on two critical thresholds that describe the lower and upper limits of soil moisture available to plants, referred to as the wilting point (WP) and field capacity (FC), respectively. Moisture in the soil at and below WP would require capillary pressures beyond the capacity of most vegetation (-1,500 kPa) to extract additional moisture from the soil. During heavy precipitation events, soil moisture conditions can exceed FC such that the additional moisture will be quickly drained (to deeper depths or horizontally) before plants can access that moisture and is therefore unavailable to the plant as well. Given these conditions, AvailableWater is defined as the current soil moisture observation at UMRB mesonet stations (ObsSM) minus WP (eq. 2) with WaterCapacity derived as FC minus WP (eq. 3). As a result, FAW is computed as shown in equation 4. 

FAW = AvailableWater / WaterCapacity (eq. 1)

AvailableWater = ObsSM - WP (eq. 2)

WaterCapacity = FC - WP (eq. 3)

FAW = (ObsSM - WP)/(FC - WP) (eq. 4)

The WP and FC thresholds were defined for each station depth using a combination of soil property information from USDA’s gNATSGO database (i.e., percentages of  sand, silt, and clay) to drive the Rosetta version 3 model. This model estimates the van Genuchten parameters allowing one to build a water retention curve for each station depth. The station depth’s WP and 

FC were extracted along this curve at -1,500 kPa (WP) and -33 (FC) kPa, respectively. For station depths that had over 52% sand or a combination of higher percentages of sand (~42% or more) and silt (~28% or more), the -10 kPa threshold was used to define FC.

Layer composites (e.g., topsoil, subsoil) of FAW were derived as the sum of the 24-hour averaged AvailableWater ending Tuesday at 12Z for each depth in the layer divided by the sum of the WaterCapacity over each depth in that layer. For instance, topsoil FAW is computed as the sum of the 2-inch and 4-inch depths’ hourly (h) averaged AvailableWater observations, divided by the sum of their respective WaterCapacities. This process is repeated for the subsoil (8-inch and 20-inch) and column (2-inch, 4-inch, 8-inch, and 20-inch) layers, respectively.

TopFAW = sum(mean(ObsSM2inh-WP2in),mean(ObsSM4inh-WP4in)) 
/ sum((FC2in-WP2in),(FC4in-WP4in))

Fractional available water (FAW) estimates the amount of water in the soil available for plant use; however, the ability of vegetation to access and retain soil moisture varies by species (e.g., native grasses can retain more moisture than crops). Categorized soil moisture helps identify when plants are moisture limited based on critical theta (CT). Similar to wilting point (WP) and field capacity (FC), CT is another soil moisture threshold that represents the condition at which further drying reduces vegetation transpiration. 

Using these three thresholds, it is possible to classify soil moisture conditions into four broad categories that are representative of USDA’s soil moisture categories: Very Short (representing soil moisture below the WP), Short (soil moisture between WP and CT), Adequate (soil moisture between CT and FC), and Surplus (soil moisture greater than FC). Categorizing soil moisture conditions in this way allows one to easily distinguish between stations where soil moisture conditions are water limited for vegetation development. 

The WP and FC thresholds were defined for each station depth using a combination of soil property information from USDA’s gNATSGO database (i.e., percentages of  sand, silt, and clay) to drive the Rossetta version 3 model. See more details under the “Fractional Available Water” product description. Critical theta (CT) represents the soil moisture condition at which further reductions in soil moisture limits evapotranspiration processes from plants. In the literature, CT is often estimated by identifying the wettest soil moisture condition at which values less than this threshold correlate with warmer land-surface temperatures. Given the lack of an available record to confidently identify this threshold, the CT was simply defined as the halfway point between WP and FC (eq. 1).  

Critical Theta = (FC + WP)/2 (eq. 1)

Categorical conditions were derived for each station depth over the 24-hour period ending 12Z Tuesday. Hourly soil moisture observations (SMobs) were placed into one of the four categories based on the observation’s relationship to WP, CT, and FC thresholds:

• “Surplus”: SMobs is greater than or equal to FC
• “Adequate”: SMobs is greater than or equal to CT but less than FC
• “Short”: SMobs is greater than or equal to the WP but less than CT
• “Very Short”: SMobs is less than the WP 

The hourly categorical values for each depth were then aggregated by layer based on the most dominant hourly condition category across the depths (i.e., greatest frequency). For instance, the topsoil (2-inch and 4-inch depths) categorical value is based on the most frequent category reported among the 48 reported categories (24 observations for each of the two depths). In the event of a tie, the wetter of the two dominant conditions was chosen. The same approach was applied to the subsoil and total column layers. For the total column, it is important to note this approach weights the upper soil layers (2-inch and 4-inch) slightly more than the deeper depths (8-inch and 20-inch) given the higher likelihood of similar conditions observed from sensors in close proximity. That said, once a sufficient period of record is established to more fully estimate CT (surface temperature response to soil moisture) this limitation will be resolved.