Land Use and Land Cover Change (LUCC)
--David Skole (dave@igapso.sr.unh.edu),
Complex Systems Research Center, Institute for the Study of
Earth, Oceans, and Space; University of New Hampshire, Morse
Hall, Durham, New Hampshire 03824.
LUCC in the Context of Global Change Research
Over the coming decades, the global effects of land use and cover change (LUCC) may be as significant, or more so, than those associated with climate change. Moreover, land use and cover change are known and undisputed aspects of global change; it is an important human-caused global environmental change that is with us now. In addition to their central role in affecting climate, many facets of human health and welfare are directly connected with LUCC, including biological diversity, food production, and the origin and spread of infectious disease. Yet, we don't know enough about these important human-caused agents of global change. It is a testament to this paucity of knowledge that an accurate global map of agricultural activity does not now exist, nor do we have good measurements of agricultural expansion and the concomitant loss of natural ecosystems, particularly forests.
Classically, the empiricism of observation and measurement is the core of good science. An improved understanding of LUCC must begin with documenting the rate and extent of major changes worldwide. However, it is equally important to move the science from the assessment of pattern to the analysis of process. Hence, furthering our understanding of the complexities and dynamics of LUCC will require a comprehensive, interdisciplinary examination of its underlying causes. LUCC is sufficiently complex that research must include a wide range of scientific and scholarly disciplines, including demography, economics, political science, systems ecology, and other related fields.
The importance of an interdisciplinary perspective was recognized early in the development of the LUCC Core Project, and is manifested in its joint sponsorship by the IGBP and the IHDP. From inception, the planning and implementation of the project has actively engaged both the natural and social science communities, and this will continue to be an important modus operandi in the future.
However, it is not prudent to engage these two communities separately. Rather, we should entrain the intellectual contribution of the interdisciplinary LUCC community itself -- those scientists and scholars who are actively involved in the study of LUCC in its global change context. In this way, we will not only engage research teams comprising of members from both of C. P. Snow's "two cultures", but we will also engage those individuals who bridge the two cultures in their own research: the demographer who uses remote sensing to help elucidate the spatial orientation of settlement over time, or the systems ecologist who uses socio-demographic models to help elucidate the process of habitat fragmentation (as two simple examples).
LUCC as an Agent of Change
The contemporary state of the world's land cover is a constantly changing mosaic of cover types determined by both the physical environment and human activities. These changes in land cover can have profound global consequences. Here I will illustrate the various dimensions of LUCC as an agent of global change through three examples: the global carbon cycle, biological diversity, and infectious disease.
The Global Carbon Cycle
Understanding the global carbon cycle is central to our understanding of global change. The increase in atmospheric carbon dioxide is attributed to two anthropogenic sources: fossil fuel combustion and land cover conversion. Although the current net flux of carbon is dominated by the fossil fuel source term, biogenic net sources contribute approximately 25 - 30% of the total. However, over the last two hundred years the contribution from land cover conversion has been approximately equal to that of fossil fuel combustion. Reconstruction of the last two hundred years of land cover change will be an important part of carbon cycle research.
There are two ways to provide this long-term historical view. The first is to document land cover changes directly using historical reconstruction of changes in land use, particularly agricultural expansion. This would require the interdisciplinary efforts of historians and biogeographers. An example of this kind of work is the reconstruction of land cover change in South and Southeast Asia by John Richards and his colleagues and worldwide by Richard Houghton and his colleagues. Because of the nature of the data, reconstructions are usually spatially and temporally aggregated. Commonly, time steps are a decade or more and data are compiled at the country or province level.
The second approach is to use models to "backcast" the history of land cover change from current distributions of land cover. The current distribution could be obtained from satellite remote sensing data, such as the global 1 km AVHRR land cover dataset being compiled by the IGBP-DIS. Integrated models, which utilize socio-demographic and economic information as forcing functions, could be used in conjunction with rule-based models and synoptic historical data at various intervals to derive spatially and temporally disaggregated analyses. In essence, such an approach which integrates actual observations, either for the current time period using satellite data or back in time using historical documents, with model-based estimates would be similar to the assimilation techniques, which climate and meteorological models often utilize.
While extremely valuable as a way to build the long-term datasets on land cover transformation, such an approach falls short of providing a robust and dynamic understanding of the interrelationship between land use and land cover as it is occurring today. Dynamic Vegetation Models (DVMs), in which land use change and land cover change are interrelated in terms of processes and feedbacks offer an advanced approach to coupling the human driving variables with ecosystem response functions, hydrological dynamics, atmospheric conditions, and edaphic (fire related) factors. The development of LUCC-driven DVMs will require a truly interdisciplinary analysis and integrative modeling involving most of the IGBP Programme Elements.
Such models could, for instance, account for the role of transient states of secondary succession following disturbance. New evidence from satellite monitoring in tropical forests suggests that a large fraction of disturbed areas are secondary growth. In the Amazon, approximately one-third of the deforested area is in secondary succession following abandonment of agricultural fields. Detailed multi-temporal analyses of satellite data show that deforestation is a highly dynamic process of clearing, abandonment and re-clearing, and the rates at which land is cleared or abandoned are related to the land use and management system the forest farmers employ.
In one study I conducted with Brazilian colleagues from 1986 to 1993, approximately 35% of disturbed sites were in active agriculture every year through 1993; these are sites in which the land use system is maintaining them as pastures or crops. Another 9% of the sites persisted every year as secondary growth, representing sites where long-term abandonment was occurring due to out-migration and severe losses of site fertility. But approximately 56% of the sites rotated between secondary growth and active agriculture throughout the period of observation, representing sites where the use and rotation of secondary growth is an explicit land use strategy.
In some cases, deforestation and secondary succession exist in tandem as a tightly coupled system in which secondary growth is continually recycled back into farmland. In other cases active land management maintains the land in agriculture, or the lack of active land management or population displacement results in long-term succession. Thus, in the Amazon and elsewhere, land use change influences land cover change and is integrated with dynamics of ecosystem structure, function, and response.
Within a focused research program on LUCC it will be important to frame such questions as: what human land use and land-management strategies are employed in different situations around the world, and how do they control, or interact with, the dynamics of ecosystem response to disturbance? Because deforestation and abandonment have opposite effects on atmospheric carbon dioxide, e.g., uptake vs. release, we can restate the aforementioned general question in terms specific to the carbon cycle: over time, what land use strategies determine the abundance and spatial distribution of secondary growth, how do they determine the balance between clearing and regrowth rates, and how are these land use strategies in turn influenced by ecological conditions?
Finally, it is worth noting that the current global carbon budget is not balanced. There are good reasons to suspect that there is an unaccounted "missing sink" in undisturbed forests of the world, and hence some importance is placed on emphasizing research on global ecosystem metabolism. However, a net uptake of approximately 1.5 x 1015 gC yr-1 spread over large undisturbed forested ecosystems would be difficult to measure or detect in the field at the hectare scale. Thus, models which predict such sinks must infer the magnitude of the sink as a residual calculation from estimates of fluxes associated with land cover change.
Again turning to the Amazon as an example, the rate of deforestation appears to have increased from the early 1970s to the late 1980s, reaching a peak and then declining quite dramatically through the early 1990s. At the same time, but probably lagged by several years, the rate of abandonment to secondary growth has likely increased as well, but just out of phase. Such asynchronicities in two related, but non-linear and phased pulses of land cover change have interesting consequences for atmospheric carbon dioxide and global carbon budgets.
Is it possible that while the rate of deforestation is declining, the rate of reforestation is increasing, causing the net flux of carbon to be unusually low, at least temporarily? Over the long term, what socio-economic factors are controlling the relative balance between deforestation and abandonment? How will they persist into the next century? Are these socio-economic factors influenced by interannual and long-term climate changes?
Biological Diversity
Typically, species/area curves are used to estimate the relative loss in species due to habitat loss. A simple approach would be to use rates of deforestation to estimate the change in forest habitat area. New research, however, is elucidating a fascinating new role for land use and cover change. This work points to three factors: (a) spatial geometry of ecosystem disturbance through habitat fragmentation and its relationship to land use practices, (b) the time-varying "matrix" of areas in agriculture or selective logging and secondary succession over time, and (c) historical changes in land use within "natural" ecosystems.
It is not enough to know only aggregate rates of forest loss since it is the spatial geometry of deforestation that is critically important to understanding forest fragmentation. Deforestation affects biological diversity in three ways: the destruction of habitat, isolation of fragments of formerly contiguous habitat, and edge effects within a boundary zone between forest and deforested areas. When one considers the spatial geometry of land cover change and calculates this total effect, the impact is much larger than if one used deforestation rates alone. In my own research with C. J. Tucker we found the total affected habitat to be twice as large as the deforested area.
It is not possible to predict the spatial geometry, and hence the total effect on habitat, from aggregate information on deforested areas alone. Also, within a landscape of various land uses, the dynamics of disturbance and succession create a changing matrix of vegetation types and habitats which are more complex than simply forest vs. non-forest. The spatial relationship between primary forest, deforested land, and secondary forest imposed by different land uses creates a specific spatial topology which cannot be determined by aggregate figures alone. Detailed measurements need to be made, considering, for example, regions where disturbance patterns are small and dispersed vs. large and clustered.
Information on land use can provide insights into the cause and characteristics of forest fragmentation. The size and shape of clearings are often related to the land use system being employed, whether they are large commercial farms or small-holder sites. The spatial orientation is influenced by transportation corridors, population nodes, historical foci of settlement, and local environmental conditions such as slope, soils, and rivers. Moreover, as mentioned above in the discussion of the carbon cycle, land use strategies are coupled closely to the dynamics of secondary growth. Some fundamental LUCC questions begin to emerge from this view: within a given landscape how do land use, tenure, and management influence the spatial topology of primary and secondary land covers and biological diversity?
Peter Vitousek and colleagues have estimated that as much as 40% of global net primary productivity (NPP) has already been utilized by human activities. A large component of this appropriation of NPP occurs when human land uses directly disturb natural ecosystems, such as clearing forests for agriculture or logging. The work of anthropologists, ethnobotanists, and ecologists suggests the biological diversity of large areas of so-called "natural" ecosystems can also be directly affected by human use. In what he refers to as the Empty Forest, Kent Redford and others have shown an important influence of the customary use of natural ecosystems, where humans exploit the flora and fauna of intact forests to a significant degree, particularly those in close proximity to disturbed areas.
If our concept of global change rests solely on the issue of climate change, we would likely ignore intact forests, since their influence on carbon dioxide, sensible and latent heat flux, and the atmosphere are relatively unchanged compared to logged areas and croplands. But as Redford states, "We must not let a forest full of trees fool us into believing all is well." This leads to interesting land use and human dimensions questions to be asked if we want to look beyond climate change: How do changing patterns of consumption and human population density affect biological diversity, even in intact ecosystems? How does the spatial orientation of land use affect intact ecosystems?
Infectious Disease
There are very good data relating the incidence of certain human diseases to changes in land cover. Malaria rates (per thousand individuals) increased fivefold in the Amazon between 1975 and 1990 as deforestation rates increased dramatically over the same period. Regions of most rapid deforestation had the highest infection rates. The epidemiology of disease is complex, multivariate, and not linearly related to disturbance rates, but the link between changes in cover/habitat and disease is becoming an important area of research.
The case of Oropouche disease reported by Robert Shope and his colleagues in the Amazon in the 1960s is indicative. The construction of the Belem-to-Brasilia highway resulted in an outbreak of a flu-like epidemic, which was later attributed to a virus from a biting midge. The midge population grew explosively when settlers cleared the land, and certain crop harvesting practices associated with the land use in the region provided an ideal breeding ground.
The work of Paul Epstein at the Harvard School of Public Health shows the relationship with deforestation in Honduras, where cases of malaria rose from 20,000 to 90,000 from 1987 to 1993. But he also notes that in addition to ecological effects from road building and deforestation, it is the interaction between land cover change and climate that is important.
Along with an emphasis on these kinds of established diseases, there is increasing speculation on the relationship between land cover change and the origin of new disease. There is considerable work to be done in this area, but like no other example the case of infectious disease illustrates the direct link between global change, LUCC, and human health.
LUCC as an IGBP Program Element
The global effects of land use and cover change is an emerging and important area of research. Over the past four years an international community of scientists has been making the case for its inclusion into the IGBP framework, first under the auspices of an ad hoc working group (IGBP Report #24/HDP Report #5), then under the auspices of a formal core project planning committee (IGBP Report #35/HDP Report #7).
There have been a number of special conferences and symposia which have also helped focus the discussion, including the 1991 Global Institute in Snowmass, Colorado, (published by Cambridge University Press, 1994), the 1992 Ecological Society of America Symposium on Global Impact of Land Cover Change (published as a special issue of BioScience, May 1994), and the LUCC Open Science Meeting in Amsterdam, The Netherlands in 1996.
The development of a Core Project on LUCC comes at an important moment in the evolution of the IGBP. Other Core Projects, such as GCTE, BAHC, LOICZ, and IGAC urgently need to incorporate LUCC dynamics in their research. The global synthesis and integrated models provided by GAIM will now need to include the transient states caused by LUCC. The 1 km AVHRR/Global Land Cover dataset being developed by IGBP-DIS will soon be ready for widespread distribution.
At this stage of IGBP development it is becoming increasingly evident that the research community which the LUCC Core Project joins will be pressured in two important ways: (i) to produce timely scientific results and policy-relevant information -- in my own view through focused research campaigns or projects, such as the Large Scale Amazon Basin Experiment (LBA) or the IGBP Transects, and (ii) to strengthen the inter-program-element nature of our research, linking across rather than within the Core Projects and Framework Activities.
LUCC should be a critical nexus for this kind of IGBP research. At the same time, LUCC should provide a bridge between the human dimensions and the geophysical/biophysical dimensions of global change. Moreover, LUCC can provide direct links to policy, as it relates to climate, biodiversity, agriculture, and human health.
[Table of Contents]
[Previous]
[Next]
