Jennifer Koch

Examining fire-prone forest landscapes as coupled human and natural systems

Fire-prone landscapes are not well studied as coupled human and natural systems (CHANS) and present many challenges for understanding and promoting adaptive behaviors and institutions. Here, we explore how heterogeneity, feedbacks, and external drivers in this type of natural hazard system can lead to complexity and can limit the development of more adaptive approaches to policy and management. Institutions and social networks can counter these limitations and promote adaptation. We also develop a conceptual model that includes a robust characterization of social subsystems for a fire-prone landscape in Oregon and describe how we are building an agent-based model to promote understanding of this social-ecological system. Our agent-based model, which incorporates existing ecological models of vegetation and fire and is based on empirical studies of landowner decision-making, will be used to explore alternative management and fire scenarios with land managers and various public entities. We expect that the development of CHANS frameworks and the application of a simulation model in a collaborative setting will facilitate the development of more effective policies and practices for fire-prone landscapes.

Using an agent-based model to examine forest management outcomes in a fire-prone landscape in Oregon, USA

Fire-prone landscapes present many challenges for both managers and policy makers in developing adaptive behaviors and institutions. We used a coupled human and natural systems framework and an agent-based landscape model to examine how alternative management scenarios affect fire and ecosystem services metrics in a fire-prone multiownership landscape in the eastern Cascades of Oregon. Our model incorporated existing models of vegetation succession and fire spread and information from original empirical studies of landowner decision making. Our findings indicate that alternative management strategies can have variable effects on landscape outcomes over 50 years for fire, socioeconomic, and ecosystem services metrics. For example, scenarios with federal restoration treatments had slightly less high-severity fire than a scenario without treatment; exposure of homes in the wildland-urban interface to fire was also slightly less with restoration treatments compared to no management. Treatments appeared to be more effective at reducing high-severity fire in years with more fire than in years with less fire. Under the current management scenario, timber production could be maintained for at least 50 years on federal lands. Under an accelerated restoration scenario, timber production fell because of a shortage of areas meeting current stand structure treatment targets. Trade-offs between restoration outcomes (e.g., open forests with large fire-resistant trees) and habitat for species that require dense older forests were evident. For example, the proportional area of nesting habitat for northern spotted owl (Strix occidentalis) was somewhat less after 50 years under the restoration scenarios than under no management. However, the amount of resilient older forest structure and habitat for white-headed woodpecker (Leuconotopicus albolarvatus) was higher after 50 years under active management. More carbon was stored on this landscape without management than with management, despite the occurrence of high-severity wildfire. Our results and further applications of the model could be used in collaborative settings to facilitate discussion and development of policies and practices for fire-prone landscapes.

Participatory Modeling to Assess Climate Impacts on Water Resources in the Big Wood Basin, Idaho

Many of the natural resource issues we face today cannot be fully understood using normal scientific methods or theories. When uncertainties or decision stakes are high, participation by an “extended peer community” can be critical to studying and managing complex systems since local expertise on land use and ecosystem processes may not otherwise be captured. We conducted a case study of participatory modeling to support natural resource management under a changing climate. Over a 2-year period, we convened a Knowledge to Action Network (KTAN) of scientists and stakeholders interested in exploring the impacts of climate change in Idaho’s Big Wood Basin. Through a series of workshops and webinars, participants worked together to (1) pose relevant research questions collaboratively and iteratively, (2) create models of the future landscape under a changing climate regime to explore their questions, and (3) increase knowledge about how scientific information is both used and generated within those models in order to interpret the results. We began by developing a conceptual model of the river basin and progressed to two quantitative models—a system dynamics model and a spatially-explicit integrated model—in order to explore research questions generated by the KTAN. Alternative scenarios were developed and simulated in the final quantitative model to assess the range of impacts that external drivers and management decisions could have on the system.

Analyzing the relationship between urbanization, food supply and demand, and irrigation requirements in Jordan

The landscape surrounding urban areas is often used as farmland. With the observed expansion of urban areas over the last decades and a projected continuation of this trend, our objective was to analyze how urbanization affects food supply and demand in The Hashemite Kingdom of Jordan. We used a chain of simulation models covering components of the atmosphere (climate simulations), biosphere (crop yield calculations), and anthroposphere (simulations of urban expansion and land-use change) to calculate the effect of farmland displacement on land and water resources (hydrosphere). Our simulations show that the displacement of farmland itself has hardly any effect on cropland demand, crop yields, or irrigation water requirements. These results indicate that Jordan has sufficient productive areas available to buffer effects of urban expansion on food production for the next decades. However, this picture changes dramatically once we include changes in socioeconomy and climate in our simulations. The isolated effect of climate change results in an expected increase in irrigation water requirements of 19 MCM by 2025 and 64 MCM by 2050. It furthermore leads to an increase in cropland area of 147 km2 by 2025 and 265 km2 by 2050. While the combined analysis of urban expansion, climate change, and socioeconomic change makes optimistic assumptions on the increase in crop yields by 2050, the results still indicate a pronounced effect on cropland demands (2700 km2) and a steep increase in irrigation water requirements (439 MCM). Our simulation results highlight the importance of high resolution, spatially explicit projections of future land changes as well as the importance of spatiotemporal scenario studies at the regional level to help improving water planning strategies.