Development and Validation of the Tahoe Project Sediment Model
SNPLMA Project Number P052
Final Report
Investigators
Dr. William Elliot, Rocky Mountain Research Station
Dr. Erin Brooks, University of Idaho
Drea Traeumer, Em Hydrology
December, 2014
1
Project Abstract
Now more than ever there is a great need for scientifically defensible upland decision support tools in
the Lake Tahoe basin. The recognition of the excess build-up of forest fuels and the continued struggle
with maintaining or improving clarity in the lake has put enormous pressure on decision-makers in the
basin. Managers are faced with stiff time constraints to meet TMDL requirements while thinning forests
to reduce the risk or intensity of wildfire in the basin. The project developed an upland decision support
tool to assist managers in the selection and assessment of site-specific management options to reduce
forest fuel loads and to evaluate effectiveness of sediment mitigation practices. The online Tahoe Basin
Sediment Model (TBSM) is based on the WEPP model and will be parameterized using existing Tahoe
experimental databases. TBSM is a flexible web-interface tool which will assess the effects of site
specific management practices on sediment transport and delivery from a treated hillslope to a channel.
One of the primary focuses was to develop training materials that clearly describe how to apply the tool
to many of the common fuel management and sediment management practices employed by decisionmakers in the basin. The web interface was developed such that new information can be incorporated
into the tool as it becomes available from on-going research in or near the basin. TBSM will have an
immediate impact in the basin as it provides the scientifically defensible predictions that watershed
managers in the basin so desperately need.
Goals and Objectives
The goal of this project was to develop the Tahoe Basin Sediment Model (TBSM) online decision support
tool to be used by forest managers and planners in the Tahoe Basin to assess hillslope-scale fine
sediment (<20 um) loads for the most common forest upland management practices in the basin under
current and future climate scenarios. The primary objectives of the project were to:
1. Compile a database of existing upland rainfall, runoff, and erosion experiments in the Tahoe
basin.
2. Develop WEPP input files from existing datasets in the basin for the most common forest upland
management practices.
3. Develop a climate generation tool that creates current and future climate files for any project
location in the Tahoe basin.
4. Develop the TBSM with user friendly protocols for evaluating the effects of alternative
management practices on fine sediment loads.
5. Validate TBSM loading estimates for fine sediment (<20um) from current and proposed
monitoring projects within the basin).
6. Publish results in a peer reviewed journal
Activities
Survey for Tahoe Basin Needs
The first phase of this project included two surveys. The first survey was to determine from Basin
stakeholders their needs for soil erosion prediction technology. There were 20 surveys distributed, but
there were only three returned. The returns did, however, represent the three main stakeholder groups
in the basin, a regulatory agency, a management agency, and a fire protection agency.
2
The survey found that the respondents were most interested in impacts of mechanical fuel reduction
systems and burning, some interest in the importance of road design, but low interest in recreational
impacts. The respondents were all interested in predictions for average annual upland erosion, sediment
delivery, surface runoff, lateral flow and base flow. The water board respondents were also interested in
nutrients and snow water equivalent predictions, and the fire district respondents in soil water content
(for fire risk) and vegetation regeneration. The water board was particularly interested in monthly
distributions of predictions.
Survey of Available Data
A second survey was carried out to identify potential data sets for improving erosion models in the
basin. This survey was targeted to 14 researchers who were believed to have collected data related
to soil erosion in the Tahoe Basin in recent years. Three scientists responded to this request for
information. The Forest Service Lake Tahoe Basin Management Unit provided a link to their
database of publications, some hydraulic conductivity values from field measurements using
permeameters; however, they did not have any data from rainfall simulations. El Dorado County
provided similar information, with a database of what was considered relevant data. The county
also had data related to properties of sediment leaving roads. Both of these Forest Service and
county databases included gray literature as well as peer reviewed publications.
The most useful information was provided by Dr. Randy Foltz, USFS Rocky Mountain Research
Station and Drea Traeumer, co‐PI, who have been collecting rainfall simulation data in the Tahoe
basin. Dr. Foltz provided information on the erodibility properties of roads from his SNPLMA Round
7 research project.
Complementary Rainfall Simulation
One of the sources of soil erodibility data was to be from the Traeumor round 8 project. Initial rainfall
simulations carried out by Traeumor were not considered adequate for either the Round 8 study on jack
pot burning or this study. Subsequent rainfall simulations by Dr. Randy Foltz were carried out. These
results combined with other studies in the basin resulted in the soil properties for the Tahoe Basin
summarized in the appendix.
Soil Erodibility Studies
There are four factors that describe soil erodibility in the WEPP Technology: Interrill erodiblity, rill
erodibility, rill critical shear, and saturated hydraulic conductivity. In modeling soil erosion in the Tahoe
Basin, there are generally three conditions to consider: forests, roads, and other disturbed conditions,
such as road fills, constructions sites, or ski slopes. For forested areas, the disturbance associated with
the area is critical for defining soil properties. The main disturbance categories are: 1) Undisturbed; 2)
Mechanically disturbed; and 3) Burned. Within the Mechanically disturbed category, which is generally
applied to thinning operations, the ground cover remaining after disturbance is considered as well as a
reflection of the degree of disturbance. Typically, for simple thinning, a reduced ground cover is used
with a “shrub” soil to describe the soil erodibility after the disturbance. For major understory thinning,
which may even include removal of biomass for fuel, or a seed tree harvest, then a “tall grass” condition
soil condition would be recommended along with the ground cover after the treatment. For a
prescribed burn or low severity wildfire, a “low severity fire” condition is recommended, along with the
3
associated ground cover. In all these forest cases, data collected from different forest conditions by
Robichaud were considered as the primary source of information to describe soil erodibility. For roads in
the Tahoe Basin, data collected by Foltz (2008 and 2009) within the basin were combined with what was
learned from his earlier research and that of Luce (1999) in Oregon to develop low and high traffic road
erodibility values. For the other disturbed areas, data from two Grismer et al. publications were
considered (See Reference Section). These papers contained management practices demonstrated by
Hogan to be especially effective in mitigation, including mulching and incorporating mulch. As these
sites were highly disturbed, the erodibility values observed by Grismer were tempered with agricultural
soil erodibility values (Elliot et al., 1989). In all cases, the hydraulic conductivity reported from rainfall
simulation was generally reduced about 40 percent. This has been found to be reasonable by a number
of modelers and likely reflects the fact that simulations are carried out on very small areas of hill side,
surrounded by large areas where there is no simulation. Thus the soil matric potential of the
surrounding areas will tend to “draw” water into the soil. In addition, on steep slopes, there is likely
considerable subsurface lateral flow from the simulation site to drier downslope areas that would not be
occurring should the entire hill slope be experiencing the same rainfall or snow melt event. Once the
main erodibility parameters were determined, they were compared to each other, and a continuum of
properties developed so that the model will respond in the way the a user would expect. This
adjustment is following the guidelines established by Foster et al. (1987) in the original WEPP User
Requirements document, that the model should perform in the way that a user expects. Thus the
ranking of hydraulic conductivity is from sand to silt to alluvium, and from Forest to Grass to disturbed
to fire to roads in treatment. The interrill erodibility is similar, with silt the most erodible, followed by
sand and alluvium, whereas with rill erodibility, sand is the most erodible, followed by silt and alluvium.
These erodibility values have a similar ranking for vegetation for texture. The treatments, however, are
different with the disturbed soils being the most erodible, and then fire, roads, grass and forest soil in
decreasing erodibility. In all cases, the cover effects tend to overshadow the soil erodibility effects. A
final inclusion for soil properties in the Tahoe basin database is that forest soils are considered the
deepest, reflecting the fact that deeper soils are often needed to sustain forests, and the trees
themselves add to the water holding capacity of a site, which can only be accounted for in the current
version of WEPP by specifying a deeper soil. Deeper soils frequently have slightly less surface runoff, but
may generate more lateral flow which can form spring lines at the base of hills, creating unexpected
increases in surface runoff predictions.
Future Climates
Future climate scenarios were develop at for eight SNOTEL locations across the Tahoe Basin. Coates et
al. (2010) used two Global circulation models, the Geophysical Fluid Dynamics Laboratory Model (GFDL)
and the Parallel Climate Model (PCM) to provide 100 year records of daily future climate projections for
the A2 and B1 emissions scenarios at a 12 km grid scale for the Tahoe basin. At this resolution there are
12 points with daily projections within or very close to the Tahoe Basin, see Figure 5-1 in Coates et al.
(2010). These daily data were disaggregated into hourly data by John Riverson from Tetra-tech using
observed hourly data trends using historic weather records at SNOTEL sites in the basin, see section 4.3
of Coates et al. (2010). These data were used to drive the LSPC hydrology and water quality model. The
LSPC model breaks the Tahoe basin down into many subwatersheds and air temperature and
precipitation data for the various watersheds LSPC are developed for each watershed using adiabatic
lapse rates with elevation.
4
1800
Precip
Tmax
20
Tmin
1600
15
1200
10
1000
5
800
600
0
Annual Avg. Temperature (C)
Annual Precipitation (mm/yr)
1400
400
-5
200
0
1989
-10
2009
2029
2049
2069
2089
Figure 1. Observed and ‘uncorrected’ future climate predictions of total precipitation, average annual maximum
temperature, and average annual minimum temperature for the Heavenly SNOTEL. Thick solid lines are
observed data for 1989-2005. Notice the shift in observed between forecasted temperatures.
In order to be consistent with the Coates (2010) study we developed future climate files based on the
same downscaled climate data. Rather than large watershed, the Tahoe Sediment Basin Model provides
climate data at specific SNOTEL locations. Future climate scenario at eight SNOTEL sites in the basin
were developed using scaling factors that relate the future climate scenarios at one of the 12 locations
to the SNOTEL point through the use of 800 m resolution, 30 year average PRISM maps for Tmax, Tmin,
and Precipitation following the approach of Brooks et al. (2010).
When comparing these future climate forecasts to the observed data collected at the site we noticed in
some cases that the observed trends in temperature and precipitation did not match. Figure 1 shows
observed annual maximum temperature, minimum temperature, and precipitation as well as the
forecasted data for the Heavenly SNOTEL. As indicated by this figure the trend in observed annual
maximum temperature is a degree or two below the trend in the forecasted data. In contrast the
observed minimum annual temperature is slightly warmer than the forecasted data. This discontinuity
is likely caused by scaling problems using the average annual values from 800 m cells PRISM maps.
Micro-topography near the SNOTEL station could play a factor in controlling the micro-climate.
In order to provide continuous simulations that transition from observed data to future forecasts we
applied a secondary correction to the future climate scenarios. The maximum and minimum
temperature was shift by fixed, constant amount and the precipitation data was corrected using a fixed
scaling factor. These corrections factors were calculated by comparing the average annual values over
the previous 10 years to the forecasted data for the same period. Figure 2 shows the corrected future
5
1800
Precip
Tmax
Tmin
20
1600
15
1200
10
1000
5
800
600
0
Annual Avg. Temperature (C)
Annual Precipitation (mm/yr)
1400
400
-5
200
0
1989
-10
2009
2029
2049
2069
2089
Figure 2. Observed and ‘corrected’ future climate predictions of total precipitation, average annual
maximum temperature, and average annual minimum temperature for the Heavenly SNOTEL. Thick solid
lines are observed data for 1989-2005. Notice the shift in temperatures has been corrected.
climate data for the same Heavenly SNOTEL, notice a much smoother transition from the observed data
to the forecasted data. These corrections were made to both the A2 and B1 scenarios.
One of the challenges with working with multiple climate scenerios is that they are independent
predictions so simulations cannot be compared on a day to day basis nor even a year to year basis.
Figure 3 compares the Future climate scenarios predicted for both the A2 and B1 scenarios. Notice that
wet years in both scenarios do not match. It is for this reason that we will be recommending that users
only make comparison between the climate scenarios using at least 10 year simulations. Figure 3 shows
the 10 year running average for each scenario. The web interface provides the users with the ability to
simulate specific time periods for each climate file and therefore provides the users with the ability
isolate blocks of time.
The online interface is currently programmed so that users select the nearest SNOTEL station and initial
year to start their future climate scenario, like 2030, and the interface then abstracts the following 20year predicted climate for the scenario (A2 or B1) desired.
Phosphorus Data
Phosphorous pathways in agricultural watersheds are mainly associated with surface runoff, detached
sediments, and tile drainage water. The dominant pathway in most cases is associated with detached
sediments, while phosphorous dissolved in surface runoff and tile drainage are of less importance.
Agricultural phosphorus delivery models have tended to focus on how management practices such as
6
2000
10 per. Mov. Avg. (Precip. A2)
20
1800
10 per. Mov. Avg. (Precip B1)
15
1400
1200
10
1000
800
5
600
400
Annual Avg. Temperature (C)
Annual Precipitation (mm/yr)
1600
0
200
0
2001
-5
2021
2041
2061
2081
Figure 3. Future A2 and B1 climate predictions of total precipitation, average annual maximum temperature,
and average annual minimum temperature for the Heavenly SNOTEL. Thick solid lines are B1 predictions. The
solid black line and the dotted black line are the 10 year moving average trendline of the A2 and B1
precipitation forecasted data.
manure spreading and chemical fertilizers affect the availability of soluble reactive phosphorous (SRP)
for runoff, and the concentration of phosphorous adsorbed to soil aggregates and particles (particulate
phosphorous). The concentrations of phosphorous in eroded sediments, surface runoff, and drain tile
flows are then used in runoff and erosion models to predict phosphorous delivery.
In forested watersheds, surface runoff and erosion are frequently minimal, are generally associated with
wildfire. In the absence of wildfire, the dominant flow paths for runoff are either shallow lateral flow or
base flow (Elliot, 2013; Srivistava et al., 2013). Phosphorous concentrations in the soil are usually much
lower than in agricultural settings. Recent research has found that the concentration of SRP in the upper
layers of the soil that are the source of shallow lateral flow are much greater than measured in surface
runoff (Miller et al., 2005). These observations suggest that a phosphorous delivery model is needed for
forest watersheds that can include the current surface runoff and sediment delivery vectors, as well as
delivery from shallow lateral flow.
In order to develop a model that can predict phosphorus delivery with lateral flow, a hydrologic model
that includes shallow lateral flow as well as surface runoff and sediment delivery is needed. The Water
Erosion Prediction Project (WEPP) model has such a capability (Dun et al., 2009; Srivistava et al., 2013).
The WEPP Model is a physically-based distributed hydrology and erosion model. WEPP uses a daily time
step to predict evapotranspiration, plant growth, residue accumulation and decomposition, deep
seepage, and shallow lateral flow. Whenever there is a runoff event from precipitation and/or
7
snowmelt, WEPP predicts infiltration, runoff, sediment detachment and delivery (Flanagan and Nearing,
1995). WEPP has both a hillslope version and a watershed version. In recent years, the deep seepage has
been used to predict groundwater base flow (Elliot et al., 2010, Srivastava et al., 2013), further
increasing the model’s hydrologic capabilities.
The WEPP Model
The WEPP model was originally developed to predict surface runoff and erosion and sediment delivery
from agricultural, forest and rangeland hillslopes and small watersheds (Laflen et al., 1997). Inputs for
the model include a daily climate file, soil file, topographic information, and a management or
vegetation file. Within the model, WEPP completes a water balance at the end of every day by
considering infiltration, deep seepage, lateral flow and evapotranspiration. Surface runoff is estimated
on a subdaily time step using an input hyetograph based on the daily precipitation depth, duration, and
peak intensity and the soil water content, using a Green and Ampt Mein Larson infiltration algorithm
(Flanagan and Nearing, 1995; Dun et al., 2007). The deep seepage is estimated for saturated conditions
for multiple soil layers if desired using Darcy’s Law. Evapotranspiration is then estimated using either a
Penmen method or Ritchie’s Model. The lateral flow is then estimated for layers that exceed saturation
using Darcy’s law for unsaturated conditions as downslope conditions may not be saturated (Dun et al.,
2007). Duration of surface runoff is dependent on storm duration and surface roughness (Flanagan and
Nearing, 1995) and lateral flow duration is assumed to be 24 hours on days when lateral flow is
estimated.
If requested, WEPP generates a daily “Water” file that contains precipitation and snow melt amounts,
surface runoff, lateral flow, deep seepage, and soil water content (Flanagan and Livingston, 1995). Table
1 shows a part of the water file for the Tahoe City, CA climate for Julien Days 70-78 (March 11-19). On
day 70, precipitation (P) was all rain, on day with no snowmelt, on day 71 rainfall combined with melting
snow, and days 77 and 78 were snowmelt only days. Surface runoff (Q) only occurred on day 78, while
lateral flow was occurring every day. The soil became saturated on day 72 so that deep seepage
occurred. During these 9 days, the total precipitation was 31.6 mm, total surface runoff was 14.73 mm,
the lateral flow was 18.93 mm, the deep seepage was 0.13 mm, the soil water content increased by
73.38 mm and the snow water equivalent on the surface decreased by 106 mm. Development is ongoing
to add the deep seepage to a temporary groundwater reservoir, and from that to use a linear flow
model to predict base flow from a sub watershed as a fraction of the volume of that reservoir
(Srivastava et al., 2013).
WEPP predicts delivered sediment in five sized categories: primary clay, silt and sand particles, small
aggregates made up of clay, silt and organic matter, and larger aggregates consisting of all three primary
particles and organic matter. The sediment size classes and properties are summarized in Table 2 for a
coarse sandy loam soil that is widespread in the Tahoe Basin. The fraction of sediment in each size class
delivered from a hillslope or a watershed is presented appended to the main output file from WEPP. In
addition, WEPP calculates a specific surface enrichment ratio (SSR) which is the ratio of the sediment
surface area in the clay and organic matter fraction in the delivered sediment divided by this value for
the soil on the hillslope. This ratio was intended to be used to assist water quality modelers to
determine the increase in concentration of a pollutant in the delivered sediment compared to the
sediment on the hillslope. For example if the phosphorus content in the soil was 500 mg/kg and the
enrichment ratio was 2.2, the concentration of phosphorus in the delivered sediment would be 1100
mg/l
8
Table 1. Example of information in the WEPP water.txt file output file. The climate was for Tahoe City, CA.
The variables are: Day=julian day; P= precipitation; RM=rainfall +snowmelt; Q=daily runoff; Ep=plant
transpiration; Es=soil evaporation; Dp=deep percolation; latqcc=lateral subsurface flow; Total Soil
Water=Unfrozen water in soil profile; frozwt=Frozen water in soil profile; SWE=Snow Water Equivalent on the
surface.
Day
P
RM
Q
Ep
Es
Dp
latqcc
Total-Soil
frozwt
SWE
mm
mm
mm
mm
mm
mm
mm
Water(mm)
mm
mm
70
8.4
8.4
0.00
0
2.05
0
0.28
129.73
0
258.79
71
2
42.68
0.00
0
3.09
0
0.88
168.43
0
218.12
72
0.3
17.74
0.00
0
3.2
0.01
1.89
181.08
0
200.67
73
6.9
30.09
0.00
0
2.35
0.02
2.77
206.03
0
177.49
74
13.7
4.57
0.00
0
2.08
0.02
2.77
205.69
0.04
186.62
75
0.3
0
0.00
0
1.27
0.02
2.54
201.89
0
186.92
76
0
0
0.00
0
3.52
0.02
2.26
196.09
0
186.92
77
0
11.48
0.00
0
1.73
0.02
2.77
203.06
0
175.44
78
0
22.69
14.73
0
1.12
0.02
2.77
207.11
0
152.75
Estimating Phosphorous Concentrations
Phosphorus delivery from a hillslope will either be adsorbed to eroded sediment, (particulate
phosphorus of PP) or be dissolved in surface runoff, subsurface lateral flow, or base flow (soluble
reactive phosphorus or SRP). Concentration of phosphorus in sediment depends on the mineralogy of
the soil and particle size. Phosphorus dissolved in solution depends on the geology and the flow
pathways (surface, lateral or base flow) that runoff follows.
For this paper, we will focus on developing concentration values that are typical of the Lake Tahoe Basin.
A similar procedure can be applied to other watersheds. Within the Tahoe Basin, there is a long history
of measuring phosphorus in streams. There are 13 major streams in the Lake Tahoe Basin with sample
density ranging from 129 samples collected from Trout Creek to 1414 at Incline Creek (Figure 4). These
samples were generally grab samples, often collected during peak flow times of the year. The
subsequent analyses reported total phosphorus (TP), and SRP. We assumed that particulate phosphorus
was the difference between TP and SRP. From these data, we were able to apply a USGS model,
LOADEST, developed to transform point data into continuous flow data as a function of stream flow.
With the LOADEST tool, we were able to estimate TP and SRP throughout the year for each of the Tahoe
streams.
9
Table 2. For a forest sandy loam soil, properties of sediment size classes in eroded
sediments in the WEPP model.
Class
1
2
3
4
5
Diameter
(mm)
0.002
0.010
0.030
0.300
0.200
Specific
Gravity
2.60
2.65
1.80
1.60
2.65
Sand
0.0
0.0
0.0
85.4
100.0
Particle Composition (%)
Silt
Clay
Organic
Matter
0.0
100.0
250.0
100.0
0.0
0.0
80.0
20.0
50.0
7.1
7.5
18.8
0.0
0.0
0.0
Using the observed samples, and the LOADEST software, we developed a model for phosphorus
concentrations throughout the year (Figure 2). Observed values tended to exceed LOADEST predictions
as they were often collected during large runoff events, and were not representative of the more
normal concentrations during each season or when averaged for a number of years.
In addition to the sample concentrations we also had the hydrographs for each of the streams that were
sampled. Figure 6 is a typical hydrograph where we have estimated the relative contribution of each of
the flow paths (surface, lateral and base) using the WEPP model water file. Figure 6 shows that the base
flow is the dominant flow path from July until snowmelt the following April, that surface runoff only
occurs at times of peak flow rates, and that lateral flow is the dominant flow path during higher flow
rates in the late spring. Combining this information with the results shown in Figure 5, it is apparent that
the SRP in surface runoff is likely less than 0.01 mg/L, whereas SRP concentrations in lateral flow and
base flow are likely to be around 0.02 mg/l. Concentrations are the lowest during March and April when
surface runoff is contributing to runoff and diluting lateral and base flow, but higher from June onwards
when lateral flow and base flow are the main sources of water in the stream system. Total phosphorus
delivered, however, is likely to be the highest during the peak flow times associated with snow melt in
April and May
In addition to TP and SRP, we had the suspended sediment concentrations of the samples that had been
collected from the streams in the basin. We were able to estimate the phosphorus concentration on
suspended sediments by dividing the calculated PP concentration by the suspended sediment
concentration. We found that the concentration varied within the Tahoe Basin depending on the
geology, mainly volcanic in the northwest and decomposed granite in the southeast part of the basin
(Figure 4).
Estimating Fine Sediment Delivery
In addition to phosphorus, stakeholders in the Lake Tahoe Basin are also concerned about fine sediment
delivery (Coats, 2004). In this context, “fine sediment” is generally considered to be sediment particles
and aggregates less than 10 to 20 microns in diameter. Such particles can remain suspended in lakes for
a considerable period of time as vertical currents due to wind shear and temperature gradients are
sufficient to prevent the particles from settling (Coats, 2004). It is these small particles combined with
increased algal growth because of phosphorus enrichment that have caused the lake to lose some of its
clarity in recent decades.
10
Figure 4. Concentration of phosphorus
(mg/kg) adsorbed to delivered
sediment from watersheds within the
Tahoe Basin.
Figure 5. Observed and predicted
phosphorus concentrations averaged
across all years for a typical watershed.
The distribution of particle size delivery from hillslopes or watersheds given in the WEPP output file
(Table 2) can be parsed to determine the amount of sediment in each particle size category by
summing the delivery of a given size primary particle with the fraction of that particle contained in the
aggregates. In WEPP, clay primary particles are less than 4 μ diameter, and silt particles are between 4
and 62.5 μ. To simplify modeling, we assumed that within the silt textural category that the distribution
of particle sizes was linear. Thus if the user needed to know the amount of sediment less than 10 μ, the
number could be determined by adding all of the clay fraction as primary particles and in aggregates to
the (10-2.5)/(62.5-2.5) fraction of the silt delivered as primary particles and in aggregates.
11
Figure 6. Example hydrograph based on WEPP hydrology for Blackwood Creek (Elliot et al., 2010)
Interfaces
In order to make this technology useful to mangers, an interface was developed similar to the Forest
Service Disturbed WEPP online Interface for the WEPP model (Elliot, 2004). Figure 7 shows the input and
output interfaces for the Tahoe Basin Sediment Model (TBSM). The user is asked to select a climate,
dominant geology, vegetation conditions for the upper or treated part of a hill, and lower or stream side
buffer part of the hill. In the case of an undisturbed condition, or a post wildfire condition, the upper
and lower portions of the hill may have the same vegetation.
The climate database for the TBSM includes one NOAA station within the Tahoe Basin as well as five
nearby climates. In addition, climate statistics have been added to the database for seven NRCS Snow
Telemtery (SNOTEL) stations located within the basin.
The user is also asked to provide the phosphorus concentrations in the surface runoff, lateral flow, and
sediment. Earlier versions of the interface were designed for the user to enter the phosphorus
concentration in the soil, and the model would then adjust this value using the specific surface
enrichment ratio from the WEPP output. We found, however, that it was easier to obtain the
concentration of total and soluble reactive phosphorous from instream monitoring rather than
concentrations of phosphorus in the soils themselves, so the current interface is designed to use the
concentrations shown in Figure 4 based on instream data. The interface could be altered for other
applications where onsite particulate phosphorus concentrations are readily available and be designed
to use delivered sediment with a delivery ratio as previously discussed.
Figure 7b shows the output screen for the TBSM. Each phosphorus path (sediment, surface runoff and
lateral flow) is presented so that the users will be able to determine the dominant pathway for the
condition they are modeling. In the example shown in Figure 7b for a prescribed burn with a buffer, the
greatest source of phosphorus is in the delivered sediment. This is often the case in disturbed forests
12
Figure 7a. Input Screen for the Tahoe
Basin Sediment Model using a
SNOTEL Station from within the
Tahoe Basin for the weather and
phosphorus concentrations from
Figures 4 and 5.
Figure 7b. Output screen for the
Tahoe Basin Sediment Model.
(Stednick, 2010). In undisturbed forests, the greatest source of SRP is likely to be in the shallow lateral
flow (Miller et al., 2005).
The fine sediment category between 4 and 62.5 μ is specified on the input page (Figure 7a) and the
total delivery per unit area is calculated from the predicted sediment delivery, and presented on the
output page.
Discussion
This approach to modeling phosphorus and fine sediment delivery was developed for the Lake Tahoe
Basin. The principals that are described here for estimating delivery of phosphorus can be applied to any
condition where the input variables are known. The TSPM interface assumes a PP concentration
attached to stream sediment. In other conditions, it may be more appropriate to link the PP
concentration to the onsite concentration, and apply a specific surface enrichment ratio to the delivered
sediment. With this interface, using the large PP concentrations in stream sediments (1000 – 2500
mg/kg), we may be over predicting the delivery of PP to the stream. Elliot et al. (2012) reported onsite
concentrations of 4-22 mg/kg, and concentrations on coarse sediments collected from rainfall
simulation of 160-475 mg/kg. The increasing concentrations of PP from soil to upland eroded sediments
to stream sediments is due to the specific surface enrichment, and further work on the interface may be
necessary to make sure the high instream concentrations are linked to the delivery of clay-size material.
In the Tahoe basin, clay generally accounts for around 2 percent of the soil fraction.
13
Another interesting hydrologic feature of coarse forest soils is that unless the soils are highly disturbed,
there is little surface runoff. Comparing the hydrograph in Figure 3 to the SRP concentration variability
in Figure 6 suggests that when surface runoff does occur, SRP concentrations are low, but when lateral
flow or subsurface flow dominate the runoff, SRP concentrations increase. The net effect of integrating
the runoff and concentrations values in these two figures suggests that total SRP delivery is the greatest
when runoff is the greatest. It also suggests an interesting twist to managers: if managers seek to
minimize surface runoff, lateral flow is likely to increase (Srivastava, 2013), and so will the concentration
of SRP leaving the hillslope. Surface runoff itself will deliver less SRP, but it will also be the mechanism
that delivers sediment, and may then dominate the TP budget.
The TBSM does not consider channel processes. In steep forest watersheds, stream channels and banks
tend to be coarse, having minimal impact on adsorbing or desorbing TP. Forests with finer textured or
higher organic materials in stream beds or banks are more likely to have a moderating influence on TP
delivery, adding to SRP during times of low stream SRP concentration and reducing SRP during times of
high concentration. Work is ongoing with a SNPLMA Round 12 project to incorporate some of these
channel processes.
The interface clearly shows the link between sediment delivery and TP delivery. Past watershed research
has shown that sediment budgets from forest watersheds are dominated by wildfire, with sediment
delivery following wildfire as much as 100 times greater than associated with undisturbed forests (Elliot,
2013). Such sediment pulses will likely dominate delivery of phosphorus in the same way as they
dominate the sediment budget. Managers will need to consider the effects of management practices
not only on immediate sediment delivery, but also on the effects that management may have on
reducing the probability or severity of wildfires because of those activities (Elliot, 2013).
Workshops
A one‐day training workshop was held at the Tahoe Regional Planning Agency on November 14, 2012,
which was well‐attended by regulators, land managers, and consultants. A list of those that attended is
presented as Table 3. Personnel from the USFS Lake Tahoe Basin Management Unit were unable to
attend the 2012 workshop; therefore, a second training workshop was held for USFS personnel on
March 14, 2013 at the USFS Lake Tahoe Basin Management Unit. A list of attendees for the 2013
workshop is presented as Table 4. Both were all‐day workshops. Attendees were first introduced to the
WEPP model and the TBSM interface through a Power Point presentation. Attendees then ran TBSM
simulations on‐line by working through worksheet exercises that applied TBSM to upland forest
scenarios that included undisturbed forest, fuels management, roads/road decommissioning, cut and fill
slopes, harvest treatments, ski areas, and snow‐making corridors. Worksheet examples were designed
specifically to demonstrate to attendees how TBSM can be applied to evaluate baseline conditions and
alternative treatment scenarios on fine sediment delivery.
User Guide
The TBSM User Guide was drafted, with worksheet exercises from the training workshops included
as appendices to assist users. It is currently under review.
14
Table 3 TBSM Workshop Attendees at the November 14, 2012 Workshop at the Tahoe Regional Planning
Agency.
Name
Cushman, Doug
Byrne, Bryan
Dugan, Maxwell
Hagan, Tim
Hirt, Brian
Knust, Andrew
Landy, Jack
Pilgrim, Kevin
Praul, Chad
Sribe, Laurie
Stucky, Daniel
Van Huysen, Tiffany
Vollmer, Mike
Walck, Cyndie
Agency/Company
LRWQCB
UNR
UNR
Carndo ENTRIX Consulting
Califorina Tahoe Conservancy
NDOT
EPA
USFS RMRS
Environmental Incentives Consulting
LRWQCB
Atkins Consulting
USFS
TRPA
CA State Parks
Table 4. 2013 TBSM Training Workshop Attendees (March 14, 2013 at USFS Lake Tahoe Basin
Management Unit).
Name
Agency/Company
Craig Oehuti
USFS
Nicole Brill
USFS
Barbara Shanley
USFS
Genevieve Villamaire
USFS
Melanie Greene
Hauge Brueck & Associates
Technical Presentations
Elliot, B. Brooks, E., Traeumer, D., Bruner, E. (2012) Predicting phosphorus from forested areas in the
Tahoe Basin. Presented at the Conference on Environmental Restoration in a Changing Climate,
Tahoe Science Conference, 22-24 May, 2012, Incline Village, NV.
References
Brooks, E.S., W.J. Elliot, J. Boll, and J. Wu. 2010. Final Report, Assessing the Sources and Tr4ansport of
Fine Sediment in Response to Management Practices in the Tahoe Basin using the WEPP Model.
Final Report, Round 7 SNPLMA, submitted USDA-FS Lake Tahoe Basin Management Unit. Accessed
on 9/28/2012 at
http://www.fs.fed.us/psw/partnerships/tahoescience/documents/final_rpts/P005FinalReportJan2
011.pdf
Coats, R. 2004. Nutrient and sediment transport in streams of the Lake Tahoe Basin: A 30-year
restrospective, pages 143-147. USDA Forest Service Gen. Tech. Rep. PSW-GTR-193. USDA Forest
Service Pacific Southwest Research Station.
15
Coats, R. J. Reuter, M. Dettinge, J. Riverson, G. Sahoo, G. Schladow, B. Wolfe, M. Costa-Cabral. 2010. The
effects of climate change on Lake Tahoe in the 21st Century: Meteorology, Hydrology, Loading,
and Lake Response. Final Report, Round 8 SNPLMA, submitted USDA-FS Lake Tahoe Basin
Management Unit. Accessed on 9/28/2012 at
http://terc.ucdavis.edu/publications/P030Climate_Change_Project_Final_Report_2010.pdf
Copeland, N.S. and R.B. Foltz. 2009. Improving erosion modeling on forest roads in the Lake Tahoe
Basin: Small plot rainfall simulations to determine sateruated hydraulic conductivity and interrill
erodiblity. Presented at the Annual International Meeting of the American Soc. of Ag. & Bio.
Engrs, 21-24 June, Reno, NV. St. Joseph, MI: ASABE. 11 p.
Dun, S., J.Q. Wu, W.J. Elliot, P.R. Robichaud, D.C. Flanagan, J.R. Frankenberger, R.E. Brown and A.D. Xu.
2009. Adapting the Water Erosion Prediction Project (WEPP) Model for forest applications.
Journal of Hydrology, 336(1-4):45-54.
Elliot W.J. (2004) WEPP internet interfaces for forest erosion prediction. Jour. of the American Water
Resources Association 40, 299-309.
Elliot, WJ (2013) Erosion processes and prediction with WEPP technology in forests in the Northwestern
U.S. Trans ASABE. 56(2): 563-579.
Elliot, W. J., A. M. Liebenow, J. M. Laflen and K. D. Kohl. 1989. A compendium of soil erodibility data
from WEPP cropland soil field erodibility experiments 1987 & 88. Report No. 3. USDA-ARS,
National Soil Erosion Research Laboratory, W. Lafayette, IN. 316 p.
Elliot, B. Brooks, E., Traeumer, D., Bruner, E. (2012) Predicting phosphorus from forested areas in the
Tahoe Basin. Presented at the Conference on Environmental Restoration in a Changing Climate,
Tahoe Science Conference, 22-24 May, 2012, Incline Village, NV.
Elliot, W., E. Brooks, T. Link, S. Miller. 2010. Incorporating groundwater flow into the WEPP model.
Procs. 2nd Joint Federal Interagency Conference, 17 June – 1 July, Las Vegas, NV. 12 p
Flanagan, D. C. and Livingston, S. J. (eds.). (1995) WEPP User Summary. NSERL Report No. 11. USDA
Agriculture Research Service, National Soil Erosion Research Laboratory, W. Lafayette, IN. USA.
131 p.
Flanagan D.C. and Nearing, M.A. (1995) ‘USDA – Water Erosion Prediction Project: hillslope profile and
watershed model documentation.’ USDA-ARS National Soil Erosion Research Laboratory, NSERL
Report No. 10. (West Lafayette, Indiana)
Foltz, R.B., H. Rhee and W.J. Elliot. 2008. Modeling changes in rill erodibility and critical shear stress on
native surface roads. Hydrological Processes. DOI: 10.1002/hyp.7092.
Foster, G. R. and L. J. Lane, compilers. 1987. User Requirements: USDA-Water Erosion Prediction
Project (WEPP). NSERL Report No. 1. W. Lafayette, IN: USDA-ARS, National Soil Erosion Research
Laboratory.
Grismer, M.E., A.L. Ellis and A. Fristensky. 2008. Runoff sediment particle sizes associated with soil
erosion in the Lake Tahoe Basin, USA. Land Degrad. Develop. 19: 331-350.
16
Grismer, M.E. and M.P. Hogan. 2005. Simulated ranfall evaluation of revegetation/mulch erosion
control in the Lake Tahoe Basin - 3: Soil treatment effects. Land Degrad. Develop. 13:1-13.
Laflen JM, Elliot WJ, Flanagan DC, Meyer CR, Nearing MA (1997) WEPP-predicting water erosion using a
process-based model. Jour. of Soil and Water Conservation 52(2), 96-102.
Luce, C. H., and T. A. Black. 1999. Sediment production from forest roads in western Oregon. Water
Resources Research 35(8): 2561-2570.
Miller, W.W., D.W. Johnson, C. Denton, P.S.J. Verburg, G.L. Dana and R.F. Walker. 2005. Inconspicuous
nutrient laden surface runoff from mature forest Sierran watersheds. Water, Air and Soil Pollution
163:3-17.
Robichaud, P. R. 1996. Spatially-varied erosion potential from harvested hillslopes after prescribed fire
in the interior northwest. Ph. D. dissertation. Moscow, ID:University of Idaho. 219 pp.
Srivastava, A. 2013. Modeling of hydrological processes in three mountainous watersheds in the U.S.
Pacific Northwest. PhD Dissertation. Pullman, WA: Washington State University. 170 p.
Srivastava, A., M. Dobre, J. Wu, W. Elliot, E. Bruner, S. Dun, E. Brooks, and I. Miller. 2013. Modifying
WEPP to improve streamflow simulation in a Pacific Northwest Watershed. Trans. ASABE 56(2):
603-611.
Stednick, J.D. (2010) Effects of fuel management practices on water quality, Chapter 8. 149-163. In Elliot,
W.J., I.S. Miller and L. Audin (eds.). Cumulative Watershed Effects of Fuel Management in the
Western U.S. Gen. Tech. Rep. RMRS-GTR-231. Fort Collins, CO: U.S. Dept. of Agric., Forest Service,
Rocky Mountain Research Station.
Wagenbrenner, J. W., P. R. Robichaud, and W. J. Elliot (2010), Rill erosion in natural and disturbed
forests: 2. Modeling Approaches, Water Resour. Res., 46, W10507, doi:10.1029/2009WR008315.
Wikipedia. 2010. Albedo. Online at <http://en.wikipedia.org/wiki/Albedo>. Accessed May, 2010.
17
Appendix 1. Soil Database for the Tahoe Basin Sediment Model
The variables shown here are in the same format, with the same units as described in the WEPP User
Summary (Flanagan and Livingston, 1995)
# Soil parameter database for Tahoe Basin model -- modified from Disturbed WEPP
#
by W Elliot, May, 2010
#
# Granitic
#
'Skid' 'Granitic' 1
0.2 0.75 2700000 0.001 4
10
300 90
2 4
3 rfg
'HighF' 'Granitic' 1
0.1 0.75 1800000 0.0005 4
15
300 90
2 3
3 rfg
'LowF' 'Granitic' 1
0.15 0.75 1000000 0.0003 4
20
300 90
2 4
3 rfg
'Sod_G' 'Granitic' 1
0.15 0.75 750000 0.0001 4
25
350 90
2 4
3 rfg
'BnchG' 'Granitic' 1
0.15 0.75 600000 0.00008 4
30
400 90
2 5
3 rfg
'Shrub' 'Granitic' 1
0.15 0.75 500000 0.00006 4
35
500 90
2 5
3.5 rfg
'YForst''Granitic' 1
0.1 0.5 400000 0.00004 4
40
600 90
2 5
4 rfg
'OForst''Granitic' 1
0.1 0.5 250000 0.00003 4
45
800 90
2 6
4 rfg
'Bare' 'Granitic' 1
0.2 0.75 300000 0.001 4
25
300 90
2 2
2 rfg
'Mulch' 'Granitic' 1
0.2 0.75 300000 0.001 4
30
400 90
2 4
2.5 rfg
'Till' 'Granitic' 1
0.15 0.75 300000 0.001 4
35
500 90
2 6
3 rfg
'Lowr' 'Granitic' 1
0.2 0.75 225000 0.0013 4
10
200 90
2 1
2 rfg
'Highr' 'Granitic' 1
0.2 0.75 900000 0.005 4
10
200 90
2 1
2 rfg
#
#
# Volcanic
#
'Skid' 'Volcanic' 1
0.2 0.75 3000000 0.0008 1.5 8
300 65
7 4
8 rfg
'HighF' 'Volcanic' 1
0.1 0.75 2000000 0.0004 1.5 10
300 65
7 3
7 rfg
'LowF' 'Volcanic' 1
0.15 0.75 1500000 0.0002 1.5 15
300 65
7 4
7 rfg
'Sod_G' 'Volcanic' 1
0.15 0.75 1000000 0.00008 1.5 20
350 65
7 4
8 rfg
18
'BnchG' 'Volcanic' 1
0.15 0.75 900000 0.00006 1.5 25
400 65
7 5
8 rfg
'Shrub' 'Volcanic' 1
0.15 0.75 800000 0.00004 1.5 30
500 65
7 5
8.5 rfg
'YForst''Volcanic' 1
0.1 0.5 700000 0.00003 1.5 35
600 65
7 5
9 rfg
'OForst''Volcanic' 1
0.1 0.5 600000 0.00002 1.5 40
800 65
7 6
9 rfg
'Bare' 'Volcanic' 1
0.2 0.75 750000 0.0008 1.5 20
300 65
7 2
7 rfg
'Mulch' 'Volcanic' 1
0.2 0.75 750000 0.0008 1.5 25
400 65
7 4
7.5 rfg
'Till' 'Volcanic' 1
0.15 0.75 750000 0.0008 1.5 30
500 65
7 6
7.5 rfg
'Lowr' 'Volcanic' 1
0.2 0.75 250000 0.001 1.5 8
200 65
7 1
7 rfg
'Highr' 'Volcanic' 1
0.2 0.75 1000000 0.004 1.5 8
200 65
7 1
7 rfg
#
#
# Alluvial
#
'Skid' 'Alluvial' 1
0.2 0.75 2500000 0.0006 1
6
400 60 10 5
12 rfg
'HighF' 'Alluvial' 1
0.1 0.75 1500000 0.0003 1
8
400 60 10 4
11 rfg
'LowF' 'Alluvial' 1
0.15 0.75 1000000 0.0002 1
10
400 60 10 5
21 rfg
'Sod_G' 'Alluvial' 1
0.15 0.75 900000 0.00006 1
15
450 60 10 5
22 rfg
'BnchG' 'Alluvial' 1
0.15 0.75 800000 0.00005 1
20
500 60 10 6
12 rfg
'Shrub' 'Alluvial' 1
0.15 0.75 700000 0.00003 1
25
600 60 10 6
13 rfg
'YForst''Alluvial' 1
0.1 0.5 600000 0.00002 1
30
700 60 10 6
13 rfg
'OForst' 'Alluvial' 1
0.1 0.5 500000 0.00001 1
35
900 60 10 7
13 rfg
'Bare' 'Alluvial' 1
0.2 0.75 600000 0.0006 1
15
400 60 10 3
10 rfg
'Mulch' 'Alluvial' 1
0.2 0.75 600000 0.0006 1
20
500 60 10 5
12 rfg
'Till' 'Alluvial' 1
0.15 0.75 600000 0.0006 1
25
600 60 10 6
13 rfg
'Lowr' 'Alluvial' 1
0.2 0.75 240000 0.0008 1
6
200 60 10 1
10 rfg
'Highr' 'Alluvial' 1
0.2 0.75 950000 0.003 1
6
200 60 10 1
10 rfg
19
#
#
# Rock/Pavement
#
'Skid' 'Rock/Pave' 1
0.2 0.75 100 0.000001 20 0.1
300 60
5 1
4
rfg
'HighF' 'Rock/Pave' 1
0.1 0.75 100 0.000001 20 0.1
300 60
5 1
4
rfg
'LowF' 'Rock/Pave' 1
0.15 0.75 100 0.000001 20 0.1
300 60
5 1
4
rfg
'Sod_G' 'Rock/Pave' 1
0.15 0.75 100 0.000001 20 0.1
300 60
5 1
4
rfg
'BnchG' 'Rock/Pave' 1
0.15 0.75 100 0.000001 20 0.1
300 60
5 1
4
rfg
'Shrub' 'Rock/Pave' 1
0.15 0.75 100 0.000001 20 0.1
300 60
5 1
5
rfg
'YForst' 'Rock/Pave' Does not exist
00
'OForst' 'Rock/Pave' Does not exist
00
'Bare' 'Rock/Pave' 1
0.2 0.75 100 0.000001 20 0.1
300 60
5 1
4
rfg
'Mulch 'Rock/Pave' 1
0.2 0.75 100 0.000001 20 0.1
300 60
5 1
4
rfg
'Till' 'Rock/Pave' 1
0.15 0.75 100 0.000001 20 0.1
300 60
5 1
5
rfg
'Lowr' 'Rock/Pave' 1
0.2 0.75 100 0.000001 20 0.1
200 60
5 1
4
rfg
'Highr' 'Rock/Pave' 1
0.2 0.75 100 0.000001 20 0.1
200 60
5 1
4
rfg
#
# rfg = 0.5 for all Rock/Pavement soils
# Lower element for Rock/Pave to be Alluvial in all cases
20