Penn State logo
Search: This site | People | Departments | Penn State
Riparia
wetland photo
Related Links

research project details

Project Title: Analysis of BMP implementation, performance, and maintenance in Spring Creek, an agriculturally-influenced watershed in Pennsylvania
Investigator(s): R. P. Brooks, R. F. Carline, D. Weller, G. Constantz, R. Stedman, J. S. Shortle, K. Boomer, M. McK. Easterling, J. A. Bishop, K. Mielcarek, J. Beck, G. Smith, C. Pfeiffer (Pennsylvania State University, Smithsonian Environmental Research Center, Canaan Valley Institute, in collaboration with USDA-ARS Laboratory)
Sponsor: U.S. Department of Agriculture, CEAP Watershed Projects 2006

Research Project Objectives: The study has four interrelated objectives and corresponding questions/hypotheses, as described below:

  1. Landscape Characterization - Coarse vs. Fine Resolution GIS Analyses Objective: To determine at what spatial scale [coarse (=30 m) vs. fine (< 1m) remote data resolution; reach vs. subwatershed] the performance of BMPs can be assessed and subsequently aggregated on a watershed basis.

    Questions/ Hypotheses: BMP performance at the watershed scale can be more accurately assessed, understood, and predicted by using GIS analyses based on fine-resolution data (LIDAR and digital photography) that distinguishes among different hydrogeomorphic (HGM) settings, than similar analyses using readily available coarse-resolution data.

  2. Hydrologic and Landscape Modeling of BMP Performance Objective: To quantify the relative importance of agricultural uplands, riparian zones, and streams channels as sediment sources, and to quantify the relative benefits of upland, riparian, and in-stream BMPs on stream sediment loads.

    3. Questions/ Hypotheses:

    • Are sediment loads affected more by in-stream processes (e.g., stream bank erosion), by riparian buffer quality, or by upland factors (e.g., erosion, runoff)?
    • Are in-stream, riparian, or upland BMP’s more effective in reducing sediment loads?
    • How can in-stream, upland, and riparian BMPs be combined to maximize water quality benefits on a watershed basis?

      We propose to use a multi-model framework to evaluate how landscape characteristics at different scales influence stream water quality. Results will improve our ability to predict the relative importance of near-stream versus upland factors and assess the value of applying BMPs in different parts of the watershed (e.g., in-stream, within a stream buffer, or across the agricultural lands in a watershed) that have different HGM settings.

  3. Ground-based Monitoring and Ecological Analyses

    4. Objectives:

    • To determine if reach-level measures of abiotic (substrate composition) and biotic (macroinvertebrates and fish) variables are useful surrogate measures of water quality to describe effects of BMPs on stream ecosystems.
    • To determine if reach-level abiotic and biotic variables vary spatially with regard to the number of BMPs implemented upstream of study reaches.
    • To assess the time required to measure stream responses to BMPs over a range of 1 to 12 years.

    Questions/ Hypotheses:

    • Can other abiotic or biotic variables be used as surrogates of traditional (but costly) water quality measures to assess the effectiveness of BMPs? Hypothesis: Integrative abiotic and biotic measures of water quality provide better and more efficient assessments of BMP performance than traditional chemical measures of water quality. We have simultaneously collected data on chemical, physical, and biological factors from several experimental units to use for these comparisons.
    • If one is attempting to use reach-level measures of abiotic and biotic variables, what is the relation between spatial distribution of the reaches, the locations of BMPs, and the degree of measured responses? Hypothesis: Reach-level responses are affected by the number, type, and HGM setting of upstream BMPs. By combining spatially explicit response data with landscape analyses and hydrologic modeling we can characterize the BMP factors that affect downstream responses.
    • Measuring stream responses to BMPs is costly process. To make the most efficient use of post-treatment monitoring funds, one should begin monitoring after treated areas have stabilized. How long should one wait before instituting post-treatment monitoring? Hypothesis: There is an optimal time after which to begin post-construction monitoring of BMP performance. This will lead to recommendations for monitoring on future projects.
  4. Socio-Economic Analyses and Outreach

    5. Objectives:

    • To explore the factors that affect effective farmer adoption of BMPs and citizen perception of water quality.
    • To develop methods for integrating and communicating the complexity of water quality measures derived from multiple measures and multiple sites to stakeholders by watershed. Our objective is to increase understanding of water quality changes resulting from implementation of BMPs by communicating study results through various media (e.g., workshops, websites, etc.).

    Questions/ Hypotheses:

    • Hypothesis: The reasons that farmers will or will not implement and maintain BMPs can be understood using properly designed and implemented social science instruments. Similarly, the perceptions of citizens within a watershed can be documented and understood using these techniques.
    • Hypothesis: Standardized procedures can be developed to manage a diversity of data types, and communicate results to stakeholders. (Using our extensive experience in engaging a wide range of stakeholders, we will conduct surveys and workshops within our study watersheds to gauge the acceptance and understanding of BMP performance measures.)

Progress Report: (study in progress)

Products: None available at this time

Summary:

Assessing the performance of conservation BMPs in agricultural watersheds across the U.S. is extraordinarily difficult because of the diversity of conservation practices, the variety of hydrogeomorphic (HGM) settings in which they are implemented, and the breadth of geographic regions involved. Spring Creek watershed, an intensively studied 11-digit HUC watershed in central Pennsylvania (see Figure 1), has the necessary long-term data on water quality for assessing a suite of BMPs in an agricultural watershed.

Our investigative team comes from three institutions – Pennsylvania State University, Smithsonian Environmental Research Center, and Canaan Valley Institute - representing the fields of stream and wetland ecology, landscape ecology, fisheries biology, hydrology, resource economics, and rural sociology. We have carefully selected experimental units to use that focus our data collection and analysis on BMP performance, and allowing us to factor out the impact of underlying HGM variation. Using a combination of field data, landscape analyses, and hydrologic modeling from the experimental reaches and other watershed segments, we will determine if evaluations at the reach scale can be compiled predictively to characterize the condition of watersheds. By combining ground-based measurements with fine-resolution LIDAR and digital photography data, we will be able to more precisely map topographically and hydrologically distinct reaches (e.g., HGM settings) that vary in their response to BMPs both spatially and temporally. Building on recent surveys in Spring Creek, new focus groups, interviews, and surveys will be conducted to gain more understanding about the motivations to participate in BMP implementation, maintenance, and monitoring. The study will be conducted over 3 years (2007-2009).

Map Spring Creek Watershed Figure 1. Spring Creek Watershed, Pennsylvania with subwatersheds, ambient water quality sampling stations of the Spring Creek Watershed Community (red dots), and treated stream/riparian reaches (Slab Cabin and Cedar Run subwatersheds – highlighted reaches).