Recipient Organization
BLUE FOREST CONSERVATION
171 5th St
Lake Oswego,OR 97034-3029
Performing Department
(N/A)
Non Technical Summary
Forests across the Western U.S. are at a tipping point - overgrowth, a warming climate, drought, and insect infestations have ravaged tens of millions of acres of land, increasing the risk of wildfire and threatening water resources, air quality, communities, homes, and habitat. Forest restoration (the strategic removal of brush and shrubs and the selective thinning of trees to return forests to a healthier state) is a proven, but often underutilized, tool that can decrease these risks while also increasing water quantity. While forest restoration is well known to decrease the risk of severe wildfire, its impact on water quantity in regions such as California are not as well understood. As a result, potential beneficiaries of restoration like utilities do not account for its value when analyzing the cost/benefit of an investment in restoration projects. Without the ability to account for water quantity considerations, utilities often cannot make the economic case to invest in restoration projects as the benefits are too uncertain. We intend to change that through advances in research paired with an innovative public-private financing model that will empower utilities and other downstream beneficiaries to invest in forest health.For Phase II, we propose to continue to develop a remote-sensing-based watershed-scale toolkit for determining water-yield changes following forest restoration in California's Sierra Nevada, applied to a specific project as well as to the larger landscape to identify where opportunities for water-yield gains may be greatest. To create this framework, we will build upon our Phase I work to evaluate the accuracy of existing physically based hydrologic models, compared to results predicted using different remote sensing data for assessing water-yield changes following forest restoration.The purpose of the toolkit is to enable a more comprehensive approach to prioritizing restoration and reduce the uncertainty of water benefits associated with restoration projects. Success for this technology consists of widespread adoption by utilities and other potential customers across California, with the potential to scale to other high-exposure western states. The basic toolkit, which is likely to consist of a visualization dashboard summarizing expected water quantity benefits, can be offered as a free resource to all utilities, water-dependent companies, and land managers with a paid option for more in-depth analysis. The free toolkit will allow a customer to quantify and visualize fire risk and potential for water quantity gains on a watershed level. In order to determine the precise impacts to water quantity as a result of a restoration project in a specific location, the granular analysis option would be required with the customer either hiring our researchers for a consulting engagement and/or moving forward with us to package watershed investments. For those watersheds that would be a good fit for investment, the toolkit could serve as a powerful lead generation device.
Animal Health Component
40%
Research Effort Categories
Basic
20%
Applied
40%
Developmental
40%
Goals / Objectives
Our overall R&D goal is to develop a customizable and scalable landscape scale water quantity evaluation framework that can be broadly applied to quantify changes in watershed yield following forest restoration across the western US. This framework and associated dashboard will 1) provide a stakeholder engagement tool for non-experts that communicates the potential of forest restoration to result in water quantity benefit and wildfire risk reduction, and 2) enables Blue Forest Conservation to better evaluate opportunities for forest restoration and constrain uncertainty from prediction of economic value to stakeholders. Shifting the remote sensing-based evapotranspiration estimate from MODIS to Landsat will improve the resolution of the NDVI-ET product, improve detection of lighter vegetation changes over smaller areas, and reduce uncertainty of landscape-scale water balance calculations from these spatial data. The web interface will be designed with water utilities and stakeholders in mind, with the goal of communicating water benefits and wildfire risk to non-technical personnel such as governing boards, program managers, and other decision-makers with a wide variety of backgrounds at the very small to very large organizations with whom we work.Technical Objectives of Phase IIDetermine areas of potential water yield enhancement in California using PRISM precipitation data and LandSat NDVI-based evapotranspiration data.Evaluate changes to vegetation and evapotranspiration following historical forest management actions and wildfires on federal and private land in California.Implement the remote sensing water yield assessment for the North Yuba River pilot project. Research if the water utility stakeholder finds the forest water use data (evapotranspiration) sufficiently useful for stakeholders to estimate water quantity gains.Develop an interactive map to communicate water yield enhancement opportunities and wildfire hazard risk covering within California. Collaborate with utility stakeholders to design the optimal data visualization dashboard for communicating forest restoration benefits.
Project Methods
Phase II consists of three tasks and all Phase II activities will be overseen by Co-PI's Nicholas Wobbrock and Dr. Phil Saksa of BFC with research performed by Dr. Roger Bales, Dr. Ben Bryant, and colleagues.Task 1: Develop a statewide spatially-explicit water balance for California to evaluate historical changes to forest ecosystems and water use, and determine where water yield enhancement opportunities exist in the state.This task will combine spatially-gridded precipitation data (PRISM, 800 meter resolution) with evapotranspiration calculated from NDVI (Landsat, 30 m resolution), to estimate spatially-explicit runoff (30 m resolution) for California from 1985 to 2017. The runoff product will be based on a simple water balance equation that assumes long-term subsurface storage is a neutral term:Runoff = Precipitation - EvapotranspirationThe runoff product will be validated for the Sierra Nevada using the full natural flow calculated by the California Department of Water Resources for each of the 12 major watersheds, providing 396 total years of hydrology data for comprehensive evaluation.The spatially-explicit runoff will then be used to determine where in California water yield enhancement opportunities exist. This simple remote sensing approach provides a first order assessment of potential water yield benefits with forest restoration efforts. For ease of communicating water yield enhancement opportunities to stakeholders, initial maps will coarsely be divided into three tiers: High, moderate, and low water yield potential.Additional granularity would be available for specific regions, utilities, and forest restoration projects with stakeholders who have engaged Blue Forest Conservation to develop a Forest Resilience Bond.The products developed in this task will enable Blue Forest Conservation to engage utility stakeholders with a specific forest restoration project and show 1) the potential to enhance water yield, 2) an estimate of the additional volume of water that could be expected, and 3) an estimate of the vegetation recovery rate.Task 2: Implement water yield assessment on a Forest Resilience Bond pilot project.This task will result in the implementation of the remote sensing water yield assessment on the 150,000 acre North Yuba River forest restoration, the initial pilot project of the Forest Resilience Bond. Blue Forest Conservation has been collaborating with the Yuba County Water Agency (YCWA), Sierra Nevada Conservancy, and the Tahoe National Forest supervisor's office as we develop the Forest Resilience Bond for the Yuba Project, all providing letters of support for this application. The North Yuba River provides 80% of YCWA's inflows to the 970,000 acre-foot New Bullards Bar Reservoir which is used for flood control, recreation, water supply storage, and hydropower production with the 340 MW New Colgate Powerhouse. We used the MODIS NDVI approach from Phase I to include a historical water balance in our Forest Resilience Bond proposal to Yuba County Water Agency.We also showed how using the remote sensing assessment can capture local changes to vegetation and evapotranspiration from disturbances such as the Bassetts Fire in 2006, ignited by a lightning storm. The wildfire decreased evapotranspiration by 8 cm the following year within the 2,000 acre fire perimeter, but changes to the water balance were not significant over the entire watershed. Vegetation recovery within the fire boundary is inconsistent and shows why it will be desirable for utilities to monitor regrowth post vegetation disturbance such as wildfire or restoration.The paired catchment remote sensing approach used in Phase I, updated with the Landsat NDVI estimation of evapotranspiration, will be used to determine changes to the water balance following forest restoration treatments during the initial two years of the project. This task will rely on a paired watershed Before-After Control-Impact (BACI) experimental design.Task 3: Develop interactive dashboard for stakeholder engagement and investment.This task will result in a spatial product that uses the tiered water yield opportunity developed in task 1 and overlays the risk of a large high-severity wildfire in California. Annual wildfire risk of forested regions is produced by the U.S. Forest Service from a nationwide fire modeling effort using FSim, which provides an absolute indication of annual fire risk at a point in time. While adequate for characterizing relative risk across space, this metric does not straightforwardly translate to metrics of actionable interest to stakeholders. For example, a 2% annual burn probability may register as relatively minor, but such a probability implies a greater than 50% chance of fire in the watershed over a 35-year period (or 1% over 30 years ≥ 25% chance of fire). Starting with burn probabilities over long time periods, we will explore which metrics are of most interest to stakeholders.Importantly, the fire risk estimates interact with the water yield impact, because water yield is a function of vegetation state, which is both a function of time since treatment and time to disturbance (fire). Therefore, when estimating the expected value of water yield benefits from restoration activities, the fire risk should be taken into account. Drawing on existing process-based modeling frameworks, NDVI time series, and field measurements described in the literature, we will use a generalized logistic functional form and a state-transition model to characterize the vegetation state over time, and in-turn refine the post-treatment water benefits to account for regrowth and fire dynamics. These calculations will be a new contribution to this development space, which to date has emphasized water yield impacts without interacting growth and fire dynamics.All of the spatial information described in this task will be displayed in a dashboard that can be used by Blue Forest to screen and vet opportunities, and to engage stakeholders in understanding the connections between the health of their watershed and the benefits they may realize. The various proposed dashboard features will be explored with stakeholders as we research what visualization layout, data, and summary metrics are most useful for them to make decisions regarding resources and land management, and which are most helpful for champions to communicate the value of forest restoration to their own stakeholders.Besides simply toggling between display of individual spatial layers, we expect the dashboard will have the following capabilities, some of which we will construct in anticipation of stakeholder needs to explore further, and some of which can be expanded during iterative testing:Dynamically adjust display resolution to facilitate regional level exploration, or zooming on a specific watershed.Adjust-single layer input values: For individual layers, users can adjust key input values through manual input or slider bars. For example, a user could enter 10, 20 or 30-year fire probabilities depending on what they found most decision-relevant. Similarly, the user could adjust the marginal dollar value assigned to additional water yield or the discount rate.Create a weighted blend of specific layers: Highlight "hot-spots" as a weighted product of fire risk and water benefits, to assess the where there are synergies in reduction in fire risk and water yield gains, versus where they are in tension.Visualize the layers in context by overlaying on base maps such as contour, terrain, satellite or common maps including locations of cities, water and utilities infrastructures.