07 July 2011

Forests and Water, part 2: Razing Arizona...

Author's Note: the following is derived from a project proposal to the National Science Foundation (NSF) that I co-authored in 2009 with a colleague at Northern Arizona University (NAU).  More context on the narrative given here is provided in part 0 of this series.  A brief review of forest - water connections, forest health, and fire susceptibility is given in part 1 of this series.

ASTER false-color image of the northeastern portion of the Wallow Fire
area in Arizona and New Mexico, obtained 21 June 2011 with the NASA
Terra Earth observation satellite. A more complete description of this
image is available from the NASA Earth Observatory.
As I post this, the 2011 Wallow Fire in the the Apache National Forest in eastern Arizona and western New Mexico (pictured in part 1) has become the largest wildfire in Arizona's history (~540,000 acres) and is finally under a significant level of containment.  The Wallow Fire was started, it seems, by people; it was fueled, however, by an ongoing southwestern drought and nearly a century of forest and wildlands mismanagement that have left massive quantities of vulnerable tinder in place.  Wildfire policy in the U.S. grew from federal response to the three-million-acre Great Fire of 1910 in the northwestern states, the subject of The Big Burn by Timothy Egan (another book on my reading list).  A strict policy of fire suppression, established by Congress in the aftermath of that event, has been managed by a hobbled and underfunded U.S. Forest Service with little regard for changing priorities, climatic conditions and improved ecological science over the past century.

Would this proposed work that I have been posting on my blog have saved the Apache-Sitgreaves Forest in Arizona?  Not likely; we would be right in the middle of the project this summer, had it been funded.  Might we have helped to enhance cooperation and communication among stakeholders and managers, and contributed to changes in forest management policy for the better so that something like the Wallow Fire was less likely to happen once we had completed out project?  We like to think so, but only time would tell if we had been successful.  As it is, however, the legacy policies are not working, and new directions in research and policy-making in order to preserve our forested lands are still needed.

We seek to aid, by the collection of relevant datasets and tools and their application in appropriate physical modeling and decision support systems, ongoing efforts at adaptability and resilience in the forest management community. We will address both short-term disturbances, such as forest fires and meteorological drought, and long-term regime changes due to factors such as climatological drought, invasive species and urban development. Academic scientists will work with forest and other resource managers and stakeholders to find innovative methods for the integration and analysis of data from traditionally disparate sources, facilitating the development of new predictive and decision support tools. The project proposed here will provide standardized data access at the intersection of forest health, climate conditions and water supplies beginning in the American Southwest, but the methods and tools developed will be available for application in any location of need. A customized, map-based platform will support interoperability of information and datasets, physical and parametric models, and decision support tools related to water supply, forest, fuels and fire management information and to hydrologic, meteorological and climatological datasets.

We propose to integrate pre-existing tools and to develop new methods within a common platform that will facilitate monitoring of climate, hydrologic and ecological variables and their integration with land cover, fire fuels and forest health information at landscape and watershed scales for strategic planning and management applications. Much of these datasets have historically been gathered for specific but separate uses; this work will integrate those with a network of researchers, land managers, modelers and forecasters, and educators in a common framework for information analysis and dissemination. This effort is aimed at the community of data providers, application developers, and a broad range of end users including land and resource management agencies, environmental organizations, and local and state governments.

We envision a network of scientists and subject experts, datasets, modeling tools, visualization tools, scenario comparison templates and management-level decision-makers in a horizontally- and vertically-integrated, fully interactive framework for the collaborative support of decisions related to forest health and assessment of the impacts those decisions may have on water resources in areas of interest. Overall, the scientific process that leads to informed decision-making is bracketed by people with an interest in the resources at issue. These are stakeholders with individual ideas, values, and levels of understanding. Combining the community with science addressed here presents an area of complex feedback: the health of a forested area is determined in part by the physiography, meteorology and climatology of a region, three factors which determine almost entirely the water resource availability in that same area. In forested areas, especially forests subject to fire disturbances, the cycles of water, carbon and nutrients are inextricably intertwined and every effort must be made to consider these natural processes together.

According to the National Research Council (NRC), considerable expansion in the number and variety of technologies have led to little improvement in the effective use of quantitative modeling and decision support tools (DSTs) by decision-makers [1,2,3]. At the same time, there have been calls to find more meaningful, participatory ways to bring the widely-varied judgments of stakeholders into environmental decision-making [4,5]. We recognize that the technologically-aided consensus-building process must remain inclusive, informative, educational, flexible, and responsive. Where scientific assessment and public policy intersect, such practices as subjective judgment, probability assessment and logical coherence are often more relevant than rigorous, deductive hypothesis testing as in traditional applications of the scientific method.

Decision-makers, both managers and policy-makers, can act only on the information that is made available, and the timeliness and accuracy of that information becomes important as long as the information itself maintains the context of the problems at hand. Analytical tools that help visualize a research result or a model output are often insufficient for good decision-making when they do not address the context, value choices and uncertainty of the results [3]. Our academic team with extensive community experience is well-suited to advise resource managers on the integration of science in decision support while not also dictating the outcomes of the managers' decision-making processes.

Recognizing the need for a more holistic approach to forest health management and community protection, in 2003 Arizona Governor Janet Napolitano initiated a series of Forest Health conferences and established both the Forest Health Advisory Council and the Forest Health Oversight Council, now known collectively as the Arizona Forest Health Council (AFHC). Primary concerns were the conditions of Arizona’s forests, the risks of fire from various causes and in sensitive locations, and the impacts of local, state and federal efforts to mitigate fire risks over time.

Fire Regime Condition Classes in forested areas across Arizona,
indicative of forest health conditions, from the 2007 Statewide 
Strategy for Restoring Arizona's Forests [6]. This map was
generated with data from LANDFIRE, a cooperative program of
the USFS and DOI. Under traditional regimes of land management
and fire suppression, areas in yellow (red) have diverged
noticeably (significantly) from a natural fire cycle, leading to
potentially catastrophic conditions.
The forerunning organizations to the AFHC promoted an agenda that included the development of an Arizona Fuels, Information, Restoration, and Education Mapping and Assessment Program (FIREMAP). Council efforts resulted in 2007 in a Statewide Strategy for Restoring Arizona's Forests [6], which assessed forest health with respect to wildfire susceptibility across much of the state (figure at right) and presented a 20-year road-map toward the restoration of forest health, the establishment of sustainable forest-based commerce, and the protection of rural communities from wildfire. Upon acceptance of the report, the Governor mandated that the AFHC should develop a "scorecard" system to track and highlight critical management activities for the health of Arizona’s forests. Such an approach requires integrated information from numerous data sources and output from multiple physical models of the forest ecological system, applied in locations throughout the state, to approach successful utility at the state and regional levels. The computational and community networks proposed here may provide a significant advancement on the implementation of such a management framework.

Several other organizations have been involved in forest-oriented assessments in Arizona over the past several years and could be engaged as network members in an effort to include all knowledgeable participants in stakeholder-oriented decision processes. The U.S. Geological Survey (USGS) completed their Southwest Region Gap Analysis Program (SWReGAP), covering a five-state area in the southwestern U.S., in 2005. The Nature Conservancy collaborated with the USDA Forest Service (USFS) to complete a Southwest Forest Assessment Project in 2007 oriented on forest health in Arizona and New Mexico. Both of these projects will be discussed as components of the proposed network. Finally, the Food, Conservation, and Energy Act of 2008 (the "2008 U.S. Farm Bill," enacted by Congress over Presidential veto) requires every state to conduct a cross-jurisdictional assessment of forest resource conditions, trends and threats (Pub.L. 110-234, Title VIII, Subtitle A, Sect. 8002). According to staff at the USFS Regional Office, the assessment for Arizona was to be completed in late 2010. Our proposed work aimed to gather recommendations from the assessment advisory team and for this project to support follow-on needs identified by the USFS statewide assessment effort.

Efforts to integrate the latest science with the objectives of stakeholders are already underway across the Southwest and are showing impacts on strategic planning for water and forest management. Efforts include the NAU Forest Ecosystem Restoration Analysis (ForestERA) project and the University of Arizona WildfireAlternatives (WALTER) program, both of which have developed datasets and tools to support decision-making processes. These programs have been used to identify preferred solutions among stakeholders with diverse objectives [7,8]. Workshop participants have indicated that solutions are more easily identified when map-based information is available, as it allows them to visualize better the magnitude and spatial context of perceived conflicts [9].

The proposed work brings together knowledge in computer science, informatics, physical process modeling, data visualization, remote sensing, forest health, water resources management, climatology and meteorology, and ecology and hydrology. On the technical level it is necessary to adopt standards that are consistent with the larger community of practice in each of these fields. In models and analytical tools it is desirable to understand both the physical processes and relevant community values, so that each component is oriented properly. The highest level of service must then provide interoperability between data sets, analysis tools, physical models, and the decision-making community. Decision interoperability is a level of operation at which community leadership, analytical capability, communication, consensus-building, and mastery of the computational networks at hand become most useful. It is the decision that is the goal, and all efforts must be oriented toward facilitation of the decision-making process while maintaining spatial and geographic context for the stakeholders and their value choices.

Interoperability between spatial analysis centers, such as the NAU Geospatial Research and Information Laboratory (GRAIL), and efforts to bring together datasets and analytical tools through common-platform systems, will offer significant benefits for timeliness, display and use of disparate datasets to support the community of forest managers, stakeholders and researchers. As the importance of water- and climate-related conditions become more evident to forest managers, GRAIL researchers must often compile datasets for individual projects that include streams and water bodies, riparian areas, water storage tanks, springs, seeps, groundwater wells, impaired waters, drinking water reservoirs, watershed delineation, and climate conditions such as precipitation and drought indices. Obtaining and compiling these data as geospatial products is a time-consuming process. Programs like LANDFIRE depend on local spatial data products; the recent LANDFIRE update process required fire perimeter information from numerous sources. Forest treatment information with metadata, as from FIREMAP, could be included in future update processes. Providing local, interoperable datasets from cross-jurisdictional sources to national programs will improve product quality for local use. The work of local, state and national facilities will benefit greatly from a system that facilitates transfers of spatial data with adequate metadata. In cases where high-demand datasets are not of sufficient quality or timeliness, improved dataset interoperability through a common platform will certainly assist resource management efforts.

In addition to building the computational network for data and model interoperability, we look forward to engagement with subject experts, educators, students, and other stakeholders in a process of community-oriented requirements development for the protection of forest health and water resources in forested areas. We have planned several workshops over the course of the proposed project where direct interaction with the community will help us generate and approach new requirements for datasets and models that are not already identified here. These efforts at building both computational and interpersonal network are aimed at greater reliance on peer-to-peer communication, information exchange among peers with common objectives, and the ability for participants in consensus-building activities to develop widely varied and interactive practices to meet the needs of growing variability in decision-making activities.


[1] NRC (National Research Council), 1996: Linking Science and Technology to Society’s Environmental Goals. National Forum on Science and Technology Goals. National Academies Press, 544 pp., ISBN 978-0-309-05578-9.

[2] NRC (National Research Council), 1997: Building a Foundation for Sound Environmental Decisions. Committee on Research Opportunities and Priorities for EPA. National Academies Press, 104 pp., ISBN 978-0-309-05795-0.

[3] NRC (National Research Council), 2005: Decision Making for the Environment: Social and Behavioral Science Research Priorities (G.D. Brewer and P.C. Stern, eds.). Committee on the Human Dimensions of Global Change. National Academies Press, 296 pp., ISBN 978-0-309-09540-2.

[4] EPA (U.S. Environmental Protection Agency), 2001: Improved Science-Based Environmental
Stakeholder Processes
. Commentary by the EPA Science Advisory Board (EPA-SAB-EC-COM-01-006), Washington, D.C.

[5] Sisk, T.D., J.W. Prather, H.M. Hampton, A.N. Aumack, Y. Xu, and B.G. Dickson, 2006: Participatory landscape analysis to guide restoration of ponderosa pine ecosystems in the American Southwest. Landscape and Urban Planning, v. 78, pp. 300-310, doi:10.1016/j.landurbplan.2005.10.003.

[6] Arizona Governor’s Forest Health Advisory and Oversight Councils, 2007: Statewide Strategy for Restoring Arizona’s Forests. 140 pp. Available on-line at http://www.azforests.info/.

[7] Hampton, H.M., E.N. Aumack, J.W. Prather, Y. Xu, B.G. Dickson, and T.D. Sisk, 2005: "A spatial decision support system for forest restoration." In The Colorado Plateau II: Biophysical, Socioeconomic, and Cultural Research (C. Van Riper III and D. Mattson, eds.), University of Arizona Press, ISBN 978-0-8165-2526-3.

[8] Hampton, H.M., E.N. Aumack, J.W. Prather, B.G. Dickson, Y. Xu, and T.D. Sisk, 2006: "Development and transfer of spatial tools based on landscape ecology principles: Supporting public participation in forest restoration planning in the southwestern U.S." In Forest Landscape Ecology: Transferring Knowledge to Practice (A. Perera, L. Buse, and T. Crow, eds.), Springer, ISBN 978-0-387-34242-9.

[9] Noss, R.F., P. Beier, W.W. Covington, R.E. Grumbine, D.B. Lindenmayer, J.W. Prather, F. Schmiegelow, T.D. Sisk, and D.J. Vosick, 2006: Recommendations for integrating restoration ecology and conservation biology in ponderosa pine forests of the southwestern United States. Restoration Ecology, v. 14, pp. 4-10, doi:10.1111/j.1526-100X.2006.00099.x.

No comments: