Miller D.,Earth Systems Institute
USDA Forest Service - General Technical Report PNW-GTR | Year: 2013
Large woody debris is recognized as an important component of stream geomorphology and stream ecosystem function, and forest-land management is recognized as an important control on the quantity (and size and species distributions) of wood available for recruitment to streams. Much of the wood present in streams comes from adjacent forests, and riparian management practices now reflect our understanding of the role these forests play in modulating and maintaining stream environments. In steep terrain, slope failures also carry wood (and sediment) to streams from upslope source areas. In these environments, periodic inputs of wood and sediment from landslides and debris flows also play an important role in stream geomorphology and ecosystem dynamics. Channel environments are naturally dynamic systems. Depending on where you are in the channel network, discharge can vary from none in the summer to bed-scouring, channel-avulsing floods in the winter. Slope failures also drive variability in this system. Deposition of wood and sediment occur at discrete points in time and space, thereby creating temporal and spatial variability in channel environments. Redistribution and decay of deposited materials over time further add to this variability, and act to hide the original source of these materials, thereby masking the role of landsliding in setting stream environments. Landslide effects thus depend on when and where you look, and can be dif cult to discern if the landslide occurred some time ago. This makes efforts to anticipate the effects of landsliding very challenging. Are we interested in the short term? The long term? Are we interested in a single reach? Or effects over a basin? Observations also suggest that landslide effects depend on a host of factors, including valley geometry, channel geometry, the quantity and size of sediment and wood in the deposit, the amount of wood and sediment already in the channel, and the amount of wood and sediment that enter the channel over the lifespan of the landslide deposit. This sets the stage for considering slope failure as an upslope source of stream wood, particularly if we are to consider in-stream wood in the context of a stream ecosystem. I will brief y review the evidence on which to base conceptual and empirical models for identifying and characterizing upslope landslide source areas, and for placing them into a channel-network context. Then I'll illustrate the data-analysis and modeling approaches that we and our collaborators have been experimenting with to identify upslope source areas for stream wood and for anticipating the in-stream consequences of management decisions in those areas. These methods span a range of complexity. At the most basic, we use digital elevation data coupled with empirical models to identify the source areas and runout tracks for landslides that could potentially carry material to specified portions of the channel network (e.g., fish-bearing streams). To gain insights to effects of management, we couple stand-growth, wood recruitment, and landscape dynamics models to estimate wood abundances over time and space.
Bidlack A.L.,Ecotrust |
Bidlack A.L.,University of Alaska Southeast |
Benda L.E.,Earth Systems Institute |
Miewald T.,U.S. Fish and Wildlife Service |
And 2 more authors.
Transactions of the American Fisheries Society | Year: 2014
Ecosystem management requires information on habitat suitability across broad scales; however, comprehensive environmental surveys in remote areas are often impractical and expensive to carry out. Intrinsic Potential (IP) models provide a means to identify on a broad scale those portions of the landscape that can provide essential habitat for various freshwater fish species. These models are derived from watershed patterns and processes that are persistent and not readily affected by human activities. We developed an IP model for rearing habitat of Chinook Salmon throughout the Copper River watershed (63,000 km2) in southcentral Alaska, utilizing digital elevation models, expert opinion, and field surveys. Our model uses three variables-mean annual flow, gradient, and glacial influence-and adequately predicts where probable habitat for juvenile Chinook Salmon occurs across this large landscape. This model can help resource managers map critical habitat for salmon throughout the Copper River watershed, direct field research to appropriate stream reaches, and assist managers in prioritizing restoration actions, such as culvert replacement. Intrinsic Potential modeling is broadly applicable to other salmonid species and geographies and may inform future work on the ecological impacts of climate change in polar and subpolar river systems. © American Fisheries Society 2014.
Litschert S.E.,Earth Systems Institute |
Brown T.C.,Rocky Research |
Theobald D.M.,Colorado State University
Forest Ecology and Management | Year: 2012
Wildfires play a formative role in the processes that have created the ecosystems of the Southern Rockies Ecoregion (SRE). The extent of wildfires is influenced mainly by precipitation and temperature, which control biomass growth and fuel moisture. Forecasts of climate change in the SRE show an increase in temperatures, bringing warmer springs with earlier runoff and longer fire seasons. Increasing wildfire extent and intensity would affect human safety, livelihoods, and landscapes. Our summary of historical wildfire records from the national forests of the SRE from 1930 to 2006 revealed an order of magnitude increase in the annual number of fires recorded over the full time period and in the number of large fires since 1970. We developed a model of percent burned area in the SRE for the period 1970-2006 using temperature and precipitation variables (R 2=0.51, p=1.7E-05). We applied this model to predict percent burned area using data from two downscaled global circulation models (GCMs), for the Intergovernmental Panel on Climate Change Special Report Emissions Scenarios A2 (projects high increases in temperature) and B1 (projects lower temperature increases), for the time period 2010-2070. The results showed increasing trends in median burned areas for all scenarios and GCM combinations with higher increases for the B1 scenario. The results suggest that precipitation increases could at least partially compensate for the effect of temperature increases on burned area but the strength of this ameliorating effect of precipitation will remain uncertain until the GCMs are further developed. © 2011 Elsevier B.V.
Fullerton A.H.,National Oceanic and Atmospheric Administration |
Burnett K.M.,Oregon State University |
Steel E.A.,U.S. Department of Agriculture |
Flitcroft R.L.,Oregon State University |
And 5 more authors.
Freshwater Biology | Year: 2010
1. In this review, we first summarize how hydrologic connectivity has been studied for riverine fish capable of moving long distances, and then identify research opportunities that have clear conservation significance. Migratory species, such as anadromous salmonids, are good model organisms for understanding ecological connectivity in rivers because the spatial scale over which movements occur among freshwater habitats is large enough to be easily observed with available techniques; they are often economically or culturally valuable with habitats that can be easily fragmented by human activities; and they integrate landscape conditions from multiple surrounding catchment(s) with in-river conditions. Studies have focussed on three themes: (i) relatively stable connections (connections controlled by processes that act over broad spatio-temporal scales >1000 km2 and >100 years); (ii) dynamic connections (connections controlled by processes acting over fine to moderate spatio-temporal scales ~1-1000 km2 and <1-100 years); and (iii) anthropogenic influences on hydrologic connectivity, including actions that disrupt or enhance natural connections experienced by fish.2. We outline eight challenges to understanding the role of connectivity in riverine fish ecology, organized under three foci: (i) addressing the constraints of river structure; (ii) embracing temporal complexity in hydrologic connectivity; and (iii) managing connectivity for riverine fishes. Challenges include the spatial structure of stream networks, the force and direction of flow, scale-dependence of connectivity, shifting boundaries, complexity of behaviour and life histories and quantifying anthropogenic influence on connectivity and aligning management goals. As we discuss each challenge, we summarize relevant approaches in the literature and provide additional suggestions for improving research and management of connectivity for riverine fishes.3. Specifically, we suggest that rapid advances are possible in the following arenas: (i) incorporating network structure and river discharge into analyses; (ii) increasing explicit consideration of temporal complexity and fish behaviour in the scope of analyses; and (iii) parsing degrees of human and natural influences on connectivity and defining acceptable alterations. Multiscale analyses are most likely to identify dominant patterns of connections and disconnections, and the appropriate scale at which to focus conservation activities. Published 2010. This article is a US Government work and is in the public domain in the USA.
Fullerton A.H.,National Oceanic and Atmospheric Administration |
Jensen D.,National Oceanic and Atmospheric Administration |
Steel E.A.,National Oceanic and Atmospheric Administration |
Miller D.,Earth Systems Institute |
McElhany P.,National Oceanic and Atmospheric Administration
Environmental Modeling and Assessment | Year: 2010
Complex relationships between landscape and aquatic habitat conditions and salmon (Oncorhynchus spp.) populations make science-based management decisions both difficult and essential. Due to a paucity of empirical data, models characterizing these relationships are often used to forecast future conditions. We evaluated uncertainties in a suite of models that predict possible future habitat conditions and fish responses in the Lewis River Basin, Washington, USA. We evaluated sensitivities of predictions to uncertainty in model parameters. Results were sensitive to 60% of model parameters but substantially so (|partial regression coefficients| >0.5) to <10%. We also estimated accuracy of several predictions using field surveys. Observations mostly fell within predicted ranges for riparian shade and fine-sediment deposition, but large woody debris estimates matched only half the time. We provide suggestions to modelers for improving model accountability, and describe how managers can incorporate prediction uncertainty into decision-making, thereby improving the odds of successful salmon habitat recovery. © Springer Science+Business Media B.V. 2008.