Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 From Principles to Prediction: Research Needs for Forest Hydrology and Management Forest hydrology has built a strong foundation of general principles (Table 3-1) concerning the direct effects of forest management on hydrologic processes (Table 3-2) from plot studies, process studies, and watershed experiments (Chapter 3). The challenge now is to apply these principles to predict how hy- drologic processes will respond to many forms of change in forest landscapes. Forest hydrologists have long recognized the need to understand indirect and interacting effects of forest management at much larger spatial scales and longer temporal scales than is possible in plot studies, process studies, and wa- tershed experiments. Indirect effects are responses to forest management that are displaced in time or space, such as fire suppression leading to insect out- breaks that affect forest hydrology. Interacting effects occur when two or more management practices coincide, such as when post-salvage logging and road building have a different collective effect on forest hydrology that differs from their individual effects. This chapter examines the research challenges faced by forest hydrology as it moves from principles to prediction at larger spatial scales, at longer temporal scales, and in a changing social context. The chapter concludes by outlining the potential for improved cumulative watershed effects analysis that could provide the predictions needed by forest and water managers in the twenty-first century. SPATIAL RESEARCH NEEDS A key unresolved issue in forest hydrology is how to apply the findings of hydrological studies in one area to a different area or how to scale up the find- ings to large watersheds and landscapes (defined in Chapter 1, Box 1-1). Al- though forest hydrologists have confidence in the general principles of hydro- logic responses to forest management and disturbance (Chapter 3), they cannot predict precisely how forest management will affect hydrologic processes in specific places other than those that have been intensively studied. Most forest hydrology studies are conducted in small watersheds that are instrumented to measure streamflow and other hydrologic properties. However, the sum total of the area studied by forest hydrology is only a tiny fraction of the watersheds in the United States, and hydrologists recognize the need to extend hydrologic knowledge from âgauged basinsâ (watersheds that have measured records) to ungauged basins. Predictions are most needed in ungauged places to better un- derstand hydrologic effects where conflicts sometime arise: in water supply sys- tems for agriculture and cities; rivers where endangered and threatened aquatic 75
76 HYDROLOGIC EFFECTS OF A CHANGING FOREST LANDSCAPE species occur; large water bodies such as the Chesapeake Bay; and many, many others. Forest hydrology is adopting a landscape perspective to examine spatial pat- terns of forests and associated hydrologic processes and to link principles from plot- and small watershed scales (up to several square kilometers, see Chapter 3) to predictions at larger spatial scales (hundreds of square kilometers). Within a watershed, forests can be located in the headwaters, along riparian corridors, in woodlots in agriculture lands, and in urban or suburban areas (Figure 4-1). Based on its intra-basin position, a forest fulfills various water-related functions with respect to water quantity and quality. For example, forests in headwaters influence water yield and the quality of water delivered to downstream areas. Riparian forests located along streams throughout a watershed provide key func- tions for protecting streams from inputs of sediment (Naiman and Decamps, 1997), nutrients, and herbicides (Peterjohn and Correll, 1984; Lowrance et al., 1997); provide wildlife habitat for terrestrial and aquatic organisms (Barton et al., 1985; Darveau et al., 1995); and support a diversity of other functions (Ris- ser, 1995). Riparian forests have been greatly altered by economic develop- ment, and they are the focus of many forest management guidelines designed to preserve water quantity and water quality. Research Need: Process studies are needed to determine how forests, par- ticularly riparian forests, affect water quantity and quality according to their position within a watershed. Hydrologists use models to predict water quantity and quality in watersheds where there are no measured records. Since 2004, a working group for Predic- tion in Ungauged Basins (PUB) of the International Association of Hydrological Sciences (http://www.hydrologic science.org/pub/about.html) has developed methodologies for assessing uncertainty in hydrologic predictions arising from uncertainties in landscape properties and climate inputs, choice of model struc- ture, and methods of information transfer from gauged to ungauged watersheds (Sivapalan, 2003). Most hydrological models are developed and tested for gauged basins and subsequently are validated and applied to ungauged areas. However, models that have been fitted to data in small, gauged watersheds often provide inaccurate or imprecise predictions when they are (1) extrapolated to other small forested headwater basins, (2) extrapolated to future time periods, or (3) applied to large watersheds. This problem of prediction in ungauged basins has preoccupied hydrology researchers for several decades, and is compounded by a lack of information about how direct hydrologic effects interact under the multiple sets of specific conditions that occur in changing forest landscapes. By examining forest hydrologic processes under a wide range of conditions, land- scape-scale studies could provide data and understanding to help extend basic forest hydrology principles to make predictions needed by water managers.
FROM PRINCIPLES TO PREDICTION 77 Headwater forests Forest harvest, fire, mortality from insects and disease, exurban sprawl Small forested watersheds Removal & re-establishment of upstream riparian buffer strips riparian forests downstream reservoir riparian forests Removal & restoration of wetlands downstream wetland agriculture Creation and removal of woodlots and agricultural buffer strips downstream urban/suburban Urban forestry plantings FIGURE 4-1 Changes in forests in various parts of watersheds collectively contribute to cumulative watershed effects. Bold italicized font illustrates a few of the changes occurring to forests in these parts of watersheds in the United States. Harvest, fire, insects, and pests in headwater forests alter quantity and quality of water delivered to downstream ar- eas. Modifications of riparian forests and wetlands affect quantity, quality, and timing of water delivered from lands adjacent to streams and rivers. Modifications of forestsâsmall woodlots, windbreaksâon agricultural lands and urban forestry can influence water quan- tity and quality. Research Need: Landscape-scale studies to improve predictions of hydro- logic responses in large watersheds and landscapes based on general princi- ples of hydrologic responses to forest management (Tables 3-1 and 3-2) de- veloped in small, homogeneous watersheds Road networks are a pervasive feature of forest landscapes (Figure 4-2). The location and density of roads in a watershed can influence the hydrologic effects of roads. In many forested areas, roads are concentrated in the valley bottoms immediately adjacent to streams, meadows, and wetlands. These roads have a particularly high potential for delivering runoff and sediment to streams, lakes, and aquatic ecosystems. The legacy of midslope roads from past logging practices is also of concern, because these have a high potential for subsurface flow interception, connectivity to the drainage network, and initiating shallow landslides. Considerable research effort has been devoted to modifying road design and management to mitigate erosion, and to developing techniques
78 HYDROLOGIC EFFECTS OF A CHANGING FOREST LANDSCAPE FIGURE 4-2 A dense network of forest roads in the Cascade Mountains of western Ore- gon. Photo courtesy of A. Levno, U.S. Forest Service, retired. to decommission roads (Madej, 2001; Megahan et al., 2001; Ice et al., 2004; Switalkski et al., 2005). The hydrologic effects of road networks at large scales, and the effects of road decommissioning are not widely studied. Research Need: Landscape-scale studies of the effects of road networks and road decommissioning on water quantity and quality in larger watersheds and landscapes, particularly during extreme storm events TEMPORAL RESEARCH NEEDS Water management systems have been designed and operated under the as- sumption that hydrologic variables such as annual water yield, while varying over time, can be predicted reliably based on instrument records. However, in- creased understanding of long-term variability and trends in climate has under- mined this assumption (Milly et al., 2008; Barnett et al., 2008). If precipitation and streamflow vary over the long term, future annual water yields may fall short of the levels that water supply systemsâand attendant agricultural and urban developmentâwere designed to provide. Given this context, it is critical for forest hydrologists to extend beyond general principles to make predictions
FROM PRINCIPLES TO PREDICTION 79 of how forest management and disturbance affect hydrologic response on time scales that exceed those of most forest hydrology science. Some forest and stream management plans now include the historical range of variability, which presupposes that (1) past conditions and processes provide context and guidance for managing ecological systems today; and (2) distur- bance-driven spatial and temporal variability is a vital attribute of nearly all eco- logical systems (Landres et al., 1999; Perera et al., 2004; Poff et al., 1997). The historical range of variability helps characterize the variation in quantity, qual- ity, and timing of streamflow from forests. It can also be used to establish base- lines for assessing change in water from forests over time. Forests and their associated hydrologic processes change on time scales ranging from decades to hundreds, or even thousands, of years (Table 3-2). The temporal context for understanding forest hydrologic processes involves expand- ing the temporal scale into the past to consider the effects of past forest practices and into the future to project and anticipate changes in land use and climate. Many different kinds of legacies of past human activities affect forests in the United States (USGCRP, 2000). In some areas, native forests have been con- verted to agricultural and urban uses, and forests have regrown on abandoned agricultural lands in others. Roads and expansion of urban areas have frag- mented forests into smaller, less-contiguous patches and created new drainage patterns. Fire suppression has changed the structure and community composi- tion of many forests, especially in those with otherwise active fire regimes. Ex- otic species introductions, grazing by domestic animals, predator eradication, and timber harvesting methods have changed forest cover and species composi- tion as well. Future urban and suburban development and climate change are expected to continue to alter forest cover and species composition. Human ac- tivities will modify forests in the future (USGCRP, 2000), and future legacies will reflect current, regional forest histories (NRC, 2002). Research Need: Long-term predictions of hydrologic responses to forest management, including harvest, roads, and fire suppression, over decades to centuries based on general principles of hydrologic responses to forest management (Table 3-2) Regional Forest Histories in the United States Each region of the United States has a different history of forest conversion and management by humans that has yielded forests of different types with dif- ferent capacities to produce clean, abundant water. Hydrologic effects of re- gional forest histories in most areas were not documented and may be difficult, but not impossible, to reconstruct. Although data collection and recording at many experimental watersheds ceased in the 1970s and 1980s, these forested properties remain in the public domain and hydrologic data collection could be reactivated at these sites. If resumed, streamflow and water quality monitoring
80 HYDROLOGIC EFFECTS OF A CHANGING FOREST LANDSCAPE at these sites could be very informative about the hydrologic effects of long-term changes in forests, land use and land cover. For example, research conducted at the Coweeta Hydrologic Laboratory (and the Long-Term Ecological Research [LTER] Network) shows that pine plantations, which occupy much of the forest area in the Southeast, use more water than native deciduous forests (Swank et al., 1988). These types of assessments strengthen the understanding of hydro- logic effects of forest management activities. Research Need: Studies that compare hydrologic responses to forest man- agement and long-term changes in forest species among the various regions of the United States. Legacy of Exotic Species Many forested landscapes in the United States are affected by nonnative species introductions, a legacy of past human effects on landscapes. Introduced species cause profound ecological effects (Mack and DâAntonio, 1998), but their direct and indirect effects on hydrology are not as well documented. Possible declines in water yield resulting from the invasion of riparian zones by exotic tree species such as salt cedar (Tamarix spp.) and Russian olive (Elaeagnus an- gustifolia) have been a major source of concern in the arid southwestern of the United States (Vitousek, 1990). Riparian vegetation in southwestern rivers, in- cluding nonnative salt cedar and Russian olive, as well as native cottonwood, may use up to one-third of the water lost along the river, or amounts equivalent to irrigation withdrawals and evaporation (Dahm et al., 2002). Simple strategies to increase water yield in arid regions by tree removal, including exotic invasive tree species, have not produced consistent, demonstrable increases in water yield (Shafroth et al., 2005). Hydrologic processes in forests also are affected indi- rectly by defoliation and tree mortality due to introduced insects (e.g., gypsy moth, hemlock wooly adelgid) and diseases (chestnut blight). Research Need: Studies of the direct effects of exotic tree species introduc- tion or removal and indirect effects of introduced insects and diseases, on water quantity and quality from forests. Legacy of Fire Suppression The legacy of fire suppression, practiced since European colonists arrived and especially since 1910 in the western United States, may have effects on for- est structure and water (Clark, 1990; Baker, 1992; Covington and Moore, 1994). One consequence of fire suppression has been an increase in stem density and leaf area in forests (Johnson et al., 2001; Graham et al., 2004). Another conse- quence of fire suppression has been an increased susceptibility to insect and pest
FROM PRINCIPLES TO PREDICTION 81 outbreaks and greater vulnerability to defoliation during outbreaks, particularly in forests where the suppression of fire has led to crowding of trees and in- creased stress (Bergeron and Dansereau, 1993; McCollough et al., 1998; Power et al., 1999). As time since the last fire increases and forest age exceeds the natural fire return interval, outbreaks of insects such as spruce budworm (Cho- ristoneura fumiferana), spruce beetle (Dendroctonus rufipennis), and mountain pine beetle (Dendroctonus ponderosae) are more likely, and mortality from out- breaks is higher. Connections between fire suppression and insect epidemics have been documented in Alberta, British Columbia, the Rocky Mountains, and the Pacific Northwest (Bergeron and Leduc, 1998; Bebi et al., 2003; Taylor and Carroll, 2004). In small paired watersheds, forest mortality as the result of insects and pests produces a short-term increase in water yield and some transient effects on water quality (Swank, 1988; Lewis and Likens, 2007), but in the longer term, foliage regeneration may lead to decreases in water yield after insect infestations (Swank et al., 1988). However, very few studies have addressed the hydrologic consequences of fire suppression and insect outbreaks (Love, 1955; Bethalmy, 1974, 1975; Alila et al., Year). The longer-term, indirect and interacting effects of fire suppression on forest ecology, water yield, and water quality over large watersheds are difficult to estimate because the changes in forest density and composition are largely undocumented. Data exist and could be used to assess the effects of fire suppression on water yield and quality in some experimental watersheds owned and managed by federal agencies (the U.S. Forest Service [USFS] and other agencies); however, these data have not been analyzed for these purposes. Research Need: Landscape-scale, long-term studies of the effects of fire suppression and insect and disease outbreaks on water quantity and quality from forests. Legacy and Future of Grazing, Predator Eradication, and Predator Reintroduction Grazing by domestic livestock and native animals and the consequences of eradication and reintrodution of predators have potential but largely unquanti- fied effects on hydrology from forested watersheds. Grazing of forests by do- mestic livestock in the nineteenth and twentieth centuries had long-term conse- quences for forest ecology, water timing, and water quality (Rummell, 1951, Johnson, 1952; Armour et al., 1994). Forest grazing by sheep and cattle left biophysical legacies in forest landscapes. It suppressed fire, helped convert the original park-like forests of the interior western United States into dense stands of less fire-tolerant tree species, and changed the physical environment by re- ducing fire frequencies, compacting soils, reducing water infiltration rates, and increasing erosion (Belsky and Blumenthal, 1997; Graham et al., 2004). Con-
82 HYDROLOGIC EFFECTS OF A CHANGING FOREST LANDSCAPE tinued mobility of sediment accumulated in stream channels from elevated up- land erosion or historic agriculture in forests in the past (e.g., Platts, 1981) is another landscape-scale legacy of forest grazing that may still be influencing water quality. In forests throughout the United States, eradication of native predators has led to increases in populations of deer and other browsers, reducing the cover of tree seedlings and saplings, particularly in riparian forests (Terborgh et al., 2001). Ecologists call the indirect effects of predator removal on forest vegeta- tion âtrophic cascades,â whereby predators exert âtop-downâ control on primary production and growth of vegetation (Polis et al., 2000). In Yellowstone Na- tional Park, ecological studies indicate that wolf eradication increased browsing and largely eliminated riparian aspen forests; the reintroduction of wolves and associated trophic cascades may lead to riparian vegetation expansion in some areas (Ripple and Beschta, 2007). Despite a rapidly expanding literature on ecological trophic cascades, very little work has examined the indirect hydro- logic effects on water yield and quality from trophic cascades and predator eradication and reintroductions. Research Need: Studies of the indirect and interacting effects on water yield and quality of reduced grazing by domestic cattle and sheep and of predator removal and reintroduction on ungulate browsing in riparian for- ests. Future Climate Change Effects on Forests Climate changes over the past half-century are likely to have major effects on water quantity and timing from forests; some of these effects are already ap- parent. Climate change effects on forests will occur through (1) direct effects on precipitation type, snowmelt timing, and amount of precipitation; (2) indirect effects on disturbances in forests, including fire, wind, insects, and pests; and (3) indirect effects on vegetation species ranges including both natives and exotics. A number of modeling studies have investigated direct hydrologic effects of climate change at the large watershed, regional, national, and global scales (Bar- nett et al., 2004). Simulated future climate in the Columbia River Basin indicate a shift in the timing of water availability through reduced snowpack, earlier snowmelt, higher evapotranspiration in early summer, and earlier spring peak flows, leading to reductions of 75 to 90 percent in April-September runoff vol- umes; studies of direct climate change effects on hydrology for Montana and California produced similar results (Running and Nemani, 1991; Lieth and Whitfield, 1998; Miller et al., 2003; Dettinger et al., 2004). These changes are expected to exacerbate conflicts over limited dry season river flows among en- ergy production, irrigation, instream flow, and recreational uses (Hamlet and Lettenmeier, 1999). Simulated hydrologic effects of climate change show an increase in competition for reservoir storage between hydropower and instream
FROM PRINCIPLES TO PREDICTION 83 flow targets developed in response to the Endangered Species Act listing of Co- lumbia River salmonids (Payne et al., 2004). In the Colorado River basin, simu- lated future climate scenarios show a reduction in water storage, reduced hydro- power production, and an increase in the number of years in which reservoir releases do not meet demand (Christensen et al., 2004). These predictions of direct effects of climate change on water yield from forested basins do not take into account many potential indirect and interacting effects of forest responses to climate change (Clark, 1990; Stocks et al., 1998; Dale et al., 2001, Walther et al., 2002). For example, Westerling et al. (2006) assert that climate warming is responsible for the increased frequency of wild- fires and longer fire seasons in the western United States. Widespread outbreaks of pine beetle in British Columbia and the Rocky Mountains also are attributed to climate warming (Taylor and Carroll, 2004; Hicke et al., 2006). By altering forest disturbance, climate change may indirectly affect water yield and water quality. In addition to effects on forest disturbances, climate change is expected to alter forest productivity and species composition (Aber et al., 1995, Houghton, 1995; USGCRP, 2000). Forest productivity will change (Melillo et al., 1993). Forest species composition is changing (Brown et al., 1997; Davis and Shaw, 2001; Pearson and Dawson, 2003; Parmesan and Yohe, 2003; Thomas et al., 2004). By altering the productivity and species composition of forests, climate change may indirectly modify water quantity and quality. Research Need: Studies of indirect and interacting effects of climate change on water quantity and quality through effects on forest disturbance, structure, and species composition. SOCIAL RESEARCH NEEDS The social context of forest and water interactions has changed since the mid-twentieth century and continues to shift today. Current issues involving forests and water encompass practices that extend beyond the traditional scope of timber production and now include multiple public and private groups, past and future land use trends, and non-market resource valuation and trading schemes. Changing forest landscapes also include rapid changes in the public policy setting. Today, with growing populations in or adjacent to forestlands and more stringent polices regulating forest management, there are many more groups of people influencing where and how forests should be grown or pre- served, and this influences forest water resources. Urban and Exurban Development Future expansion of homes, commerce, and industry replacing forests in ur-
84 HYDROLOGIC EFFECTS OF A CHANGING FOREST LANDSCAPE ban, suburban, and exurban areas is likely to produce hydrologic effects. In the last few decades of the twentieth century, âexurban sprawlâ (sprawl develop- ment in the farthest fringes of metropolitan areas) changed demographic patterns (Alig, 2006). Compared to urbanization, sprawl is more diffuse, is more wide- spread, and affects more area per unit of population. Through exurban sprawl, human populations spread into rural areas, extend the amount of impervious area, and fragment remaining blocks of forest. Across the United States, the proportion of area classified as rural area has declined in most counties, housing has fragmented many large forest patches, and housing density has increased over large exurban areas (Theobald, 2005; Goetz et al 2004; Auch et al., 2004). Continued exurban sprawl is expected to reduce forest cover throughout the United States, with potential consequences for water yield and quality for mu- nicipalities. In the western United States, census data from 1960 to 2000 show development spreading, especially along the coast (Travis et al., 2005). By 2040 a large swath of development is projected along the west coast, and the foot- prints of larger interior cities such as Phoenix, Denver, and Salt Lake City are projected to increase. Rural valleys across the mountain Westâfor example, in western Coloradoâexhibit marked exurbanization by 2040 (Travis et al., 2005; Theobald, 2005). The trend is evident in the eastern United States, too. Popula- tion increases and exurban sprawl are occurring in the south and southeastern United States (Theobald, 2005), New England (Foster et al., 2004), and the mid- Atlantic United States (Goetz et al., 2004). Direct consequences of these pat- terns are an increase in impervious area (faster runoff), and with more wildland- urban interfaces, houses come into greater contact with wildland processes, in- cluding windthrow of trees, landslides, and fire (Radeloff et al., 2005). Much is known about the localized and larger-scale effects of urbanization on hydrology, but it is not clear whether the hydrologic effects of exurban sprawl can be pre- dicted by these past studies because of differences between sprawl and urbaniza- tion. The new patterns of sprawl and development in forested landscapes create an opportunity for new interdisciplinary studies involving hydrologists, ecolo- gists, economists, and social scientists to improve and communicate understand- ing of the value of forests in their role of producing water. In Arizona, econo- mists worked with hydrologists to translate biophysical responses to treatments into production functions that capture economic impacts of forest change on hydrology (Baker and Ffolliott, 1998). Interdisciplinary research by hydrolo- gists, ecologists, managers, and economists can help to translate research and monitoring results into economic terms that have meaning to decision makers and policy analysts. Research Needs: â¢ Landscape-scale analyses that assess the effects of exurban sprawl on water quantity and quality.
FROM PRINCIPLES TO PREDICTION 85 â¢ Collaborative research among hydrologists, ecologists, and econo- mists and social scientists to improve and communicate understanding of the value of sustaining water resources from forests. Changing Forest Practices on Public Lands Environmental laws have led to reduced timber harvest on public forest- lands, and wildfires appear to be increasing in severity and extent. Yet, simulta- neously, more people have moved into the urban-wildland interface in or near forests. These factors have turned attention to protecting people and their prop- erty from forest disturbances, such as fire, landslides, and wind storms. Some contemporary management practices, therefore, cater to these new social condi- tions and involve new forms of timber harvest whose effects on hydrology are not well understood. Federal forests are managed for a wide range of objectives (see Chapter 2). Today, many of the federal forest lands are managed to conserve terrestrial and aquatic species and protect water, practices that constrain the amount and extent of timber harvest (USDA and USDI, 1994; Christensen et al., 1996; Thomas, 1996). On public lands, legal requirements for species protection, forest preser- vation, and fire in effect limit forest management options to a very narrow scope. One noted example of contemporary management practices is the Northwest Forest Plan (1996), which restricts cutting of mature and old-growth forests and mandates wider riparian buffers to protect the habitat of an endan- gered species, the spotted owl. After fire, salvage logging may occur on public forest lands (Donato et al., 2006; Thompson and Spies, 2007). Ongoing road decommissioning has occurred in key watersheds of the Aquatic Conservation Strategy of the Northwest Forest Plan (Box 2-1). Other contemporary forest management practices derived from the Healthy Forests Restoration Act (2003), including thinning for fuel reduction and forest restoration. In western forests where conflicts arise over the allocation of limited summer low flows between endangered species, agriculture, and other uses, managers need to understand the effects of contemporary practices, such as various levels of forest thinning, buffers, and salvage logging, on streamflow and aquatic ecosystems. Research Need: Studies of effects on water quantity and quality of contem- porary forest management on public lands, including thinning for fuel re- duction or forest restoration, salvage logging, road decommissioning, and redesigned riparian buffers. CUMULATIVE WATERSHED EFFECTS Cumulative watershed effects (CWEs) include any changes that involve wa- tershed processes and are influenced by multiple land use activities. Assess-
86 HYDROLOGIC EFFECTS OF A CHANGING FOREST LANDSCAPE ments of CWEs use interdisciplinary approaches at the large watershed scale and attempt to include longer temporal scales (Reid, 1993; MacDonald, 2000). Assessing CWEs requires an understanding of physical, chemical, and biologi- cal processes that route water, sediment, nutrients, and other pollutants from uplands to downstream areas (Sidle, 2000). Research on CWEs attempts to establish cause-effect relationships among forests and water over large spatial scales; however, CWEs are elusive to quan- tify and are especially difficult to convey in terms applicable to policy and man- agement. Many of the methods for evaluating cumulative effects have encoun- tered technical, legal, or political problems because they have not explicitly ad- dressed the complexity of biophysical interactions spread over large areas and long time frames (Grant and Swanson, 1991). Spatially explicit studies of hydrologic responses to forest management dis- play how land uses and land cover across large watersheds interact to influence the quantity, timing, and quality of streamflow. In many parts of the United States, novel analyses represent watershed processes in large watersheds and landscapes. Three case studies are presented: Oregon (Box 4-1), the Chesa- peake Bay (Box 4-2), and New England (Box 4-3). Each of these cases reflects: (1) satellite image interpretation of land cover in heterogeneous watersheds; (2) public participation in envisioning future scenarios in the watershed; and (3) spatially explicit alternative future land cover patterns and engineered features using geographic information systems and spatial models (Baker and Landers, 2003; King et al., 2005). In the first example (Box 4-1 and Figure 4-4), the Environmental Protection Agency (EPA) funded work to reconstruct historical forest cover and project future forest cover under alternative scenarios for future development in the for- ested Willamette River drainage basin, a 30,000 km2 area in Oregon that in- cludes the largest cities in the state. In the Chesapeake Bay drainage basin, re- gional coordination and large watershed-scale modeling are used to help the five states in the basin to meet nutrient reduction goals; forests are a critical part of each state's strategy, particularly riparian forest buffers and conserving existing forests, and state goals (Box 4-2 and Figure 4-5). Finally, the New England case (Box 4-3 and Figure 4-6) shows forest management on public land for the pur- pose of protecting municipal drinking water in a 75 km2 watershed in Connecti- cut and Massachusetts by limiting harvest in forests close to the reservoir on steep slopes or unstable soils. While all of these examples display the power and effectiveness of geospa- tial analysis and ways to combine, analyze, and communicate complex data, none of them explicitly focuses on CWEs or water quantity and quality. How- ever, all three of these studies represent the spatial patterns of forests in large watersheds, implicitly draw on principles of forest hydrology, and aggregate scientific principles relevant to policy making at the large-watershed scale. These studies illustrate the potential for forest hydrologists to use geospatial and geostatistical tools to analyze and display hydrologic processes in large, hetero- geneous watersheds, and they are excellent examples of how forest hydrology
FROM PRINCIPLES TO PREDICTION 87 and spatially explicit CWE research could be done. With these types of toolsâ including the rapid advancements in and increased availability of spatial data, greater power of geographic information systems, and gains in scientific knowl- edgeâforest hydrology could develop a reasonable path forward to assessing CWEs and fulfilling other needs of forest management in the twenty-first cen- tury. Research Need: Spatially explicit assessments of CWEs in large watersheds that connect and communicate watershed processes, changing land cover, management issues, and public participation. SUMMARY Expanding from general principles (Table 3-1) of hydrologic responses to forest management and disturbance (Table 3-2) to predicting responses in changing forest landscapes will involve meeting research needs that extend the spatial and temporal scales and social considerations in forest hydrology. These extensions to the existing rich body of forest hydrology science can improve the application of forest hydrology to address forest management issues in the twenty-first century. Most forest hydrology research has been conducted at small spatial scales, over short time scales, and in watersheds with homogeneous land cover. Today, forest and water managers need predictions of direct, indi- rect, and interacting hydrologic responses to changing forest landscapes and guidance in applying these predictions at the scales of large watersheds, land- scapes, and regions, over multiple decades. Spatial, temporal, and socioeco- nomic factors and climate are important sources of change in forested landscapes, and each of these has research needs and challenges associated with improving forest hydrology applications to manage forests for forest and water resources. This chapter describes the challenges and the research needed to ad- dress them. Cumulative watershed effects are discussed and examples are given of how the science of forest hydrology can use existing technologies to quantify and communicate hydrologic effects over large spatial scales and in basins with heterogeneous land cover.
88 HYDROLOGIC EFFECTS OF A CHANGING FOREST LANDSCAPE BOX 4-1 Integrating Stakeholder Perspectives to Manage Future Forest Landscapes in the Willamette Valley, Oregon The Willamette Basin Alternative Futures Analysis is a novel environmental assess- ment approach (http://oregonstate.edu/dept/pnw-erc/index.htm) that facilitates consensus building and helps communities make decisions about land and water use (EPA, 2002). It combines technological capacities for reconstructing past landscapes and modeling future landscapes with public consultation to provide a long-term, large-area perspective on the combined effects of multiple policies and regulations affecting the quality of the environ- ment and natural resources within a geographic area. In this process, community members articulate and understand their different viewpoints, priorities, and goals. 2 The Willamette River drains an area of nearly 30,000 km between the Cascade and Coast Range Mountains in western Oregon (see Figure 4-4). Forests occupy two-thirds of the basin, mostly in upland areas, while much of the lowland valley area has been con- verted to agricultural use (43 percent) and urban and rural development (11percent). Ore- gonâs three largest cities, Portland, Salem, and Eugene-Springfield, are located in the val- ley, along the Willamette River. About 2 million people lived in the basin in 1990. By 2050, the basin population is expected to nearly double, placing demands on land and water resources and creating challenges for land and water use planning. In the mid-1990s, recognizing the need for an integrated strategy for development and conservation, Oregon Governor John Kitzhaber initiated basinwide planning efforts and created the Willamette Valley Livability Forum (http://www.wvlf.org). Working with stakeholders, researchers outlined three alternative futures and pro- jected them through the year 2050. Plan Trend 2050 represents the expected future land- scape based on current policies and recent trends; Development 2050 reflects a loosening of current policies to allow freer rein to market forces; and Conservation 2050 places greater emphasis on ecosystem protection and restoration. All three futures assume the same population increase. The historical, present-day, and future landscapes are repre- sented as maps (see Figure 4-4) with associated assumptions about management prac- tices and water use and computer-simulated âflyoversâ of the future conditions. Research- ers used models to compare expected effects of the alternative futures on terrestrial wild- life, water availability, ecological conditions of streams, and the condition of the Willamette River (Baker and Landers, 2003; Dole and Niemi, 2003; Hulse et al., 2003; Van Sickle et al., 2003). Major changes to the Willamette River Basin since EuroAmerican settlement in 1850 include: (1) loss of 80 percent of riparian forests; (2) conversion of about two-thirds of the old-growth forest in the uplands to younger forest; (3) drying of an estimated 130 km of second- to fourth-order streams in a moderately dry summer due to consumptive water use for irrigation, municipal, industrial, and other out-of-stream water uses; and (4) 15 to 90 percent declines in wildlife habitat and abundance, and stream and river biota. Projected conversion of agricultural lands to rural and urban development between 2000 and 2050 produced smaller effects on ecosystems than conversions of riparian or upland forest to either agriculture or urban land uses. All three futures involved substantial increases in water use and declines in water availability, resulting in habitat loss and sum- mer drying of streams. Changes were greatest in the Plan Trend and least in the Conser- vation future, but even the water conservation measures incorporated into Conservation 2050 were not sufficient to reverse recent trends of increasing water withdrawals for human use. Researchers concluded that major changes in Oregonâs water rights laws would likely be needed to substantially reduce water withdrawals, but stakeholders did not consider such changes to be plausible.
FROM PRINCIPLES TO PREDICTION 89 FIGURE 4-4 Trajectories of landscape change in the Willamette River Basin, from pre- EuroAmerican settlement, to ca. 1990, to three alternative futures for 2050. Source: EPA (2002).SOURCE: Available online at http://oregonstate.edu/dept/pnw-erc/index.htm.
90 HYDROLOGIC EFFECTS OF A CHANGING FOREST LANDSCAPE BOX 4-2 How to Keep Forests on the Landscape: The Chesapeake Bay Case Study Once a primarily forested landscape, the Chesapeake drainage has undergone sev- eral centuries of settlement and conversion to agricultural and urban uses. Forests still cover 58 percent of the basin, although the area now supports more than 16.6 million peo- ple (Figure 4-5). Despite their extent, forests are estimated to contribute only 14 percent of the nitrogen and 2 percent of the phosphorus in the basin. Hence, the issues for forest hydrology become how to keep forests on the landscape and where they are most critical for maintaining basic environmental functions. Forest retention, especially in hydrologically active areas such as riparian zones, is an important practice for mitigating the effects of other land uses. The Chesapeake Bay has been the focus of multistate coordination and widespread efforts to reduce nutrients since the 1980s, while population grew by 23 percent. With lar- ger houses and more roads to serve lower-density development, impervious surfaces ex- panded at even higher rates, creating effects on water quantity and quality even where some stormwater measures were in place to mitigate the increases in peak flows. Greater total flow from impervious areas enlarged stream channels, mobilizing sediment stored from previous decades of agricultural erosion or abandoned mill dams. With less water infiltrated and slowly released from subsurface soils, summer baseflows in developed areas declined. New nutrient loads were added through more intensive management for turf and landscaping and septic or sewer systems. Although some measures of water quality have improved measurably in portions of the watershed since 1983, the waters remain stub- bornly below standards set for clarity, dissolved oxygen, and other criteria. The stakes are going up as the deadline to meet and maintain water quality standards for the Chesapeake Bay mainstem nears, with Total Maximum Daily Load limits slated for 2010 in a watershed that ranges over six states. The low rates of nutrient export from forests will be needed to meet water quality standards, as well as support wildlife and aquatic habitats dependent on the forests. The effects of development and agriculture usually have strong signals in water quality and habitat measures (e.g., King et al., 2005). Hydrologic changes associated with urbani- zation are even larger. In contrast, the changes from forest management can be difficult to distinguish from annual variation, at least where practices such as stream buffers and ap- propriate road design are in place (McCoy et al., 2000). Where the effects of forest man- agement on water quality and quantity are measurable, they are generally at least an order of magnitude less than those from other land uses. Forests are used in a variety of ways to mitigate the adverse effects of land use change in the Chesapeake Bay basin. The Chesapeake Bay Program adopted an ambi- tious riparian forest restoration goal of 10,000 miles in 2003. A 1996 goal of 2,010 miles by 2010 was met years ahead of time as a result of the new Conservation Reserve Enhance- ment Program in Maryland, Virginia, and Pennsylvania. Efforts are under way to increase forest conservation significantly. Forests retained on development sites are allotted credits towards meeting stormwater requirements in Maryland. âCritical areaâ laws in Virginia and Maryland limit density and forest clearing, especially in buffer areas, and require replace- ment of cleared forests. Urban canopy cover goals are being set to benefit from air and water quality improvements of trees in developed areas (Cappiella et al., 2005). As forest area continues to decline, forests increasingly are being retained and restored to mitigate the effects of other land uses on water quantity and quality. Forests are seen by some to be the vacant part of the landscape, waiting for a higher and better use. However, forests are essential to maintaining basic environmental func- tions, and explicit methods to maintain forest on the landscape will be needed to meet ba- sic water quality goals.
FROM PRINCIPLES TO PREDICTION 91 FIGURE 4-5 Classification of land cover west of the Chesapeake Bay. White and red areas are urban and impervious zones, yellow is agriculture, and green is forest. SOURCE: Available online at http://www.geog.umd.edu/ resac/lc2.html. Reprinted, with permission, from Stephen D. Prince (May 19, 2008). Copyright by the Department of Geog- raphy, University of Maryland
92 HYDROLOGIC EFFECTS OF A CHANGING FOREST LANDSCAPE BOX 4-3 Forest Management for Watershed Protection: New England 2 The Barkhamsted Reservoir watershed drains an area of about 75 km of Connecticut and Massachusetts, including private lands and public forestlands owned and managed by the Metropolitan District Commission (MDC) and Connecticut and Massachusetts state forests. Forests in the watershed are managed for timber production and other uses, but these uses have the potential to adversely affect water quality and quantity in the reservoir. To protect the municipal water resources, a decision support system (the Watershed Forest Management Information System), has been developed to 1. map the forest and land cover and engineered features such as roads in the wa- tershed using geographic information systems; 2. designate â forests and wetlands for conservation based on their perceived role in supplying clean water (the Conservation Priority Index [CPI]); â agricultural lands and parks for restoration; and â residential, commercial, and industrial lands for nonpoint source pollution; 3. identify â roads as sediment sources, according to their proximity to water bodies; and â culverts for failure given estimated peak discharges; and 4. spatially allocate and schedule forest harvest and silvicultural operations by posi- tion within the watershed (Barten and Ernst, 2004). Using this system of indices, each parcel is given a score that represents its conserva- tion value within the watershed (Figure 4-6). Because the information is shared among all landowners in the basin, scores can be used to build cooperation among state or federal agencies and nongovernmental organizations to focus conservation efforts. Watershed sensitivity classes constructed from the CPI (Figure 4-7) can be used to delineate and co- ordinate forest harvests. Forests around water bodies and wetlands are in the highest- sensitivity class to facilitate stream habitat, forest diversity and forest regeneration. This approach illustrates the potential for interagency cooperation, coordination, and information sharing to guide watershed management plans in large, multi-ownership watersheds. SOURCES: Barten and Ernst (2004); Zhang (2006).
FROM PRINCIPLES TO PREDICTION 93 FIGURE 4-6 Conservation priorities assigned to forests on public and private lands within the Barkhamsted Reservoir. SOURCE: Gregory, P.E., Y. Zhang, and P.K. Barten. Water- shed Forest Management Information System (WFMIS) User's Guide Version 1.0. USDA. Available online at http://www.wetpartnership.org/WFMIS%20User%27s%20Guide.pdf.
94 HYDROLOGIC EFFECTS OF A CHANGING FOREST LANDSCAPE FIGURE 4-7 Public lands classified by watershed sensitivity (class I is highest sensitivity, most restricted use; Class IV is lowest sensitivity, least restricted use) in Barkhamsted Reservoir Watershed. SOURCE: Gregory, P.E., Y. Zhang, and P.K. Barten. Watershed Forest Management Information System (WFMIS) User's Guide Version 1.0. USDA. Avail- able online at http://www.wetpartnership.org/WFMIS%20User% 27s%20Guide.pdf.