Text Size:   A+ A- A   •   Text Only
Site Image
Criterion 4 Indicator 18
Area and Percent of Forest Land with Significant Soil Erosion (Rill, Sheet, Gully, Mass Wasting, and Roadside)
Soil and water resources are the foundation of human and natural ecosystems. Erosion is commonly viewed as a major threat to soil, water, and related forest and plant resources, particularly agricultural crops. Yet, in a broad ecological context, erosion is a natural process in the building up and wearing down of the land. In this process, soil material from upland areas is removed, transported, and redeposited downslope, perhaps only a few feet or, sometimes, hundreds of miles away. Removal and transportation agents are usually water from rainfall or snowmelt, wind, and gravity. Topsoil removal from upland areas means loss of organic matter and nutrients for vegetation; loss of subsoil means reduced anchorage for roots and less material for soil formation. Soil loss creates conditions that generally reduce productivity of trees and other plants, especially when human-induced disturbances accelerate erosional processes to levels beyond those of natural systems.

Can This Indicator Be Quantified
The effects of soil erosion can be difficult to quantify. Most research has focused on estimating the amounts of sediment and debris exported from watersheds, and their impacts on downstream water quality and riparian attributes. Most of these studies have been local in nature and not well-distributed across many vegetation, soil, and geologic groups. Few studies in Oregon’s forests directly relate erosional processes to the loss of forest productivity or changes in vegetative composition. Extensive studies are costly and time-consuming to replicate in ways that compare the effects of natural with human-induced erosion. After major forest fires, most efforts are to salvage timber and revegetate burned areas, not to study long-term effects of soil loss after the fire.
Except for landslides, erosion is not a major event in moist Pacific Northwest forests because overstory and understory vegetation and ground litter protect the soil against raindrop impact. Moist climates also favor fast vegetation regrowth after most disturbances. But the construction of roads and skid trails, and some timber harvest techniques, can expose the soil and create conditions in which water can be channeled to produce gullies. Natural and human-caused forest fires also remove protective vegetative cover and expose bare soil to the erosive forces of rainfall and snowmelt.
In Oregon, sheet, rill, and gully erosion are most prominent east of the Cascades where forest vegetation is sparse, annual rainfall is low, and fire frequency is high. Landslides, or mass wasting, are most common in steep coastal and western Cascade forests that have unstable soils and receive high annual precipitation.
In lowlands, accumulated soil from upland areas provides anchorage and nutrients for new vegetation, thus fostering greater productivity. The establishment of some pioneer tree species, such as alder, benefits from soil disturbance or removal. However, other tree species, such as Douglas-fir and hemlock, are slower to get established if soil organic matter and woody debris are removed from the landscape by erosion and landslides. Downslope and downstream riparian and aquatic systems can benefit from the accumulation of sediment and debris, but municipal water quality is significantly lowered when these products wash into municipal water sources.
In summary, the beneficial or harmful effects of erosion are quantifiable only when one specifies the plant species of interest; the soil, topographical, and geological landscape where species are located; the timing and intensity of rain events; and the products or values of interest to resource mangers and planners.

It is difficult to estimate the impacts of erosion on forest productivity. Historical erosion rates in forests are not known, and past monitoring focused on agricultural rather than forest lands. Soils with high amounts of organic matter usually have less erosion than soils with low organic matter. This relationship exists because high organic matter content is also associated with good soil structure, many pores in the soil, and high infiltration rates. These three factors allow water to enter and move through the soil rather than travel over the soil surface.
If predictions come true of climate change and wetter winters in the Pacific Northwest, natural erosion rates in forests will probably increase across the state. In western Oregon, a noted increase in rain storm intensity has already raised the risk of landslides. In eastern Oregon, continued forest fuel accumulation and fire exclusion have increased the risk of soil loss after natural and human-caused forest fires. Private and public land owners now use better forest practices to harvest timber and build roads; these newer practices minimize disturbances and reduce erosion.

Data Source and Availability
Watershed studies in Oregon have included treatments and results for small and large basins. The central themes were the effect of harvesting or road building on water and sediment yields, and characterization of landslide processes. One study provided a broad literature review and synthesis of information about surface erosion and mass movement events on forest landscapes in the Pacific Northwest. Another study reviewed the effects of timber harvesting and roads on water quantity and quality and erosional processes. In another study, researchers reported higher stocking density and better Douglas-fir growth (measured by height, after 5 years) on 25 non-landslide areas, compared to conditions on paired 6- to 28-year-old landslide areas.
In general, researchers have been able to study vegetation response to surface erosion and landslides, after clearcutting and slash burning, for a year and up to a decade or more. However, for several reasons, it is not known what the long-term effects are of erosional processes that continue over several timber rotations. It is difficult to quantify the linkages between erosion effects and the physical stability and biological integrity of diverse forested sites. Also, it is very difficult to study, anticipate, and manage the impacts of chronic landslide processes, as studied in single- or paired-watershed studies; effects from one severe storm can overshadow the sediment and vegetation effects measured during the previous 30 to 50 years.
Several helpful databases are accessible through the Internet (see Table 18-1). The National Resources Inventory (NRI) provides information for 5- and 10-year trends on forest land cover and rangeland conditions on non-federal lands. Data elements that can be used to model erosion include cover factor, slope percent, and slope length; K (erodability) and T (soil loss tolerance) factors for forests were included in 1982.
The Water Erosion Prediction Project (WEPP) is developing erosion models for forested conditions, with or without roads. These models will replace the Universal Soil Loss Equation, for which K and T factors did not work well in forests.
The National Soil Characterization Database (NSSL) lists the erodability of soils and includes estimated physical properties for all soil series, phases of soil series, and components of soil units mapped on public and private lands in Oregon.
The State Soil Geographic (STATSGO) database has over 200 map units for Oregon on a 1:250,000 base map; over 25 physical and chemical soil properties and interpretation attributes exist for each soil component in each map unit. The minimum map unit or polygon is 1,544 acres, a feature that limits database use to broad planning activities. A Landslide Bibliography includes over 30 references summarizing the kinds and effects of landslides on aquatic and terrestrial ecosystems in the Pacific Northwest.
After catastrophic statewide floods and landslides in 1996, Oregon Department of Forestry (ODF) personnel mapped distance, stream channel effects, and sediment delivery from debris flows in six Coast Range and two western Cascades sites. These results will be posted on the ODF web site in summer 1999.
Table 18-1. Internet databases with information on soil erosion factors and processes affecting forests in Oregon.
Database NameContentsInternet Access
Landslide Bibliography30+ references on landslide processes and inventorieshttp://sequoia.fsl.orst.edu.lter/pubs/biblio/biblslid.htm
NRI NationalWind and water erosion datahttp://www.nhq.nrcs.usda.gov/NRI/data.html &
Resources InventoryFr 1982, 1987, and 1992http://www.nhq.nrcs.usda.gov/land/index/intro.html
NSSL National Soil Characterization DatabaseChemical & physical properties for soil series, phases, and map unit componentshttp://www.statlab.iastate.edu/soils/nsdaf/main.html
Oregon Dept. Forestry1996 landslide traitshttp://www.odf.state.or.us/fp/fld%5Fproj.htm
STATSGO State Soil Geographic DatabasePhysical & chemical properties; management interpretationshttp://www.ftw.nrcs.usda.gov/statsgo.html
WEPP Water Erosion Prediction ProjectPredict sediment loss from forest roads, trails, and burnshttp://forest.moscowfsl.wsu.edu/fswepp/

Reliability of Data
All soil mapping, correlation, and laboratory analyses follow strict national quality control and quality assurance protocols developed originally by the Soil Conservation Service, and now administered by the Natural Resources Conservation Service. However, until the mid-1980s, forest lands were mapped less intensively than agricultural lands. Thus, forested areas in older soil surveys have fewer map units for which erosion risk attributes such as slope, soil physical properties, and erodability factors can be assigned.
Small and large watershed erosion studies have had site-specific objectives, methods, and treatments. Data quality is not known for these studies.

Sediment loss and erosion effects on forest productivity can be studied at scales from small individual watersheds to basins covering many states. Sediment removal, transport, and depositional processes are the same at all scales, but the impacts and effects will vary from watershed to watershed depending on management objectives and desired outcomes. It is not easy to take results from small watersheds and apply them to large watersheds. Soil, geologic, topographic, storm, and climate attributes are so variable that scaling up from small to large watersheds involves complex modeling and assumptions.

Recommended Action for Data Collection
Existing resource data layers should be used to build a preliminary erosion and landslide risk map at a STATSGO 1:250,000 scale, across the five major land areas in Oregon. This project should use existing soil, geologic, topographic, slope, annual rainfall, and other resource data layers. Where more intensive soil surveys exist on national forests, BLM lands, and other public lands, the same data layers should be used to make more detailed prediction units for specific areas.
With these preliminary units, decisions can then be made about where it will be most effective to either model surface erosion or to identify areas susceptible to chronic and episodic landslides. In the second step, existing stand exam, inventory, growth trials, and tree genetic field trials should be used to collect information about the productivity of major forest types found across the mapped erosion and landslide index units. Third, basins should be identified where sediment yield and landslide studies exist and should be continued. Fourth, field studies of tree growth and yield assessment should be superimposed on the same basins; this will help establish relationships between soil loss and landslide occurrence with forest growth and productivity.

For more detailed definitions of these terms, see Swanson, et. al., under "Selected References" below. Also see Beschta, et. al., for a good introduction to the subject.
Dry ravel — Downslope particle movement on steep slopes in dry climates caused by gravity and animals, not precipitation or snowmelt.
Gully erosion — Incision of a channel more than 1 square-foot in cross-sectional area by a concentrated flow of water.
Mass wasting, landslides — Rapid movement of the soil mantle down steep slopes that carries along tens to thousands of cubic yards of trees, shrubs, grass, soil, subsoil, rocks, and other debris.
Rill erosion — Forming tiny gullies, irregularly dispersed, on bare or fallow land.
Sheet erosion — Removing soil almost uniformly from an entire slope.

Selected References
Adams, P. W., and J. O. Ringer. 1994. The effects of timber harvesting & forest roads on water quantity & quality in the Pacific Northwest: Summary & annotated bibliography. Oregon State University, Forest Engineering Department, Corvallis, OR. 147 pp.
Beschta, R. L., and J. R. Boyle, C. C. Chambers, W. P. Gibson, and others. 1995. Cumulative effects of forest practices in Oregon: Literature and synthesis. Oregon State University, Department of Forest Resources, Corvallis, OR.
Grant, G. E., and A. L. Wolff. 1991. Long-term patterns of sediment transport after timber harvest, Western Cascade Mountains, Oregon, USA. In: Sediment and stream water quality in a changing environment: trends and explanation, proceedings of Vienna Symposium. IAHS publication number 203; pp. 31-40.
Miles, D. W. R., and F. J. Swanson, C. T. Youngberg. 1984. Effects of landslide erosion on subsequent Douglas-fir growth and stocking levels in the western Cascades, Oregon. Soil Science Society of America Journal 48:667-671.
Swanson, F. J., and J. L. Clayton, W. F. Megahan, G. Bush. 1989. Erosional processes and long-term site productivity. In: Perry, D. A., and R. Meurisse, B. Thomas, and others, editors; Maintaining the long-term productivity of Pacific Northwest forest ecosystems, pp. 67-81. Timber Press, Portland, OR.

People Interviewed
Karen Bennett, USDA Forest Service, Siuslaw National Forest, Corvallis, OR.
Bernard Bormann, USDA Forest Service Pacific Northwest Research Station, Corvallis, OR.
Kermit Cromack, Forest Science Dept., Oregon State University, Corvallis, OR.
Gordon Grant, USDA Forest Service, Pacific Northwest Research Station, Corvallis, OR.
Mark Johnson, EPA Research Laboratory, Corvallis, OR.
Duane Lammers, USDA Forest Service, Pacific Northwest Region, Corvallis, OR.
Chad McGrath, Natural Resources Conservation Service, Portland, OR.
Dick Miller, USDA Forest Service (retired), Pacific Northwest Research Station, Olympia, WA.
Phil Sollins, Forest Science Dept., Oregon State University, Corvallis, OR.
Fred Swanson1, USDA Forest Service, Pacific Northwest Research Station, Corvallis, OR.
  1. Completed technical review of draft before its submission to ODF.