Landsats Aff

Bio-D – Solvency – Quantification

Remote sensing is key to quantifying bio-d loss for organizations – tech advancements increase its usefulness
Turner et al 3 (Woody Turner1, Sacha Spector2, Ned Gardiner2, Matthew Fladeland3, Eleanor Sterling2 and Marc Steininger4 1NASA Office of Earth Science, 2Center for Biodiversity and Conservation, American Museum of Natural History, 3Earth Science Division, NASA Ames Research Center, 4Center for Applied Biodiversity Science at Conservation International, TRENDS in Ecology and Evolution Vol.18 No.6 June, p. 306-314 “Remote sensing for biodiversity science and conservation” JMB)

The potential for modern sensors to identify areas of significance to biodiversity, predict species distributions and model community responses to environmental and anthropogenic changes is an important research topic. Underlying this effort is the assumption that certain key environmental parameters, with remotely detectable biophysical properties, drive the distribution and abundance of species across landscapes and determine how they occupy habitats. New imagery and data sets are now enabling remote sensing, in conjunction with ecological models, to shed more light on some of the fundamental questions regarding biodiversity. These tools should prove useful to those seeking to generate basic knowledge about why organisms are found where they are, as well as those asking the more applied question of where to invest conservation funds. Here, we use the term ‘biodiversity’ in its organismal sense to refer to species and certain characteristics of species, in particular their distribution and number within a given area. We also use ‘biodiversity’ more broadly to mean species assemblages and ecological communities (i.e. groups of interacting and interdependent species). There are two general approaches to the remote sensing of biodiversity. One is the direct remote sensing of individual organisms, species assemblages, or ecological communities from airborne or satellite sensors. New spaceborne systems with very high spatial (also known as hyperspatial) resolutions are now available from commercial sources. For the first time, the direct remote sensing of certain large organisms and many communities is possible with unclassified satellite imagery. Likewise, new hyperspectral sensors slice the electromagnetic spectrum into many more discrete spectral bands, enabling the detection of spectral signatures that are characteristic of certain plant species or communities. The other approach is the indirect remote sensing of biodiversity through reliance on environmental parameters as proxies. For example, many species are restricted to discrete habitats, such as a woodland, grassland, or seagrass beds that can be clearly identified remotely. By combining information about the known habitat requirements of species with maps of land cover derived from satellite imagery, precise estimates of potential species ranges and patterns of species richness are possible. Just such an approach has been employed extensively in the US GAP analysis program [1]. Of course, it is probable that no single environmental parameter drives patterns of species distribution and richness. Many possible drivers have been proposed (Table 1). Here, we focus on three often-cited environmental parameters that now lend themselves particularly well to detection because of recent advances in remote-sensing technology: primary productivity, climate and habitat structure (including topography) [2–5]. For the conservation biologist, remotely sensed imagery exposes land-cover changes at spatial scales from local to continental, letting one monitor the pace of habitat loss and conversion [6,7]. These measurements of habitat loss can be converted into quantitative estimates of biodiversity loss through the use of the species–area relationship (Box 2), which underlies many current estimates of biodiversity decline [8–12].Remote sensing provides the area component of the equation. Public and nongovernmental conservation organizations worldwide leverage their understanding of species–area relationships with imagery-based habitat classifications to estimate species losses associated with changes inland cover and land use(Box3).The challenge is to go beyond this approach to a more detailed understanding of which species are being lost and why. How can we match existing and emerging remote-sensing technologies to parameters that have clear implications for organisms and ecosystems? Here, we review evidence that indicates that we might be close to improving greatly the detection of species, ecological communities and patterns of species richness with remote sensing. We explore recent advances in technology, addressing direct and indirect approaches to the remote sensing of biodiversity. Following the discussion of each technology, we offer examples of applications of that technology to the issue at hand.

Bio-D – Solvency – Deforestation

Landsat key to regional forestry analysis
Harris et al 5 (Grant M., Clinton N. Jenkins, and Stuart L. Pimm, Nicholas School of the Environment and Earth Sciences at Duke, http://www.terpconnect.umd.edu/~cnjenkin/Harris_et_al_2005.pdf, accessed 7-6-11, JMB)

Generating forest maps at regional scales requires satellite imagery analysis. It is important to match this objective with suitable types and formats of input imagery. A primary consideration is the imagery’s spatial resolution. Generally, the smaller an image’s spatial resolution ( pixel size), the less area it maps. For example, were we to use Landsat ETM+ imagery (30 × 30 m) to map the entire Atlantic Forest, it would require approximately 75 scenes. The computational issues are significant (and the imagery costs are expensive, up to US$45,000). Additionally, we already know that more than 90% of the region is deforested and has little, if any, conservation value ( Harris & Pimm 2004). Our application demands mapping the entire region, but applying considerable effort to cover vast areas we know are deforested seems inappropriate. Region wide, an ecologically relevant map does not require high spatial detail. For example, the resolution of regional sensors spans 250 m to 1 km. Such imagery supplies adequate resolution for prioritizing subregions from an expanse >1 million km2 . Overall, maps generated from regional sensors provide an accurate picture of where forested habitat remains. To evaluate the types of satellite imagery designed for regional analyses, we used three different sources (AVHRR, SPOT VGT, and MODIS). We also evaluated the ability of a jpg composite based on Landsat TM data (the GeoCover Landsat TM mosaic, produced by the Earth Satellite Corporation, Rockville, Maryland) to map the Atlantic Forest. Lastly, our analysis included two preclassified products (MODIS Continuous Fields and a MODIS derived landcover based on MODIS imagery)
Landsat provides higher-accuracy forest data
Harris et al 5 (Grant M., Clinton N. Jenkins, and Stuart L. Pimm, Nicholas School of the Environment and Earth Sciences at Duke, http://www.terpconnect.umd.edu/~cnjenkin/Harris_et_al_2005.pdf, accessed 7-6-11, JMB)

To evaluate the location and quality of forest mapped, we compared each prediction’s spatial attributes against the Landsat ETM+-based forest cover (Fig. 3). All comparisons had forest fragments of <1 km2 removed. Such areas generally have little long-term conservation value for birds (Willis 1979; Stouffer & Bierregaard 1995; Brooks et al. 1999; Ferraz et al. 2003) other than their usefulness in building corridors at finer scales. We began by quantifying the percentage of the Landsat ETM+ forest prediction inside every 1 × 1 km pixel on the other maps. (This was also performed at 500 × 500 m for relevant predictions.) These percentages were binned into 10 equal categories, with the bin values corresponding to the proportion of Landsat ETM+ forest predicted in the coarser cells (Fig. 3, horizontal axis). The larger the bin value, the more Landsat ETM+ forest area was predicted in each of the coarser cells (indicating greater amounts of forest). Fragmented areas formed bins 5 and below ( less than 50% forest coverage). We calculated the intersection between these bins and all pixels predicted to be forest for each of the maps. For example, the upper left grid (Fig. 3) contains 16 hypothetical cells. The values in each cell represented the percentage of the Landsat ETM+ forest predicted inside them (now at the same resolution as the regional sensor with which it is being compared). The light gray shading indicates another forest map that also predicts some of these cells as forest. Of the four pixels that contained 95% forest according to Landsat ETM+ (bin 10), our coarser forest map captured three of them (75%). In other words, it predicted three cells that are 95% covered with the Landsat ETM+ forest prediction. The coarser forest prediction also mapped 60% of the pixels in bin 8 (71–80% covered by the Landsat ETM+ prediction), 67% in bin 7, and 50% in bins 3 and 4. We evaluated the six forest covers in this manner and calculated the percentage of times they intersected the bins (Fig. 3, vertical axis). Because the GeoCover Landsat TM mosaic is also resolved at 30 × 30 m, before this comparison we calculated its percent forest in 500 × 500 m pixels and considered pixels ≥60% to be forest. Lastly, we investigated the fractal dimension of areas in the Landsat ETM+ forest prediction missed by the coarser forest maps (mixed pixels). At times, the coarser resolution sensors predicted forest for pixels that were only half covered with forest according to the finer resolved Landsat ETM+ sensor ( bin 5). If the spatial arrangement of the 30 × 30 m pixels predicted to be forest by the Landsat ETM+ were spread out, it would help explain why the1 × 1 km or 500 × 500 m sensors did not map them. On the other hand, the coarse sensors may include these pixels as forest if their distribution were clumped. We found no differences in fractal dimension between these mixed pixels classified and unclassified as forest by the coarse sensors.