Winter 2000

Studies of Post-Fire Erosion in the Colorado Front Range Benefit the Upper South Platte Watershed Protection and Restoration Project

Deborah Martin

US Geological Survey

Introduction

The Buffalo Creek Fire in Colorado in May of 1996 was followed by substantial flooding and erosion two months later. This combination of events emphasized the need to consider post-fire erosional risk in the watershed planning process. The burned area is within the Pike National Forest, located on the densely populated eastern slope of the Colorado Front Range (see map). Ponderosa pine (Pinus ponderosa) is dominant on south-facing slopes, whereas Douglas-fir (Tseudotsuga menziesii) dominates on north-facing slopes. The fire burned nearly 5,000 hectares, and the majority of the burned area (63%) was subjected to a high intensity, stand-replacing fire (U. S. Department of Agriculture Forest Service, 1996). The flooding after the fire killed two people and, in the three months following the fire, transported approximately 331,000 m³ of coarse sediment into Strontia Springs Reservoir (Moody and Martin, 1998). This reservoir supplies over 75% of the drinking water to the city of Denver. This sedimentation rate is nearly 30 times the annual rate of sediment input used in designing the reservoir (Borland, 1978). The reservoir also experienced a significant degradation in water quality as a result of the input of burned material and sediment. The Denver Water Department, the agency responsible for distributing drinking water from the reservoir, estimates that it spent over $1 million in immediate clean-up efforts after the fire (Oulton, 1996).

U. S. Geological Survey Study in the Buffalo Creek Burn Area

Researchers with the National Research Program of the U. S. Geological Survey (USGS) received funding from the Denver Water Board and the U. S. Department of Agriculture Forest Service (USFS) to quantify the post-fire erosional response of the Buffalo Creek burn area to intense wildfire. This study is part of a larger effort funded by the USGS National Research Program to examine the effects of fire on water and sediment yields from watersheds in the United States and abroad. The Buffalo Creek Fire burned two 'sixth-level' watersheds: Buffalo Creek, with 22% of the watershed burned, and Spring Creek, with 79% of the watershed burned. The USFS further delineates the fourth-level Hydrologic Unit Codes of the USGS (Seaber et al., 1987) into fifth- and sixth-level watersheds (U. S. Department of Agriculture Forest Service, 1995). Spring Creek was selected for more intensive study by the USGS because its mouth is closer to Strontia Springs Reservoir, because a larger proportion of this watershed burned, and because less rehabilitation was carried out in the watershed by the Forest Service. The erosion of the Spring Creek watershed during the flood events was dramatic, and included the evacuation of unchannelized drainages, the incision of existing alluvial fans, and the widening of the main channel. Likewise, deposition of eroded sediment within the watershed was significant, producing new deposits over three meters thick in some areas. USGS researchers are using both ground-based and remote-sensing methods (see table below) to evaluate the processes and magnitudes of erosion and deposition. A significant component of this effort is the creation of elevation models to assess the magnitude of erosion and deposition within the Spring Creek watershed.

Creation of Elevation Models to Determine Erosion and Deposition

Using 1:12,000 aerial photography taken after the Buffalo Creek fire and again after the flooding, two different methods are being tested to create post-fire and post-flood elevation models that are being subtracted to determine the magnitude of erosion and deposition within Spring Creek. In the first approach, The USGS worked with Dave Wolf and Terry Johnson of the USFS Geometronics facility in Lakewood, Colorado. Terry Johnson used a 3-dimensional stereo plotter to hand pick breaks in slope that identified the channel boundary and channel surface in a test reach of Spring Creek for post-fire and post-flood aerial photography stereo pairs. He produced data files that were used by the software program HASP to produce two triangular irregular networks (TINs). From the TINs, the software produced contour maps of elevations. The post-fire and post-flood contour maps were subtracted to produce a contour map of erosion and deposition. Based on previous work using 1:12,000 aerial photography, the vertical resolution of this method is considered to be 1.5 meters. In the second approach, the USGS is collaborating with a University of Denver graduate student, Jeff Blossom, who is scanning stereo pairs of post-fire aerial photographs at a resolution of 14 microns (each pixel is 14 microns x 14 microns which equals 0.168 meters on the ground). The data file produced from the scanning step is used by the Photis ST Softcopy software to produce a digital elevation model (DEM). Two DEMs will be created, one for the post-fire set of aerial photographs and the second for the post-flood photographs. Again, the elevation models will be subtracted, creating a map that shows contours of areas of deposition and erosion for the entire Spring Creek watershed. The vertical resolution in the second approach will be determined by comparing cross-sections from the post-fire DEM to cross-sections from ground based surveys (which have a vertical resolution of 1 centimeter). Both approaches are testing the practicability of assessing watershed-scale erosion and deposition using aerial photography, an assessment that would be nearly impossible to carry out using detailed ground-based surveys.

The Upper South Platte Watershed Protection and Restoration Project

The Front Range of Colorado, extending from Colorado Springs to Fort Collins, includes an area of significant population density within the wildland-urban interface. The USFS identifies about 396,000 hectares of the Front Range in the 'Red Zone', a wildland-urban interface zone characterized by extreme risk to property in the event of a wildfire (Johnson, 1997). The Buffalo Creek burn area lies within the Upper South Platte River basin adjacent to residential areas within the 'Red Zone'. In addition to the Buffalo Creek Fire, several factors (such as the workshop sponsored by EPA and the Colorado State Forest Service, 'Using Data and Technology for Forest Health Decision-Making' (Sampson, in press), and Colorado's Unified Watershed Assessment) converged to provide the impetus for a multi-agency project to assess the status of the Upper South Platte River Basin and to help prioritize treatments for ecosystem restoration and fuel reduction efforts. The project is called the Upper South Platte Protection and Restoration Project supported by funding from the USFS, the Colorado State Forest Service, the Denver Water Department and the Environmental Protection Agency. Other agencies, including the USGS, are participating in an advisory role.

Excluding areas of the Basin which are adversely impacted by mining operations, the project analysis area encompasses more than 243,000 hectares. The major portion is located on public lands (the Pike and San Isabel National Forests), although privately owned lands, including land owned by the Denver Water Department, comprise a considerable portion of the project area. The problems confronting land management agencies in the Upper South Platte River Basin are similar to those problems in other regions of the inland west: fire suppression which has led to excessive vegetation density, abundant fuel, and species declines, and to a major shift from the historic range of variability of forest structure, composition and landscape pattern.

These problems have defined three major objectives for the Upper South Platte Watershed Protection and Restoration Project. These objectives are (Hessel, et al., 1999):

To conduct a landscape scale assessment in the Upper South Platte watershed to select sixth-level watersheds for treatment.

To use assessment findings to derive a menu of landscape-wide management opportunities which would maintain or restore appropriate watershed function.

To demonstrate and assess the effectiveness on landscape-scale forest protection and restoration practices intended to reduce the likelihood of catastrophic wildfires and their associated risk to human life, property, water quality, soil hydrophobicity, and air quality.

The key issues to be assessed by the Upper South Platte Watershed Protection and Restoration Project are soil erosion, sedimentation, and the vegetation condition as it relates to future risks of catastrophic disturbances (Patten, et al., 1999). Many people involved in the project feel that this is the first time that watershed protection and restoration treatments are being prioritized on the basis of post-fire erosional risk. A major method to reduce the post-fire erosional risk is to restore the forest to its historical range of variability. Detailed studies of the protected forest around Cheesman Lake, a reservoir on the South Platte River in Colorado, are the basis for the determination of the historical range of forest variables such as structure, species composition and pattern (Brown, et al., 1999; Kaufmann, et al., in preparation, a and b). Underlying the whole approach is a vision of restoring the forest to a condition of ecological sustainability. The historical ponderosa pine/Douglas-fir landscape was spatially heterogeneous (areas of high and low forest density and openings) related to past mixed severity fires and variable rates of tree recruitment into openings created by fire. The restructuring of the forest landscape to make it spatially more heterogeneous is also likely to reduce the probability of extensive crown fires such as Buffalo Creek.

An important component of the project is a monitoring protocol that will evaluate the results of the restoration treatments, which include prescribed burning, forest thinning and a combination of both treatments. A monitoring protocol is under development and may be applicable to other areas of the western U. S. and to other parts of the world such as Turkey and Jordan where semi-arid conditions loosely approximate climatic conditions in the project area. Activity-scale monitoring (monitoring of individual treatments on US Forest land within the sixth-level watersheds) is expected to be covered under the National Environmental Protection Act (NEPA) process.

An end member of the continuum of vegetational treatments is a stand-replacing forest fire. At the other end of this continuum is no vegetation treatment at all, essentially a continuation of the fire suppression activities that have resulted in the current status of the forest. The USGS research in the Buffalo Creek burn area will contribute detailed information about how watersheds within the Project area may respond to severe fire. The data will provide a context for evaluating the erosional response from treatment areas and control areas (no treatment) within the project boundaries. Based on previous work in the Beaver Creek watershed of Arizona (Worley, 1965), it is predicted that, initially, treated areas within the Project area will produce more sediment than untreated areas. However, treated areas should avoid the catastrophic erosion of sediment that occurred after the Buffalo Creek fire, though during extreme rainfall events, all areas may erode. Episodic supply of large quantities of sediment to the river system may be important in the healthy functioning of the Upper South Platte River ecosystem. The ultimate test of the success of the Upper South Platte Watershed Protection and Restoration Project will be whether the treatments are indeed effective in reducing large-scale catastrophic fire while insuring that the ecosystems are resilient and ecologically sustainable.

The Upper South Platte Watershed Protection and Restoration Project with its cooperation among several public agencies, its unique approach to prioritizing potential treatment areas, and the landscape-scale vision of ecosystem restoration, promises to be a model for other large-scale protection and restoration projects. A major challenge is to develop a monitoring protocol that confirms that treatments are meeting water quality requirements. In addition, the project must assess the effectiveness of treatments in addressing the key issues of: reducing fuel hazards and associated catastrophic fire risk while maintaining soil productivity; and reducing sediment sources and improving water quality (Marsh, 1999).

For additional information about components of the Upper South Platte Watershed Protection and Restoration Project, USGS research and USFS forest structure research, please contact one of the persons listed below. -

References:

Borland, W. M., 1978. Study of sedimentation problems associated with the diversion of municipal raw water from the South Platte River in Platte Canyon approximately 25 miles southwest of the City of Denver, Colorado, Consultant Report, Exhibit No. 14.

Brown, P. M., M. R. Kaufmann and W. D. Shepperd, in press. Long-term landscape patterns of past fire events in a montane Ponderosa pine forest of Central Colorado, Landscape Ecology.

Hessel, D., et al., 1999. Briefing Paper for the Upper South Platte Watershed Protection and Restoration Project.

Kaufmann, M. R., L. S. Huckaby and P. Gleason, in preparation a. Ponderosa pine in the Colorado Front Range: Long historical fire and tree recruitment intervals and a case for landscape heterogeneity, in Crossing the Millennium: Integrating Spatial Technologies and Ecological Principles for a New Age in Fire Management, Joint Fire Science Conference and Workshop, June 15-17, 1999, Boise, Idaho.

Kaufmann, M. R., C. M. Regan and P. M. Brown, in preparation b. Heterogeneity in Ponderosa pine/Douglas-fir forest: Age and size structure in unlogged and logged landscapes of central Colorado.

Johnson, S. J., editor, 1997. Forest Insect and Disease Conditions in the Rocky Mountain Region 1996, Rocky Mountain Region, U. S. Department of Agriculture Forest Service, 45 pp.

Marsh, C., 1999. Upper South Platte Watershed Protection and Restoration Project Monitoring Strategy.

Moody, J. and D. Martin, 1998. Memo to Bob Weir, Denver Water Board.

Oulton, S., 1996. Forest homes in a fire belt; Front Range map shows disaster possible, The Sunday Denver Post, August 4, 1996, p. 1A, 20A.

Patten, F., et al., 1999. Executive Summary, Upper South Platte Watershed Protection and Restoration Project.

Sampson, N., editor, in press. Using Data and Technology for Forest Health Decision Making, Pingree Park, Colorado, September 29­October 4, 1996, Journal of Sustainable Forestry, 11 (1­2).

Seaber, P. R., F. P. Kapinos and G. L. Knapp, 1987. Hydrologic

unit maps, U. S. Geological Survey Water-Supply Paper 2294, 62 pp.

U. S. Department of Agriculture Forest Service, 1995. Integrated Resource Training Guide, Denver, CO: Rocky Mountain Region, U. S. Department of Agriculture Forest Service, 600 pp.

U. S. Department of Agriculture Forest Service, 1996. Buffalo Creek Fire Burned-Area Report, May 27, 1996, 7 p.

Worley, D. P., 1965. The Beaver Creek Pilot Watershed for evaluating multiple-use effects of watershed treatments, U. S. Department of Agriculture Forest Service, Research Paper RM-13, Fort Collins, Colorado: Rocky Mountain Forest and Range Experiment Station, U. S. Department of Agriculture Forest Service, 12 p. 2