Using satellite imagery to assess macrophyte response to water-level manipulations in the Saskatchewan River Delta, Manitoba / Mark S. Baschuk, Michael D. Ervin, William R. Clark, Llwellyn M. Armstrong, Dale A. Wrubleski, and Gordon L. Goldsborough.
Material type: TextSeries: Wetlands. 32(6) 1091-1102 Publication details: 2012.Description: illustrations ; 28 cmLOC classification:- BAS
Item type | Current library | Collection | Call number | Status | Date due | Barcode |
---|---|---|---|---|---|---|
Electronic Journal | IWWR Supported Research | Non-fiction | BAS (Browse shelf(Opens below)) | Available | 16784 |
Browsing IWWR Supported Research shelves Close shelf browser (Hides shelf browser)
Includes bibliographical references (pages 1101-1102).
QuickBird satellite imagery and object-based
classification were used to map emergent macrophyte response
in wetlands that were partially drawn down during a
three-year period (2007–2010) in the Saskatchewan River
Delta, Manitoba, Canada. Generalized logistic models were
used to classify vegetation genera in three drawdown and
three high-water (control) wetlands. User accuracy of models
averaged 74 %, and was lower in partial drawdown
wetlands (72 %) than in high-water wetlands (77 %). In
partial drawdown wetlands, emergent macrophytes (especially
Typha spp. and Carex spp.) senesced significantly
after the first year of lowered water levels but senescent
areas decreased over following years. Plants were not completely
dead in dewatered areas and emergent stands recovered
by the third year after drawdown. Open water area in
the partial drawdown wetlands gradually decreased due to
clonal expansion of Sparganium spp. Vegetation areas in
high-water wetlands remained unchanged throughout the
study period. Although partial water-level drawdowns temporarily
influenced the growth of emergent macrophytes
they did not expose the seed bank and did not result in
substantial, prolonged changes in spatial arrangement of
vegetation areas. The combination of high-resolution satellite
imagery, object-based classification and logistic modeling
was useful for documenting vegetation response to
management in these remote, boreal wetland systems.