Integrating remote sensing and GIS with ecology for process-based understanding and prediction at landscape to regional scales
Landscapes are mosaics of different patches and gradients, varying in size, shape, composition and spatial configuration at multiple scales. This spatio-temporal heterogeneity has often been ignored in ecology, which traditionally focused on similarities rather than differences in ecological systems. We explore the processes that create, maintain and modify landscape heterogeneity, and assess the consequences of heterogeneity for ecological functioning and biodiversity conservation in the context of global change.
We use remote sensing tools to explicitly quantify heterogeneity within landscapes and explore spatial and temporal dynamics. We make extensive use of LiDAR (laser altimetry) to map landscape and vegetation structures in three-dimensions (3-D) across a broad range of ecosystems including savannas, tropical forests, and temperate forests. We aim to improve understanding and modeling of ecosystem processes across scales and inform biodiversity conservation and land management.
Our primary study sites are located in South Africa, Australia, Brazil and Germany.
Ecological processes are seldom uniform or random in space, as landscapes contain spatial structures that mediate how energy, materials and organisms move through them. Underlying soil type and hillslope morphology are two primary controls that influence biogeochemical processes, but spatial heterogeneity of these factors is poorly accounted for in regional and global models. We aim to improve understanding of how ecological processes vary across landscapes and facilitate integration with predictive modeling efforts.
We are studying a range of hillslope catenas (topographically linked sequences of soil, water and vegetation) on different geologies across a rainfall gradient in South African savannas to better understand how climate and substrate influence biogeochemical processes (such as soil carbon storage and flux) at hillslope-scales. We are currently expanding this research to sites in Australia and South America to gain a global perspective on hillslope-scale processes in savanna systems.
The vegetation present at a given point in a landscape is a function not only of climate and environmental resources, but also of various disturbance agents acting across that landscape. In savannas, vegetation structure and biomass is often much lower than what we would expect from climate potential alone, as fire, herbivores, wind and human land-use disturbance reduce standing biomass and are major determinants of vegetation structure and dynamics. We aim to understand how disturbance effects vary spatially across landscapes, and how the relative importance of different disturbances varies with spatial and temporal context.
Fire effects on carbon storage in savannas represent significant uncertainty in global carbon budgets, driving disparities between potential and realized biomass. We are using a network of long-term fire experiments in the savannas of southern Africa, northern Australia, and South America to improve understanding of how fire influences vegetation structure and carbon storage. This research is closely connected with land managers in these systems, who are interested in the biodiversity and carbon management implications of different fire policies.
Vegetation structural diversity is a fundamental component of biodiversity. Diverse plant structures create broad arrays of habitat for other organisms to utilize, and increase the range of ecological functions that vegetation provides. In order to conserve biodiversity and ecological functioning under global change, we need to better understand how variation in climate and land-use influence ecosystem structure and dynamics. Growing appreciation of the importance of spatial scale, heterogeneity and context in ecological research has seen a rapid increase in the application of remote sensing in ecology to provide spatially continuous representations of ecosystems. Airborne LiDAR (light detection and ranging) has emerged as a valuable tool for the structural characterization of ecosystems in three-dimensions (3-D), at fine resolutions, and over large spatial extents. However, despite large advances in mapping canopy structure in high spatial resolution, we have seen limited direct application of LiDAR in biodiversity research. This disparity partly stems from the collection and analysis of LiDAR data remaining largely in the remote sensing, engineering and computer science fields, with poor integration into ecological and biodiversity science.
We aim to develop better insight into how biodiversity and ecological functioning might change under future climate and land-management conditions. I am using high resolution airborne LiDAR to explore the variability in tree architecture across gradients of land-use and climate in European temperate systems. The architecture of an adult tree is a reflection of current and historic growing conditions, as it integrates the environmental and disturbance factors that have shaped its structure from seedling to maturity. We can use this information embedded within the 3-D canopy structure of trees, at multiple spatial scales, to better understand how different drivers influence carbon storage and structural diversity.
“Ecosystem engineers” are organisms that directly or indirectly modulate the availability of resources to other species, by causing physical state changes in the abiotic or biotic environment. I am studying the ecosystem impacts of two such engineers in African savannas – termites and elephants.
Termite mounds represent nutrient and biodiversity hotspots within the broader landscape matrix. Termites build their mounds from clays and are a major source of particle and nutrient redistribution in savannas. I am using LiDAR data collected by the Carnegie Airborne Observatory (CAO, http:// cao.ciw.edu), in collaboration with Greg Asner, to map the spatial location of termite mounds and gain better understanding of the spatial distribution and density of mounds on different soil types and under different rainfall regimes. We are also conducting a range of field studies to assess the scale of termite mound influence as a forage resource for other organisms.
At the larger end of the organism spectrum, elephants modify the physical environment by breaking branches and pushing over trees. Large trees form islands of biogeochemical activity within the landscape matrix, but are disappearing in many savannas through the interaction of increasing elephant densities and fire. We are using satellite imagery and airborne LiDAR (from the CAO) to understand the rate and spatial distribution of elephant impacts on large trees across different substrate, hillslope and rainfall settings. This research is conducted in close collaboration with Carnegie and South African National Parks (SANParks) scientists to understand the ecological consequences of tree loss, and provide crucial information for the setting and evaluation of biodiversity conservation objectives.