Thesis research

Remote sensing for monitoring of wildlife habitat: Lesser Snow Geese and sub-arctic coastal marshes

Statement of Problem

Large numbers of snow geese are causing widespread destruction of the vegetation of their breeding habitats in the Arctic. Satellite imagery can be used to assess habitat destruction over large spatial and temporal scales and this data, along with ground-based information, may allow management targets for sustainable snow goose population sizes to be set.

Introduction

The number of lesser snow geese in eastern North American flyways has been increasing rapidly, and has approximately tripled between 1969 and 1993. This has had severe impacts on their breeding grounds, since the geese forage destructively between the time of their arrival and the time when above-ground plant growth begins. Because the geese depend on these same sites (coastal salt marshes) for summer grazing, the availability of summer forage has been sharply reduced by this spring damage. Heavy grazing on the remaining areas has reduced plant growth and reduced forage availability still further. At La Perouse Bay, Manitoba (the site of a breeding colony of lesser snow geese that has been studied since 1968), reduced growth and recruitment of goslings has been detected and linked to the decline of summer forage. Despite reduced breeding success, the population continues to increase, largely due to increased adult survivorship. Recovery of the salt marsh vegetation following spring damage is very slow (>15 years), particularly when heavy goose use continues. Without intervention, the destruction can be expected to continue.
Because the damage to breeding areas is large-scale and long-term, there are relatively few sources of data with which to quantify it. One of these sources is satellite data, which have been provided by Landsat since the early 1970's. Reflectance data collected by satellite are consistent, reliably archived, and sample large areas non-destructively. To assess the pattern and timing of vegetation changes, the historic vegetation first needs to be mapped. Measures of vegetation condition then need to be linked to reflectance using ground-based measurements. These data can then be scaled up to regional levels and the regional damage to vegetation types used by geese can be assessed using Landsat imagery.

On-line Poster

This poster, presented at the Ecological Society of America conference in Utah, August 2000, summarizes a good chunk of my thesis; Fig. 2 is the vegetation reconstruction. Since the poster was presented to ecologists, some remote sensing terms are defined.

 

Major papers:

2002. Gadallah, F.L. Historical vegetation reconstruction of a degraded sub-arctic coastal marsh using Landsat imagery and ancillary data. Écoscience 9:271-279. abstract

2000. Gadallah, F.L., Csillag, F. and Smith, E.J.M. Destriping multisensor imagery with moment matching. International Journal of Remote Sensing 21:2505-2511. abstract link to paper

1995a. Gadallah, F.L. and Jefferies, R.L. Comparison of the nutrient contents of the principal forage plants utilized by lesser snow geese on summer breeding grounds. Journal of Applied Ecology 32:263-275. abstract (pdf) JSTOR link

1995b. Gadallah, F.L. and Jefferies, R.L. Forage quality in brood rearing areas of the lesser snow goose and the growth of captive goslings. Journal of Applied Ecology 32:276-287. abstract (pdf) JSTOR link

PDF reprints are available.