Integrating field data

While the savannah elephant (Loxodonta africana) is listed by the International Union for Conservation of Nature (IUCN) as “vulnerable” because of declining abundance in some regions of Africa (Blanc 2008), populations in some protected areas of South Africa are growing rapidly (van Aarde and Jackson 2007). These populations can cause extensive modification of vegetation structure when their density increases (Owen-Smith 1996; Whyte et al. 2003; Guldemond and van Aarde 2007). Management methods such as culling, translocation, and birth control have not reduced density in some cases (van Aarde et al. 1999; Pimm and van Aarde 2001). Providing more space for elephants is one alternative management strategy, yet fundamental to this strategy is a clear understanding of habitat and landscape use by elephants. Harris et al. (2008) combined remotely sensed data with Global Positioning System (GPS) and traditional ethological observations to assess elephant habitat use across three areas that span the ecological gradient of historical elephant distribution. They explored influences on habitat use across arid savannahs (Etosha National Park in Namibia) and woodlands (Tembe Elephant Park in South Africa and Maputo Elephant Reserve in Mozambique). The researchers focused on three main variables—distance to human settlements, distance to water, and vegetation type. The authors used Landsat 7 ETMþ imagery to create vegetation maps for each location, employing supervised classification and maximum likelihood estimation. Across all sites, they recorded the coordinates of patches with different vegetation and of vegetation transitions to develop signatures for the maps. Elephants do not use all vegetation types, and it can be expedient to focus on presence rather than both presence and absence. Accordingly, the researchers used GPS to record the locations of elephants with the aim of identifying important land cover types for vegetation mapping. The authors mapped water locations in the wet and dry seasons using remotely sensed data and mapped human settlements using GPS, aerial surveys, and regional maps. They tracked elephants with radiotelemetry collars that communicated with the ARGOS satellite system, sending location data for most of the elephants over 24 h, and then remaining quiescent for the next 48 h to extend battery life.

Landscape fragmentation

The country of Vietnam has long been recognized as an important region for biodiversity (Sterling et al. 2006). High-profile discoveries in the 1990s of many species new to science including large ones such as the Saola (Pseudoryx nghetinhensis), an 85 kg basal member of the cattle subfamily Bovinae and the first new genus of large land-dwelling mammal described since the okapi (Okapia johnstoni) in 1901, have focused the attention of national and international conservation organizations on Vietnam and surrounding countries in mainland Southeast Asia (Hurley et al. in prep.). Conservation action for these endemic, endangered species relies on a clear understanding of trends in habitat conversion. To track deforestation rates through time in Vietnam, Meyfroidt and Lambin (2008) combined remotely sensed data with landscape metrics such as number of patches, mean patch size, mean proximity index, and total core area index. They tested their analyses across a variety of land cover studies including those using Advanced Very High Resolution Radiometer (AVHRR), Landsat, SPOT, and MODIS data sources. They found that forest cover decreased nationally from the 1980s to the 1990s and then showed an increase between 1990 and 2000, due to plantation forests as well as natural forest regeneration. However, the effects of this forest transition on fragmentation metrics noted above differed across the country. For instance, in some places, such as central Vietnam where forest cover is relatively large and well connected, reforestation led to a decrease in forest fragmentation and secondary forests recovered rapidly. However in others, such as areas in the north where forest fragmentation dates back centuries and forests have therefore long been isolated, reforestation did not seem to have an impact on continued fragmentation and habitat loss. In this chapter we detail the importance of fragmentation and landscape metrics to ecology and conservation, outlining when and where remotely sensed data can help in these analyses. We then discuss a subset of fragmentation metrics and point to some challenges in processing fragmentation data. We provide examples of composition and connectivity metrics illuminated with examples from the remote sensing literature.

Measuring and monitoring land cover, land use, and vegetation characteristics

In terrestrial biomes, ecologists and conservation biologists commonly need to understand vegetation characteristics such as structure, primary productivity, and spatial distribution and extent. Fortunately, there are a number of airborne and satellite sensors capable of providing data from which you can derive this information. We will begin this chapter with a discussion on mapping land cover and land use. This is followed by text on monitoring changes in land cover and concludes with a section on vegetation characteristics and how we can measure these using remotely sensed data. We provide a detailed example to illustrate the process of creating a land cover map from remotely sensed data to make management decisions for a protected area. This section provides an overview of land cover classification using remotely sensed data. We will describe different options for conducting land cover classification, including types of imagery, methods and algorithms, and classification schemes. Land cover mapping is not as difficult as it may appear, but you will need to make several decisions, choices, and compromises regarding image selection and analysis methods. Although it is beyond the scope of this chapter to provide details for all situations, after reading it you will be able to better assess your own needs and requirements. You will also learn the steps to carry out a land cover classification project while gaining an appreciation for the image classification process. That said, if you lack experience with land cover mapping, it always wise to seek appropriate training and, if possible, collaborate with someone who has land cover mapping experience (Section 2.3). Although the terms “land cover” and “land use” are sometimes used interchangeably they are different in important ways. Simply put, land cover is what covers the surface of the Earth and land use describes how people use the land (or water). Examples of land cover classes are: water, snow, grassland, deciduous forest, or bare soil.

Working with images

There are two very different ways to envision a satellite image: as a photograph taken with a camera, or as a visual representation of spectral intensity data quantifying the light reflecting off of objects on a planet’s surface. In working with satellite images, sometimes the objective is to highlight and accent the information in the image using tools to enhance the way the image looks—the same goal that a professional photographer might have when working in the darkroom with film or using Photoshop to manipulate digital photographs. Another objective could be to manipulate the image using automated processing methods within a remote sensing package that rely on a set of equations that quantify information about reflected light. With either approach the goal is to gain information about conditions observed on the ground. At first glance, the image in Fig. 3.1 bears little resemblance to what most people would recognize as a terrestrial landscape. After all, its predominant colors are orange and bright turquoise. The use of colors in creating a visual image allows great breadth in the types of things one can identify on the ground, but also makes image interpretation an art. Even an inexperienced interpreter can make some sense of the image; more experienced interpreters with knowledge of the color scheme in use are able to determine finer details. For example, in Fig. 3.1 some of the more prominent features are a river (blue line on the left side of the image) a gradient of different vegetation (orange colors throughout the image that go from light to dark), and burn scars (turquoise patches). Fig. 3.2 shows a portion of landscape represented in the satellite image in Fig. 3.1. The red dot in Fig. 3.1 indicates the location where the photograph was taken. This photograph shows what a human observer would see looking south (in this case toward the top of the satellite image) from the point represented by the red dot. The view in the photograph differs from the satellite image in two important ways.

Human interfaces and urban change

For the first time in human history, more people live in urban areas than in rural areas, and the patterns of suburbanization and urban sprawl once characteristic of North America are now present globally (Obaid 2007). As conservation biologists seek to prioritize conservation efforts worldwide, urbanization and agricultural development emerge as two of the most extensive processes that threaten biodiversity. Suburban and rural sprawl are significant drivers of forest fragmentation and biodiversity loss (e.g., Murphy 1988; Radeloff et al. 2005). Data on human impacts is often averaged across political boundaries rather than biogeographic boundaries, making it challenging to use existing data sets on human demography in ecological studies and relate human population change to the changes in populations of other species. Remotely sensed data can make major contributions to mapping human impacts in ecologically relevant ways. For example, Ricketts and Imhoff (2003) assigned conservation priorities (based on species richness and endemism) for the United States and Canada using several different types of remotely sensed data. For mapping urban cover, they used the map of “city lights at night” from the Defense Meteorological Satellite Program (Imhoff et al. 1997) to classify land as urbanized or not urbanized. For mapping agricultural cover, they used the USGS North America Seasonal Land Cover map (Loveland et al. 2000), derived from the Advanced Very High Resolution Radiometer (AVHRR), lumping five categories to create an agricultural land class. For ecological data, they used a compilation of ecoregion boundaries combined with range maps for over 20,000 species in eight taxa (birds, mammals, butterflies, amphibians, reptiles, land snails, tiger beetles, and vascular plants; Ricketts et al. 1999). Analyzing these data, Ricketts and Imhoff (2003) identified a strong correlation between species richness and urbanization. Of the 110 ecoregions studied, 18 ranked in the top third for both urbanization and biodiversity (species richness, endemism, or both); some of the ecoregions identified as priorities were not identified by a previous biodiversity assessment that did not include the remotely sensed mapping of urbanization (Ricketts et al. 1999).

Marine and coastal environments

New remote sensing challenges arise from the addition of the water column to the remote sensing signal. At the same time, new opportunities for use of remotely sensed data are possible in the marine environment. Marine environments can have organisms in such great abundance that they are readily monitored using remote sensing. From measuring ocean productivity, to harmful algal blooms (HABs), to fisheries management, remote sensing is a key component of many efforts to manage and conserve marine ecosystems. For example, the small giant clam, Tridacna maxima, is endangered in some areas of the Pacific, and because of commercial harvest pressure is listed in Appendix II of the Convention on the International Trade of Endangered Species (CITES, meaning they are not yet threatened by extinction but could become so if their trade is not tightly regulated). Andréfouët et al. (2005a) used field observations and remotely sensed data to study the productivity of the clam fishery in tiny (22.2 km2, including a 9.9 km2 lagoon) Fangatau Atoll (Eastern Tuamotu, French Polynesia). The fishery was under pressure due to the large (4 ton per year) export of clams to Tahiti. Remotely sensed data included a mosaic of aerial photographs (1.5 m resolution), a digital photograph taken from the International Space Station (red, green, blue, 5.6 m resolution), and Landsat TM imagery (30 m resolution). The authors classified each image of key lagoon habitats, using maximum likelihood supervised classification, with each image classified independently. They estimated the population size for the entire lagoon by multiplying the mean clam density in each habitat (from field data) by the total area of each habitat (in the maps made from the remotely sensed data). Amazingly, an estimated 23.65 ± 5.33 million clams (mean ± 95 percent confidence interval) inhabited the 4.05 km2 area of suitable habitat in the lagoon. The high spatial resolution data (1.5 m aerial and 5.6 m astronaut photography data) both gave equivalent estimates of the biomass with good estimates of accuracy, but the Landsat 30 m data overestimated the population.

Global conservation

On 24 December 1968, as they watched the half-illuminated earthrise over the surface of the moon, the crew of the Apollo 8 lunar mission captured an image that changed humankind’s view of our planet and our place on it. The earthrise image and other iconic global images like the “blue marble” photo taken by the crew of Apollo 17 in 1972 gave us, for the first time, a global view of our fragile home within the vastness of space. These early global images helped promote environmental awareness around the world and were instrumental in the development of the field of remote sensing (Lowman 1999). However, it would take some time for the research community to compile and use global-scale imagery from space in the ecological sciences. Improvements in passive and active remote sensing systems placed in orbit by national governments and the growing commercial satellite sector have given us an “end-to-end” remote sensing capability that allows us to make measurements of important environmental phenomena from very local to global spatial scales (of course, airborne remote sensing systems have long enhanced our ability to capture information at local scales). Data depicting the social and economic drivers of biodiversity loss are also available globally from a variety of sources. These different data sets can now be brought together with powerful, affordable, spatially referenced computing technologies, e.g., GIS and GPS, which were unimaginable when the Apollo missions sent back their images. The entire Apollo spacecraft’s computing power was less than that of today’s mobile phone. Taken together, these advances have made it possible to grapple with the complexities and scale of addressing conservation challenges at the global level. This chapter elaborates the role of remote sensing as one among several catalysts driving the development of new approaches to ecology and conservation biology at the global level. In the early 1980s, NASA initiated its Global Habitability program (NASA 1983; Waldrop 1986; Running et al. 2004). This program sought to answer the big question of how the biosphere partitions its energy and mass.

Linking remote sensing with modeling

Conservation biologists and natural resource managers often require detailed, accurate information on natural resources or biodiversity elements such as species, landscapes, and ecosystems. Their patterns of occurrence and their responses to environmental disturbance or change are dynamic over space and time and may be mediated by complex ecological processes. In most cases, our ability to directly measure or comprehensively map biodiversity elements is limited by human or financial resources, and logistical challenges such as difficulties in accessing terrain or short field seasons. In other situations, we might want to make quantitative inferences about, say, the kinds of environments that are most suitable for the persistence of an endangered species, or the influence of landscape modification on its highest-quality habitat. In these cases, developing models that explain and predict the patterns of biodiversity elements can help provide guidance at scales and resolutions that are not available through direct measurement. For example, Goetz et al. (2007) employed lidar data to predict the bird species richness across a 5,315 ha temperate forest reserve, the Patuxent National Wildlife Refuge (PWNR) in the eastern United States. In this study, Goetz et al. derived and mapped several measures of forest canopy structure, including canopy height, and three descriptors of the vertical distribution of canopy elements. In addition to lidar, they also used optical remotely sensed data from two dates of Landsat ETM+ to derive NDVI during the growing season and the difference between the NDVI of leaf-on and leaf-off conditions (growing season versus winter). Testing three different quantitative statistical models (stepwise multiple linear regression, generalized additive models, and regression trees) to predict bird species richness, the authors used field survey data on the birds of the PWNR that were collected at a series of fixed points across the reserve as the training data for the response variable (bird species richness). To calibrate the model, they combined the habitat descriptors with the survey data, usually reserving 25 percent of the survey data to validate each model’s results.

Protected area design and monitoring

Researchers interested in remote locations have developed monitoring schemes, sometimes called “Watchful Eye” monitoring, that use a time series of remotely sensed images to assess changes over time to a protected area or habitat. For instance, the European Space Agency (ESA) and UNESCO have set up repeat analyses of satellite imagery for World Heritage sites. The first area for which they developed this technique was the habitat of the critically endangered mountain gorilla (Gorilla berengei berengei) in the Virunga Mountains in Central Africa, including the Bwindi and Mgahinga National Parks in Uganda, the Virunga and Kahuzi-Biega National Parks in the Democratic Republic of Congo, and the trans-boundary Volcanoes Conservation Area. The project developed detailed maps of these inaccessible zones so that protected area managers can monitor the gorilla habitat. Previously, available maps were old and inaccurate (at times handmade), did not completely cover the range of the gorillas, and did not cross national boundaries. Because there was no systematic information from the ground regarding changes over time, researchers also used remotely sensed data to complete change detection analyses over the past two decades. Using both optical (Landsat series) and radar (ENVISAT ASAR) satellite data, researchers were able to quantify rates of deforestation between 1990 and 2003 and relate these rates to human migration rates into the area resulting from regional political instability. Researchers constructed the first digital base maps of the areas, digital elevation models (DEMs), and updated vegetation and land use maps. They faced significant problems in both field and laboratory activities, including lack of existing ground data, dense vegetation cover, and fairly continuous cloud cover. They therefore used a combination of ESA ENVISAT ASAR as well as Landsat and ESA Medium Resolution Imaging Spectrometer (MERIS) optical data. The radar images allowed them to quantify elevation and distances between trees and homes. Landsat and MERIS data helped identify forest cover types, with Landsat providing finer-scale images at less frequent intervals and MERIS serving lower-resolution images more frequently.

Terrain and soils

Building on the foundations of working with images and measuring land cover and vegetation in the previous two chapters, we now add the discussion of elevation and geology. Terrain attributes (such as elevation, slope, and aspect) and soil characteristics affect the distribution of most taxa and are therefore critical for effective biodiversity monitoring and conservation. Remote sensing is the primary tool for collecting terrain information from local to global scales. This chapter will provide an overview of different types of terrain data that you can collect using remote sensing methods as well as how you can visualize and analyze these data. We will also highlight applications using terrain data to illustrate how they and their derived products can aid the conservation biologist or ecologist. In addition to landforms, this section will look at how remote sensing technology can provide information about geology and soils. Box 5.1 provides an example of how elevation data and satellite imagery helped in selecting field sites. The fundamental measurement used to create terrain data is elevation. In this section we discuss elevation measurements in detail. First, we define elevation and then describe common data formats. Next we present an overview of you can acquire elevation values and then we will present examples of how elevation data are useful in biodiversity conservation and how you can use these data to improve the geometric and radiometric qualities of remote sensing imagery. This section concludes with a discussion of some of the methods used to analyze and visualize terrain data. Before discussing the use of elevation data it is useful to define what is meant by “elevation” and other related terms. A general definition for elevation is that it is the vertical distance measured from a point to a reference surface. This may seem pretty straightforward, but accurately defining the location of a point or a reference surface in the vertical dimension can be quite difficult. When measuring an elevation you need to define a reference surface. This reference surface is called a vertical datum. A common reference (or datum) for elevation measurements is sea level. Many maps, for instance, label elevation as meters or feet above sea level.