WWETAC Projects

Project Title: Development of GIS-based maps for the "top 10" invasive weeds in Crook County, Oregon

Status: Completed

Principal Investigators: Sim Ogle and Steve Dougill

E-mail Contact: Sim Ogle, sim.ogle[at]co.crook.or.us; Steve Dougill, steve.dougil[at]co.crook.or.us

Web: Crook County GIS Department

Key Issues/Problems Addressed: Crook County is a rapidly developing county located in central Oregon covering 2,991 mi2. The county's economy is driven by agriculture, forestry, rangeland management, and industrial activity. The county's geographic location, climate, geology, land uses, transportation network and multiple off-highway recreational attractions make it susceptible to noxious weeds.

Federal, state, and local agencies have all had a hand in monitoring and combating the spread of noxious weeds, but until this project a single publicly accessible point of access for this data and its spatial mapping by geographic information system (GIS) had not occurred. Additionally the five surrounding counties had not communicated on a broad scale about the noxious weed problems in their jurisdictions; therefore part of our project was to develop contacts within these counties and map any available data.

The project couples GIS, remote sensing, field sampling, field validation of likely weed spread models, and GPS photographic techniques to continue updating the presence, absence and spread of noxious weeds in Crook County using a statistically viable and repeatable methodology.

Study Objectives, Goals and Description:

Original Proposal

To develop a series of GIS-based maps that show the distribution of the "top 10" invasive weeds in Crook County, Oregon, by integrating local, state, and federal information. Use the map-based distributional and geographic features information along with knowledge of the invasive species' biology to predict future spread and intensification of these weeds.

Mod 1

To fine-tune GIS-based predictive models of invasive weeds through the correlation of model results and field sampling. To partner with invasive weed monitoring programs in Deschutes, Jefferson, Wheeler, Grant, and Harney Counties and, where possible, share and integrate invasive weeds information into one interactive GIS based mapping system. The final purpose will be to investigate the use, techniques and availability of remote sensing techniques for the detection and classification of invasive weeds on a countywide or regional scale.

Mod 2

To conduct field research and mapping efforts in the Beaver Creek sub-basin and South Fork of Beaver Creek, and the upper Crooked River watersheds. These new sources of data will fill in gaps in the counties' invasive weeds mapping efforts and lead to improved statistical and spatial analyses. Efforts to gather and analyze current invasive weed data collected in the field by local, state, and federal agencies will also continue.

Secondly, county GIS will assist in organizing and deploying weed discovery field personnel. These personnel will be equipped with hand-held Trimble GeoXT GPS/GIS units to gather spatio-temporal invasive data. Field teams will be using statistically relevant sampling methodologies reviewed and recommended by the Forest Service, Forestry and Range Sciences Lab.

Finally, the data gathered from all sources in 2007 will be used to improve County weed modeling efforts. The county GIS field research to verify model data by recording presence and absence in targeted areas. This sampling effort will use the same methodology as the discovery field teams. All the information will be analyzed statistically and communicated to the public via an updated weeds Web and FTP site and communicated to the public via an updated Weeds web and FTP site and in the form of new maps distributed to team members and posted for free downloading.

Status:

Original Proposal - Complete

Mod 1 - 80 percent Complete (still waiting for available data from neighboring counties)

Mod 2 - Began July 1st 2007

Key Findings:

Original Proposal - By May 2006 the current noxious weed picture (now at 17 weed species) for Crook County had been mapped by integrating federal, state and local data sets and knowledge. Following a review of techniques and methodology, a GIS-based modeling analysis was conducted to predict the likely areas of noxious weed growth based on climate, elevation, slope, soil type/geology, and current weed locations. These maps were printed, distributed to field teams for verification and comments and ultimately provided free of charge through an interactive ESRI ArcIMS GIS mapping resource built into a project Web site. The Web site contains more detailed methodology/results and also has downloadable maps and URL link resources (see project URL's at the end of this document).

Mod 1 - N/A will be updated in 2007

Mod 2 - N/A

Deliverables:

Original Proposal

Phase I:

  • Conduct a comprehensive needs assessment.
  • Identify available current and historical data sources at local, state, and federal levels.
  • Consult with experts at OSU, WWETAC, US Forest Service, and Bureau of Land Management to identify and develop base mapping requirements (sample size, methodology).
  • Identify software and methods required to perform indepth statistical analyses in phase II.
  • Integrate all data into a geodatabase.
  • Develop top 10 list of invasive species for Crook County.
  • Produce invasive plant species distribution and treatment maps for Crook County covering all lands.
  • One 36 by 36 inch distribution map covering the county 1:150,000 scale.
  • Individual distribution maps (size to be determined) at higher resolution covering the top 10 species.
  • Create a basic Web site and basic interactive ArcIMS mapping system for displaying the results of phase I to all project participants.
  • Produce a brief report on phase I, in particular a review of the data sets in terms of suitability for spatial modeling and the analyses conducted.

Phase II:

  • Conduct trial analysis of data to test sample distribution for autocorrelation.
  • Develop and implement, along with WWETAC scientists and local experts, appropriate GIS-based statistical analyses and methods to model (predict) the distribution of invasive species.
  • Investigate relationship between species, habitat, acreage, environmental records (where available) and land use.
  • Produce brief Phase 2 report of the research into and the methodologies chosen to model weed distribution.

Mod 1

  1. Model tuning and validation:
  • Fine tune predictive GIS-based invasive weed analysis model.
  • Develop field sampling plan along with Crook County weed master and Forest Service to correlate models results with actual field locations.
  • Develop likelihood maps targeting model data with sample sites.
  • Conduct field sampling project.
  • Analyze field validation results and feed data back into model.
  • Develop maps of fine-tuned model data superimposed with field validation sample sites.
  • Produce brief report of the research into and the methodologies chosen to model weed distribution.
  1. Communication with surrounding counties:
  • Contact relevant personnel and conduct site visits (where possible) in Deschutes, Jefferson, Wheeler, Grant, and Harney counties
  • Demonstrate weed mapping project, methodology, and Web services.
  • Seek invasive weeds data-sharing relationships with surrounding counties.
  • Integrate data into Crook County GIS/WWETAC geodatabase.
  • Integrate data into a unified geodatabase.
  • Update interactive mapping system with latest data from 3.1 and 3.2
  • Produce a brief project report with an emphasis on results, lessons learned, and the big picture of invasive weeds in central Oregon.
  1. Investigate the use of remote sensing for weed detection and classification.
  • Investigate Remote Sensing techniques, methodology, and availability for weed detection and classification in Crook County and the surrounding area.
  • Identify the resources (hardware and software) necessary to complete this project.
  • Identify research papers and contacts prominent in this field.
  • Report findings to WWETAC personnel for evaluation and consideration.

Mod 2

Field project, model tuning and updates:

  • Provide personnel, material, organizational and analytical support to county summer 2007 weed discovery surveys in the upper Crooked River and Beaver Creek/South Fork subbasins.
  • Utilize field sampling methodologies recommended by Forestry and Range Sciences Lab personnel.
  • Add new survey data to project geodatabase.
  • Sample targeted weed locations as part of continued weed model tuning and verification process.
  • Statistically analyze data.
  • Update current status weed maps and project geodatabase.
  • Update and launch the county invasive weeds Web site and publish new data to interactive ArcIMS GIS mapping site.
  • Update project FTP site with new invasive data in the form of a distrbuted ESRI geodatabase.
  • Print and distribute 2007 weed maps to project team and local agencies fighting invasive weeds.

Select Background Citations:

Everitt, J.H., Escobar, D.E.; Davis, M.R. 2000. Reflectance and image characteristics of selected noxious rangeland species. Journal of Range Management. 54: A106-A120.

Glenna, N.F.; Mundta, J.T.; Weberb, K.T.; Pratherc, T.S.; Lassc, L.W.; Pettingilld, J. 2005. Hyperspectral data processing for repeat detection of small infestations of leafy spurge. Remote Sensing of Environment. 95: 399-412.

Lass, L.W.; Prather, T.S.; Glenn, N.F.; Weber, K.T.; Mundt, J.T.; Pettingill, J. 2004. A review of of spotted knapweed (Centaurea maculosa) and babysbreath (Gypsophila paniculata) with a hyperspectral sensor. Weed Science. 53(0043-1745): 242-251.

McGowen, I.; Frazier, P.; Orchard, P. 2001. Remote sensing for broadscale weed mapping— Is it possible?

Mississippi State University, Remote Sensing Technologies Center. 1999-2001 Project Final Reports.

Rew, L.J.; Pokorny, M.L., eds. 2006. Inventory and survey methods for nonindigenous plan species. Bozeman, MT: Montana State University Extension.

Smith, R.G.; Menalled, F.D. 2006. Integrated strategies for managing agricultural weeds: making cropping systems less susceptible to weed colonization and establishment. Department of Land Resources and Environmental Sciences. Bozeman, MT: Montana State University.

USDA Forest Service Remote Sensing Applications Center online resources. A weed managers guide to remote sensing and GIS.

Selected Links:

Project Links:

Project ID: FY06TS12