Forest Investment Account (FIA) - Forest Science Program
FIA Project Y071166

    Application of remote sensing to detect changes in vegetation structure from differing land treatments in the Kootenay Region of Southeastern British Columbia
 
Project lead: Magai, Robert
Contributing Authors: Magai, Robert; Marcoux, Derek; Curran, Mike P.; Magai, Margaret
Imprint: [BC] :, 2007
Subject: Forest Investment Account (FIA), Soils, Analysis, British Columbia
Series: Forest Investment Account (FIA) - Forest Science Program
Description:
This proposal will establish a different approach by which to evaluate and estimate the impact of soil disturbance on ecosystem productivity and range health in the Long Term Soil Productivity (LTSP) study plots. The LTSP Studies were established to investigate the response of plant communities to various levels of soil disturbance (organic mater removal and soil compaction). Field-based monitoring is the principal means used to collect information on plant species presence and distribution in the LTSP plots. The percent cover of native vegetation of the ten dominant species present is monitored in mid-summer (Holcomb 1996). Such monitoring is conducted usually by qualitative visual assessment and/or quantitative assessment using statistically based plot sampling designs (Pitt et al. 1997). Field-based vegetation monitoring approaches are costly and labor intensive. Advancement in remote sensing techniques for extracting biophysical data from high definition digital aerial photography and high-resolution satellite imagery has drawn quite a lot of attention in the last decade. Digital image methods for assessing vegetation conditions are now available for operational use. These methods have the potential to provide the required information in a more objective manner, at lower cost, and with greater coverage than is attainable using field sampling. Remote sensing and image interpretation methods have been utilized in forestry management for many years. The remote sensing approach has been applied in various tasks ranging from forest thematic mapping to the detailed tree or stand characteristics survey. Vegetation is an excellent indicator of ecosystem health since it provides an estimate of the biophysical conditions of an ecosystem unit. Hence, vegetation can be used as a surrogate for determining underlying soil conditions and vice versa. The rationale of assessing and monitoring vegetation health is based on comparing plant species present on various treated or disturbed sites in relation to an undisturbed site. The control sites represent the expected successional end state as dictated by the ecological zone conditions. This project is part of a MSc. thesis through Royal Roads University. Partners in the research are the B.C. Ministry of Forests and Range and Selkirk College. The proposal will explore the feasibility of estimating percent plant cover on disturbed sites. It will also demonstrate the capability of eCognition (Definiens Imaging, Germany), an object oriented image analysis software to classification plant communities from high resolution digital photography. This research will answer the question: 'Can remote sensing technology be used to detect changes in vegetation in response to differing soil disturbance levels?' The research will be conducted on three (3) existing LTSP research installations in the east Kootenay. Five year vegetation data on treated sites has been collected on two of the three sites and the third site will be sampled this year. Data for the response of shrubs, forbs, graminoids, and mosses will be grouped and evaluated to determine what, if any change, has occurred due to differing treatments (soil compaction and organic matter removal). High resolution aerial photography at 10 cm resolution was taken on all five sites in August 2005. Based on preliminary evaluation, it appears that this resolution will permit detection of differing vegetation types, and thus be applicable for using vegetation as an indicator of land treatment and management changes. Digital imagery will be analyzed at the Selkirk College Geospatial Research Center (SGRC) in Castlegar, B.C. Plant cover estimates and species identifications will be compared to the field derived measurements to validate the results. Results of this research will have far reaching consequences on development of an acceptable protocol for assessing the effects of soil disturbance on vegetation conditions in operational applications like range condition monitoring assessments and also in research applications like on other LTSP installations.
Contact: Magai, Robert, (250) 365-1349, rmagai@selkirk.ca

    Deliverables:

Executive summary (22Kb)

To view PDF documents you need Adobe Acrobat Reader, available free from the Adobe Web Site.

Updated August 16, 2010 

Search for other  FIA reports or other Ministry of Forests and Range publications.

Please direct questions or comments regarding publications to For.Prodres@gov.bc.ca