Quick Start Reference Guide

Web Browser

Modern browsers including Firefox, Chrome, Safari, and Edge should work with WEPPcloud. For best results use a laptop/desktop sized display. Internet Explorer is not supported

User Accounts (Optional)

WEPPcloud allows users to create and store model runs. Users can also create accounts to save runs and collaborate on model runs with other users. By default, anonymous runs are always publicly available via the URL to the project. Be aware that any resources, such as soil burn severity maps, that are uploaded to the interface would be potentially publicly available. If logged in the created runs are private by default and are only available to the project owner and collaborators. Private runs can be made publicly available if collaborators wish to share them. It is suggested that the runs also be made “Read Only” to avoid having viewers accidentally alter the model results.

User accounts can be created here: https://wepp.cloud/weppcloud/register

GPDR: We ask for your first and last name to provide polite coorespondances. Your email is used as a unique indentifier for your account. None of your personal identifiable information is shared with third-parties.

Creating a Project

WEPPcloud projects are also referred to as runs, as in model runs. To start a project click on the “Start Disturbed Run (CONUS)” button. This creates a new project folder on the server.

RunID and Interface Config

The newly created project has a unique RunID, underlined in red. This RunID is unique for every project and can be used to return to the project as well as for debugging purposes. In the cloud a project folder is created with this RunID. The run ids are generated using a package called “Awesome-Codenames” and has a dictionary of 26,000 words. The codenames are used to produce ids that are memorable and easily transcribable.

A large array of functionality is provided by WEPPcloud through the use of interface configurations. Each interface has a configuration file (“config file”) that specifies datasets and parameters for specific locations or models. The config is specified after the RunID in the URL. In the image below it is underlined in blue. The “disturbed9002” config is for the US.

Selecting Map Extent

The first step is to specify the boundary for the catchment you wish to model. The extent of the map is used as the boundary for modeling. The map has controls to zoom in and out. It is also possible to double click on the map to zoom in. The map can be panned by clicking and dragging. The extent can also be specified by holding down shift will drawing a box. This is especially helpful when zoomed out.

Channel Delineation

The channel delineation is limited in the size of catchment areas that can be delineated. The limit is dependent on the TOPAZ processing parameters for Minimum Channel Length (MCL) and Critical Subcatchment Area (CSA). For this reason, it is suggested that you zoom into a level of 13 or greater.

When setting the extent it is important that the ridgeline of the catchment you wish to model is within the boundary of the map. TOPAZ will fail if the catchment falls outside of the map when delineating subcatchments.

Optional SBS Upload

A Soil Burn Severity Map can be uploaded. The SBS maps should be georeferenced .tif or .img files. They should have four thematic SBS classes: no burn, low, moderate, and high burn severity or be BARC 256 maps. The maps should have a single data band formatted as uint8 (byte) datatypes. Maps with color tables will use the table to identify SBS classes.

Maps without color tables will allow the user to specify breaks to define the soil burn severity classes.

Limited support is provided for maps with float32 or float64 data if they contain integer values. If non-integer values are identified the SBS Upload will intentionally Error to reject the map.

TOPAZ

WEPPCloud uses TOPAZ to parameterize topographic data from DEMs to create hillslope profiles called sub-catchments for each watershed. TOPAZ delineates a channel network from the DEM based on the steepest downslope path from each raster cell (pixel) from the 8 cells surrounding it (Garbrecht and Martz 1997). Adjustments can be made to the detail of the channel network by changing values of Mean Source Channel Length (MSCL) and Critical Source Area (CSA). The MSCL is the shortest length that any channel is allowed to be. The CSA defines the minimum drainage area below which a permanent channel forms (Garbrecht and Martz 1997). Setting these to low values will increase the density of channels, which is useful when defining small watersheds. From the defined channel network, the user specifies the exact watershed outlet and TOPAZ generates the sub-catchments that delineate the watershed. Each sub-catchment represents the direct contributing area for each side of the drainage (Garbrecht and Martz 1997) and has homogeneous slope and aspect (Renschler 2002).

It is highly recommended to not exceed an MSCL of 100m and a CSA of 10 ha

Minimum Source Channel Length (MSCL)

The shortest length that any channel is allowed to be. The default value is set to 100 m, which means that atleast three 30-m DEM pixel is needed to initiate the headwater channel.

Critical Source Area (CSA)

The minimum drainage area below which a permanent channel forms. The default value is set to 10 ha.

Repeatable Delineation

The watershed will delineate differently if the map extent, TOPAZ parameters, or outlet location are changed. To get the same watershed to delineate you can manually set the center location of the map and zoom level as well as specify the longitude and latitude of the outlet. The map size (in pixels) is also dependent on the window size of the browser and the browser zoom. So to repeatably delineate a watershed all of these should be the same. For comparing between scenarios it is recommended to fork projects or use the Omni Scenario Runner functionality to maintain delineation between comparisons.

Outlet

The outlet for the watershed can be defined by pressing “Use Cursor” and using the map to select the watershed outlet location. WEPPcloud will identify the nearest channel, place a marker on the map, and report the outlet location in the Outlet control. User’s can also specify the outlet using coordinates.

Subcatchment Delineation

Subcatchments whether delineated by TOPAZ or TauDEM will always have a TOPAZ Identifier as well as a WEPP Identifier. The TOPAZ identifiers (ids) are assigned in a clockwise fashion starting at the outlet. TOPAZ uses a numbering scheme where all the subcatchments end with “1,” “2,” or “3” and all the channels end with “4.” Subcatchments to the right of a junction end with “2,” and subcatchments to the left and with “3.” Subcatchments that drain into a channel end with “1.” Subcatchments and channels that are closer to the outlet typically have lower numbers.

The WEPP structure file requires sequential numbering and hence the need for having both TOPAZ identifiers and WEPP identifiers.

Watershed Touches Boundary Error

If the watershed extends beyond the extent of the map the Subcatchment Delineation will fail.

In this case the map extent/zoom should be adjusted so that the entire watershed is within the map. Channels and outlet must be re-done before attempting the subcatchment delineation.

Managements

Readily available managements for WEPP use are provided in the drop-down menu. The interface contains NLCD maps from 1985 - 2023. For SBS scenarios it is recommended to select the year prior to the fire. In some cases user’s might want to refine managements for better representation of croplands and forestland practices. No Data and Water classes should be reassigned.

Note: The current interface does not have the capability to build/edit managements for simplicity purpose. Future versions will include include this capability.

Two options are provided for management selection.

<dtDetermine by hillslope</dt>

Automatically determines management per hillslope.

Single Landuse for watershed

Manually select a management from the drop-down list.

Modifying landuses

Some interfaces allow the user to apply treatments through the map at the top of the screen. First build the managements. The landuse covers will be assigned from the management database. The cover values can be modified from the default values by clicking the gear icon. Below the selected landuse a dialog should appear to allow for setting groud cover, interill, and rill cover values. The values are saved when the selection is made.

The landuses can be viewed on the map by selecting the “Dominant Landcover” layer on the layer tab to the right of the map.

Users can also supply their own landcover maps by selecting the “Upload Landcover Map” radio and using the file selection. The maps should be single band .tif maps with classes from the landuse map file. The disturbed9002 interface uses this landuse mapping file.

(Disturbed.json Landuse Map)[https://github.com/rogerlew/wepppy/blob/master/wepppy/wepp/management/data/disturbed.json]

Soils

(US) Two databases for soils are available in WEPPcloud: SSURGO and STATSGO. The SSURGO database is used as default and is queried during the watershed delineation process. Alternatively, when SSURGO database is not available, STATSGO database will be used to build the soil files.

Three options are provided for determining soils.

<dtDetermine by hillslope</dt>

Single Soil for watershed (MUKEY)

Assign one soil for the entire watershed based on the SSURGO/STATSGO MUKEY.

Single Soil for Watershed (Database)

Select a soil for the entire watershed from a local database.

Most of the time WEPPcloud builds WEPP soil files (7778 format) parameterized from the USDA’s SSURGO/STATSGO database (for US Locale; not with WEPP-PEP or RRED). After building a summary table is provided for reviewing the soils within the catchment. The soils file names are taken from the SSURGO/STATSGO MUKEYs.

The soil building also applies parameters from the (un)disturbed land soil lookup.

Climate

The climate input processing relies heavily on CLIGEN.

WEPP uses CLIGEN climate files as input. CLIGEN is a climate generator capable of generating stochastic daily climates, single storm events, or observed climates from daily temperature minimum, maximum, and precipitation series.

CLIGEN uses station parameter or (.par) files. We use the 2016 NSERL CLIGEN database of weather stations with monthly parameters for US Locales and a GHCN based database has global coverage for Europe and Australia. The first step in building a climate is to select a weather station as a basis for the climate files. The recommended method to select a climate station is to select the “Multi-Factor Ranking (Considers Distance, Elevation, and Climate)” radio button. Once selected, WEPPcloud evaluates 2400+ stations to find the best match based on distance, elevation, and monthly climate values from PRISM. After a few seconds the best match should appear in the selection dialog, and summary climate statistics should appear below the selected station.

The selection control lists the 10 best matching stations in descending order. After the station description in parenthesis is an ordinal ranking metric (lower is better) and the distance from the centroid of the catchment. The user can select any station in the list.

CLIGEN is a stochastic weather generator but also supports generating daily timeseries climate files if daily minimum and maximum temperatures and daily precipitations are available. CLIGEN also has the ability to produce single storm climate files Please note that the WEPP outputs for Single Storm Events are different from the Continuous outputs and not all of post wepp functionality including WATAR

Climate Modes

Stochastic PRISM Modified

Modifies a station parameter file with monthly precipitation, minimum and maximum daily temperatures. 800m grid. After modifing the station parameter file CLIGEN is used to generate a daily timeseries of weather events. Recommended for BAER analysis not requiring historical comparisons.

“Vanilla” CLIGEN (Stochastic weather generation).

The daily climate is generated using the CLIGEN (CLImate GENerator) and the nearest parameter (.PAR) file. Values in .PAR files includes monthly statistical values derived using the NCDC-NOAA weather stations from 1974 through 2013. For interntional locations this may be the only option that is functional.

Observed DAYMET (GRIDMET wind)

Utilizes daily precipitation, minimum and maximum temperature, dewpoint, and solar radiation from DAYMET and daily wind speed and velocity from GRIDMET. DAYMET 1km, GRIDMET 4km. Recommended for historical analysis. Storm parameters (duration, time to peak) are generated from CLIGEN and may not be accurate to historic events.

Observed GRIDMET

Daily observed precipitation, maximum and minimum temperatures, dewpoint, solar radiation, and wind are obtained from the data source. 4km grid. This dataset is available from 1980-present. Storm parameters (duration, time to peak) are generated from CLIGEN and may not be accurate to historic events.

Future (CMIP5)

4km spatial resolution daily observed precipitation, maximum and minimum temperatures are obtained from the data source. This dataset is available from 2006-2099. Other required weather inputs in the climate files are generated using nearest CLIGEN station.

Single Storm

User-defined storm for event-wise simulation

Storm Amount

Enter the precipitation depth in inches or in millimeters.

Storm Duration

Enter the duration of the storm (in hours).

% Duration to Peak Intensity

Enter the time (in percent of the storm duration value) at which the peak intensity of the storm occurs. Note that this is not a time value, but a percentage of the total duration value which was entered in the previous field. For instance, if the peak intensity occurs at hour 4 of a 10-hour storm, then you would enter 40 (%) or 0.4 here.

Max Intensity

Enter the peak rainfall intensity of the storm in inches per hour or millimeters per hour.

Climate Spatial Mdes

Single Climate

Applies the same climate file to all the hillslopes and watershed

Multiple PRISM Revision

Uses Interpolated PRISM monthly normals to adjust climates for each hillslope.

Multiple Interpolated (Slow)

For observed DAYMET and GRIDMENT this option generates a daily interpolated climate for each hillslope.

Watershed Preparation Report

Watershed Preparation details describes hillslopes and channels associations with the aforementioned four inputs climate, management, soil, and slope. The table described here might also be useful for referencing variables/values in the output reports with subcatchment delineation map.

WEPP Advanced Options

Flowpath Processing

WEPP is ran for each flowpath in the subcatchments. The plot files are used to produce a gridded soil loss/deposition map. This is only available with the TOPAZ delineation backend.

Slope definitions

The flowpath slope definitions are constructed by watershed_abstraction routine in wepppy. For each cell in a hillslope a flowpath is constructed by walking down the slope identified by the slope grid from topaz.

Clip Hillslopes

WEPP can overpredict erosion with long hillslopes. Clipping will restrict the length and maintain area by increasing the width of the hillslope.

Soils, Managements, and Climates

Flowpath runs use the soil, management, and climate of their parent hillslope.

Gridded Output

Each flowpath is ran through wepp and a plot file is generated by wepp. This file contains distance downslope, elevation, and soil deposition/loss (loss is positive). The soil dep/loss is interpolated against the normalized distance points from the flowpaths slope file to obtain a estimate at each pixel value. Many flowpaths might run through a particular cell. The dep/loss estimates are aggregated by taking the average of the non-zero values at each cell.

Baseflow Processing

Daily simulated streamflow at the outlet of a watershed is composed of surface runoff, lateral flow (i.e. shallow subsurface stormflow) and baseflow. Baseflow is assumed to be routed to the stream outlet point from an aquifer network with a volumetric discharge following linear reservoir theory. The single aquifer is assumed to receive all water which vertically percolates below all upland hillslopes. It is assumed that the recharge to the aquifer is limited by the upslope drainage area defined by the surface topography and that there is no lateral groundwater exchange across watershed boundaries. Following linear reservoir theory the baseflow from an aquifer (Qbase in mm/day) is calculated as a fixed percentage (k, the baseflow coefficient having units of 1/days) of the amount of water stored in the aquifer (S is total water volume in the aquifer in units of mm) using the following equation:

Q_base = k * S

The linear coefficient can be best estimated by slope of the ln (time) vs ln (flow) hydrograph during drought flow conditions. The default parameter for the linear coefficient is k = 0.04 1/day.

In cases where the groundwater aquifer leaks water to either a deeper more regional aquifer or laterally across the watershed boundary or below the stream gage then users can optionally define a secondary aquifer loss (Qaq) based on second deep seepage linear coefficient (kds in 1/day units) based on the same storage amount S.

Q_aq = k_ds * S

By default, the model assumes there is no other deep aquifer loss and therefore assumes Kds = 0. The amount of groundwater storage in the aquifer on any given day (i) is determined based on the follow equation.

S_i = S_i-1 - Q_base,i – Q_aq,i

where S_i-1 is the groundwater storage volume from the previous day in the simulation. The model assumes the initial amount of water storage in the aquifer on the first day of the simulation is 200 mm.

Baseflow is simulated both at the watershed outlet and within the stream network if the upslope area above a point in the stream network is greater than user defined critical watershed area. The default critical area is 1.0 ha. This value should be assigned a larger magnitude in more arid locations.

Pollutant Processing

This is an optional selection that allows users to estimate a pollutant based on known concentrations provided by the user. The estimates are simplistic and do not account for sorption, desorption, mineralization, or immobilization. They only provide rough estimates of the pollutant being transported by both water (via surface runoff, lateral flow, and baseflow) and sediments.

FAO Penman-Monteith

FAO 56 Penman-Monteith method for reference and actual evapotranspiration (ET) developed by Allen et al. (1998) (Wu and Dun, 2004)

Run WEPP

Once the user selects all the soil, management, and climate options, all the necessary input files are automatically created for the user. At this stage the user can run the WEPP model, which will wake anywhere between a few seconds, for smaller watersheds and several minutes or even an hour, for larger watersheds. The status bar should indicate the hillslopes being ran. Once successfully completed, the summary bar will provide information of the number of hillslopes ran and elapsed time. If one or more input files are missing, the interface will error, alerting the user of the missing information.

WEPP Results

Watershed Loss Summary

The Watershed Loss Summary output frame provides average annual simulated hydrologic and sediment yield metrics in three accessible tables from the watershed outlet, each hillslope (i.e. subcatchment) in the watershed, and each stream channel in the watershed, respectively.

The watershed loss results at the outlet include both absolute mass and volume-based metrics as well as normalized mass and volume output per unit upslope watershed area including:

• average annual precipitation
• water discharge (streamflow from the watershed outlet)
• total hillslope sediment loss
• total channel sediment loss
• sediment discharge at the watershed outlet (net sediment yield delivered to the watershed outlet)
• an overall sediment deliver ratio defined as the total sediment loss from the hillslopes divided by the total sediment discharge at the watershed outlet

The Average Annual Summary for Subcatchments table provides information on:

• hillslope id (both WEPP and Topaz based ID numbers)
• landuse ID number
• soil MUKEY ID
• hillslope length
• hillslope area
• runoff
• soil loss (i.e. detachment or erosion rate mass/hillslope area)
• sediment deposition
• sediment yield (i.e. mass of sediment delivered to the outlet of the hillslope)

The Average Annual Summary for Channels table provides hydrologic and sediment transport metrics for each channel

Return Periods Report

Understanding the variability and risk of soil erosion is critical for forest managements. This section reports return period assessment for precipitation depth, runoff, peak discharge, 10-min and 30-min peak intensity, and sediment yield using the Weibull plotting position method based on annual maximum series.

Sediment Delivery Report

Provides information regarding the delivery of sediment at the outlet, from the channels, and from the hillslopes. This report also provides class particle size fractions.

Water Balance Report

The Average Annual Report provides major water balance components of each hillslope:

• precipitation
• ET (transpiration + evapotranspiration)
• surface runoff
• subsurface lateral flow
• deep seepage

Daily Streamflow (Runoff + Subsurface Lateral Flow + Baseflow) Graph

This figure shows the contribution of daily runoff, subsurface lateral flow, and baseflow to streamflow averaged over hillslopes in a watershed.

Export Functionality

Download project

Zip Archive directly from interface

Once the WEPP model runs are completed, the user can download all the input and output files as a zip archive, which contains several folders with all the maps and data used in the analysis. This functionality requires a reliable internet conneciton and can be somewhat hit or miss with large projects. For large project or unreliable connections use the wepppy-win-bootstrap download_project script.

Using wget

Projects can be acquired using wget. wepppy-win-bootstrap provides a script and instructions on downloading projects with wget.

Project Directory Structure

Description of all the available files and folders:

climate (dir)

contains all the PAR files and .cli files for each hillslope and for watershed.

dem (dir)

contains the DEM map used in the analysis as well as a topaz sub directory with all the maps created during the watershed delineation.

export (dir)

channels and subcatchments files in ArcGIS format containing topographic characteristics (such as slope, aspect, or length), input data (soil and management), and output information (runoff, lateral flow, baseflow, sediment, pollutant, etc.). The file also contains the several GeoTIFF maps used in the model run.

landuse (dir)

contains landuse map (e.g. ascii map with the 2016 National Land Cover Database (NLCD) for US Locale. The NLCD codes are translated into WEPP-equivalent managements based the mapping for the configuration.)

observed (dir)

files containing the observed data (if) provided by the user

watershed (dir)

files containing slope information for each channel and hillslope

wepp/flowpaths (dir)

if the flowpath option is selected, the WEPP model will be run for each map pixel. This folder contains a runs folder with all the input data and an output folder with the runoff and soil loss for each flowpath

wepp/output (dir)

contains all the WEPP model output files

wepp/plots (dir)

maps of gridded soil loss when flowpaths are ran

wepp/runs (dir)

folder containing all the WEPP model input files

*.nodb

JSON serialized instances of wepppy.nodb classes used by WEPPcloud. These contain metadata related to the project. They are viewable in FireFox/Notepad++ etc.

More on NoDB Files

WEPPcloud project runs are designed to be self-contained and to not rely on a centralized database. This is accomplished by storing model run information in JSON (JavaScript Object Notation) files. These are the .nodb files in the project’s root folder. The .nodb files are serialized Python class instances that greatly simplifies modeling from a programmatic perspective. Even for non-programmers these files can be informative. Loading the topaz.nodb shows the following:

Daily Water Balance Report

The totalwatsed2.csv file contains all the water balance components and sediment yield combined across the hillslopes in a watershed. The report also has daily streamflow and sediment yield at the outlet of the watershed that can be used to compare results against observed values. Users can open this file in excel and analyze the results as desired. The pivot table in excel can be used to summarize the results by calendar year or water year.

Geospatial Exports

Open Standard Geopackage

A geopackage with a subcatchments and channels layer is available for download with vector layers of the subcatchemts and channels andtopographic, landuse, soils, and wepp results.

ESRI Proprietiary Geodatabase Format

The geospatial outputs may also be available in ESRI’s proprietary Geodatabase format. Creating Geodatabases through non-ESRI software is somewhat persnickety and this functionality may be dropped in the future.

ERMiT

The ERMiT output files allow users to directly input hillslope parameters directly into Erosion Risk Management Tool (ERMiT) Batch or Disturbed WEPP Batch spreadsheets for further analysis. ERMiT and ERMiT Batch are provided to allow users to compare various post-fire erosion control treatments in a probabilistic output. The ERMiT Batch allows users to examine all of the hillslopes in a watershed at once with various output options to determine which hillslopes have the probability of the highest (or lowest) erosion rates, how erosion mitigation treatments (i.e. mulching, seeding, etc.) will affect the hillslope erosion rates and how will recovery time effect erosion rates for up to 5 years post-fire. Disturbed WEPP and Disturbed WEPP Batch allows for analyzing various forest conditions and management options on hillslope erosion and provide probability of erosion in the first year after disturbance.

Download Hillslope Input CSV for ERMiT Batch and Disturbed WEPP Batch

This link will generate an input CSV file as a zip file. Download to your computer and extract the two files. The first is a Excel spreadsheet CSV file that is the input to ERMiT Batch or Disturbed WEPP. The second is metadata file providing the user with information on WEPPcloud run date, number of hillslopes, watershed outlet, climate station selected and description of all the columns in the CSV file.