WWETAC Projects
Project Title: ICWater sediment transport module
Status: Completed
Principal Investigator: Douglas F. Ryan, USDA-Forest Service, Pacific Northwest Research Station
Collaborators and Affiliations: William B. Samuels, SAIC Corp.; William J. Elliot, USDA-Forest Service, Rocky Mountain Research Station (RMRS); Deborah Martin, USGS, Boulder; Kevin Hyde, RMRS
E-mail Contact: Douglas Ryan, dryan01[at]fs.fed.us
Key Issues/Problems Addressed: Key decision makers lack tools to assess risk to water resources including drinking water and fish species of concern due to elevated suspended sediments during and after wildland fires.
Study Goal and Objectives: A GAO Report (2002) found that decision makers responding to and planning for wildland fires lack the ability to adequately consider a wider range of values-at-risk. Most existing decision support tools focus on buildings located within the projected fire boundary during the period while the fire is burning, while fire impacts on values such as public drinking water can occur outside of this space and time. This proposal seeks to fill this identified need for improved decision support to include water–related values in emergency response and planning for fires.
Public drinking water utilities are critical infrastructure because their disruption can endanger human health and seriously impact economic activity for large segments of the public. Safe drinking water is a mainstay of public health, and many industries depend on reliable, clean water from utilities to produce their products (e.g. food processing, manufacture of electronics and medicines, etc.). Wildland fire is a catastrophic geophysical phenomenon that can threaten public drinking water primarily by introducing high levels of contaminants into surface water that utilities depend upon for source water (Figure 1) or threaten critical aquatic habitat for sensitive fish (Figure 2). Contaminants in source water can cause water-borne disease, increased costs for treating drinking water or, in extreme cases, shut down a drinking water supply due to treatment failure. An acute concern for suppliers of drinking water is spills of toxic materials into surface waters caused by wildland fire damage to facilities that store or transport toxic materials (i.e. producing a “dirty fire”).
The potentially long-lasting effects on drinking water and fish populations (disruptive effects can last months to years after a large fire) make information about water impacts a high priority need for decision makers. Depending on the contaminant source, fire-related risks to drinking water can occur as the fire is burning from fire-caused toxic spills and fire retardant application, or months to years after the fire is out, caused by rain-storm driven wash off of ash, and accelerated soil erosion or landslides from burned lands.
Figure 1: Wildland fire threats to water supply values-at-risk.
Wildland fires can be caused by deliberate attacks such as arson and by natural events such as lightning. Wildland fires can also be ignited as collateral damage from terrorist attacks that use explosives or damage electrical transmission lines. Risk of fire is greatly enhanced by extreme weather including drought and high winds. Potential attackers can easily learn when high fire danger weather is expected because warnings are widely publicized. This tool would help deter attacks that use wildland fire as a weapon against public drinking water because terrorists seek “soft” targets to achieve maximum consequences. By enabling first responders to reduce the consequences of such an attack, the proposed tool would discourage terrorists from selecting wildland fire as a weapon and drinking water as a target. The tool would be equally useful for protecting the public from consequences of fires of natural origin.
Figure 2: Wildland fire threats to critical aquatic habitat as a value-at-risk.
It is the objective of the tool that we propose to provide decision makers with critical information needed to prioritize first responders’ actions in a timely way. This will be accomplished by delivering tool results to incident command teams via the Wildland Fire Decision Support System (WFDSS) which serves all federal agencies engaged in wildland fire operations. The proposed tool would also be useful to land managers and utility operators for planning wildland fuel-reduction treatments on critical watersheds in advance of fires by identifying where drinking water is most at risk and for assessing alternative fuel-reduction strategies to manage the risk. This tool would also be useful for simulating “what-if” scenarios of fire impacts on drinking water for use in training exercises and other emergency preparedness and planning activities.
Figure 3. Schematic of the proposed tool for predicting impacts of wildland fire on public drinking water and fish for emergency decision makers in government agencies and utilities. This project addresses transport of suspended sediment from burned landscapes to drinking water and fish downstream (pathway shaded in yellow). Specifically it develops the capability for ICWater to simulate the transport of suspended sediments, a major category of post-fire water contamination.
Transport and Dispersion of Waterborne Contaminants: We will use ICWater (Incident Command Tool for Protecting Drinking Water), a currently available tool, to predict the transport and dispersion of waterborne contaminants using a hydraulic modeling engine (Samuels et al. 2006) to compute time-of-travel and dispersion. This tool uses the National Hydrography Dataset Plus (Horizon Systems Corporation 2006) to estimate river flows and calibrates them to conditions current at the time of a spill using the USGS Real-time Stream Gauging Network (NWIS). ICWater also contains GIS layers and databases to analyze waterborne threats to EPA’s national inventory of public drinking water intakes and also EPA’s national inventory of toxic storage sites, pipelines, dams, reservoirs, roads and other relevant infrastructure (Samuels 2006). The advantage of the ICWater approach is that it is operational in all 50 states, can run in both stand-alone and web-based mode, and provides a realistic answer in a matter of minutes with a minimum of external inputs. ICWater is interoperable (i.e. can exchange geo-spatial data sets and reports) with other existing incident command tools: Defense Threat Reduction Agency’s CATS (Consequences Assessment Tool Suite) and FEMA’s HAZUS Disaster Planning Tool.
ICWater is currently used by Federal, State and local emergency response agencies and water supply utilities in response, planning, training, and development of monitoring strategies for deliberate or accidental toxic contamination of surface water. Skill assessment of the RiverSpill modeling engine (embedded in ICWater) for the Potomac River, has shown agreement of travel time and concentration model estimates within 10% and 30% respectively, of dye tracer study observations.
Production and mobilization of fine sediments (Turbidity): We will use GeoWEPP, a GIS Wizard based on the Water Erosion Prediction Project (WEPP) Model to make predictions of hydrology, soil erosion and sediment delivery from burned watersheds (Flanagan and Nearing 1995). When vegetation and plant litter are consumed by severe fire, soil erosion can be greatly accelerated during subsequent, intense rainfall events. WEPP captures this behavior by predicting erosion from single, extreme rain storms as well as average annual values of erosion and runoff. This tool has been validated with field measurements of forests burned by wildfires (Elliot 2004). It has several advantages: 1) The climate data used can be observed rainfall, or the model can generate rainfall from climate statistics from NOAA weather stations or other local weather data, 2) it predicts daily runoff of water to streams, daily peak runoff, and daily sediment delivery together with the expected return frequency of these events as an indicator of the uncertainty, and 3) it predicts particle sizes distribution of delivered sediment which is important because the sediments that cause most problems for drinking water utilities have the smallest size particles (i.e. clay and silt) that produce persistent turbidity in source water.
A major technical challenge to be overcome is simulating the transport in streams of sediment generated by the burned landscape. Models of sediment transport are often complex because they must accommodate several different mechanisms of sediment transport that apply to different particle size classes (e.g. Dunham et al. 2002; Peterson and Dunham 2003; Burnett 2007). However, because fine sediment (clay and silt particles) pose the greatest risk to drinking water, we are focusing on adapting models of transport that are confined to these fine-particle fractions for inclusion in our tool. We anticipate that this approach will greatly simplify our task. The size fractions that we will not be modeling (sand-size and larger particles) propagate shorter distances and more slowly through stream systems than do clay and silt, and are less likely to cause impacts on fish or drinking water than are fine sediments.
WWETAC funds (matched by funds from the ALI Program) are supporting work to address this critical challenge: the development and testing of candidate sediment transport algorithms for incorporation in ICWater.
Status: Bill Elliot has performed a proof of concept linking the land-surface sediment production tool (GeoWEPP) to the ICWater stream transport tool. Bill Samuels has accomplished an analogous proof-of-concept effort demonstrating that ICWater can link sediment generated from post-fire erosion to fish habitat downstream (see Figure 4). These proof-of-concept efforts showed that these components could be made compatible with regard to scaling and data transfer. Neither of these proof-of-concept efforts included a realistic sediment transport mechanism, but rather used the simplifying assumption that fine fraction sediment would be transported similarly to a dissolved contaminant.
Figure 4. Example ICWater downstream trace from our proof of concept showing simulated time of travel and concentration of a contaminant from a point source (e.g. fire retardant drop that accidentally reached a stream) transported downstream to Bull Trout habitat (pink highlight). The breakthrough curve (inset) illustrates that the hypothetical contaminant is projected to be below the level of concern for fish (red line) when the contaminant reaches the fish–bearing portion of the stream between 8.5 and 10 hours after the retardant drop.
Work to further link the combined GeoWEPP and ICWater tools to WFDSS has been proposed to Department of Homeland Security for evaluating wildland fire risk to public drinking water systems (a tool called RAVAR-Water) under their BAA 08-01 (Infrastructure Protection Broad Area Announcement) and to Joint Fire Sciences (2008 RFP) for evaluating wildland fire risk to fish habitat (a tool called ICFish).
Expected Products: A prototype algorithm that can be incorporated into ICWater to describe suspended sediment transport in streams. The outcome will be computer code that incorporates the sediment transport into the ICWater Tool. The realism of the simulations of post-fine sediment transport will be tested using monitoring data of fine sediments in streams under post-fire conditions. The objective is to provide realistic estimates of suspended sediment that will be experienced by fish populations or drinking water intakes downstream of burned lands.
Background Citations:
Burnett, K.M., Reeves, G.H., Miller, D.J., Clarke, S., Vance-Borland, K. and K. Christiansen. 2007. Distribution of salmon-habitat potential relative to landscape characteristics and implications for conservations. Ecological Applications 17:66–80.
Dunham, J.B., Rieman, B.E. and J.T. Peterson. 2002. Patch-based models of species occurrence: lessons from salmonid fishes in streams. In J.M. Scott, et al. (eds). Predicting species occurrences: issues of scale and accuracy. Island Press, pp 327-334.
Elliot, W.J. 2004. WEPP internet interfaces for forest erosion prediction. J. Amer. Water Res. Assoc. 40(2), 299-309.
Flanagan, D.C. and M.A. Nearing. 1995. USDA-Water Erosion Prediction Project: hillslope profile and watershed model documentation. NSERL Report No. 10. USDA-ARS National Soil Erosion Research Laboratory, West Lafayette, IN.
GAO. 2002. Severe Wildland Fires, GAO Report 02-259. Horizon Systems Corporation. 2006. National Hydrography Dataset Plus, Horizon Systems Corporation, <http://www. horizon-systems.com/nhdplus>, July 19, 2007.
Peterson, J.T. and J.B. Dunham. 2003. Combining inferences from models of capture efficiency, detectability, and suitable habitat to classify landscapes for conservation of threatened bull trout. Conservation Biology 17(4):1070-1077.
Samuels, W.B. 2006. Riverspill: A national application for drinking water protection. J. Hydraulic Eng., 132 (4): 393-403.
Samuels, W.B., Bahadur, R., Monteith, M.C., Amstutz, D.E., Pickus, J.M., Parker, K. and D. Ryan. 2006. NHD, riverspill, and the development of the incident command tool for drinking water protection. Water Res. Impact, 8(2) 15-18.
Project ID: FY08JB50
