Global Effects of Human and Terrestrial Interactions
Project Objectives:
The interaction of human systems and terrestrial systems at a global scale, as moderated by climate and atmospheric composition, is complex. Energy use, agricultural practices, and demographics, are entwined with the desire of people to improve their well-being and to manage economic growth and its effects through well-meaning but sometimes misdirected policies. In particular, this means economic growth as rapid as can be sustained in developing countries. Left uncontrolled, the environmental effects of the increasing scale of these activities are potentially enormous and global in scale. The state-of-the-art approach to modeling these interactions is to either (1) use complex but separate models of human activity, of the atmosphere and ocean, of the terrestrial systems, and sector-by-sector models of economic and human effects or (2) simplify the system dramatically to produce a cost-benefit policy calculator and in doing say eliminate the structural details where non-linearities, multiple interactions, and complexities come into play.
Our goal is to evaluate fundamental interactions between human and terrestrial systems, driven by forces that are global in nature and moderated by climate and atmospheric composition. Our focus is on food production, an obviously a critical human need in developing countries. In preliminary work we have examined interactions among climate, tropospheric ozone, ecosystems, and climate policy costs: ozone damage increased the cost of a carbon stabilization policy by a substantial 8 to 14%. The results, if they stand up to more detailed scrutiny, mean that the current state-of-the-art approaches are inadequate to address these interactions and thus may lead to policy prescriptions that are inefficient and ineffective at meeting their goals. In this preliminary work, we investigated only the economic feedbacks on carbon mitigation costs, but the economic effects of ozone damage and climate change extend to the value of lost crop and forest production, and the goal is to develop the dynamic linkages needed to evaluate these costs and potential feedbacks.
Our approach is to dynamically couple a biophysical model of global terrestrial ecosystems with a global model of the world economy, within the context of an integrated global system model. The economic model drives agriculture, forestry, and land use. The terrestrial ecosystem model simulates carbon, nitrogen and water dynamics of vegetation in both natural and managed systems to capture the effects of multiple stressors including changing climate, atmospheric composition (CO2, tropospheric ozone) and atmospheric deposition of nitrogen on ecosystem processes. A unique aspect of this work will be the ability to model short-term effects on, for example, annual NPP and the yield of agriculturally important crops as well as longer-term phenomena such as changes in soils. These coupled models will further be coupled with an atmosphere-ocean model. Thus we will be able to investigate: 1) the effects of more realistic and complex interactions of air pollution or climate policy, separately or in combination, on vegetation including crops, 2) the resultant impacts on the economy and adaptive response to these changes including international market repercussions; and 3) the ultimate feedback on the cost of the policies, economic growth, and/or the emissions of the pollutants themselves. The effects and changes, and the complex interactions, are certainly not unidirectional. Some effects, such as N deposition and CO2 fertilization, are beneficial to vegetation while others such as tropospheric ozone are mostly damaging, and still others, such as climate change, can be either beneficial or detrimental. Similarly pollution and climate policies interact because of technological interdependence, and the cost and economic implications of these policies interact with existing economic distortions in sometimes surprising ways. This work will closely parallel and benefit from ongoing work in the MIT Joint Program to couple the terrestrial model with the atmosphere-ocean model, and to investigate the feedbacks and interactions of air pollution, climate change, and human health. This currently existing model structure together with new work to dynamically couple economic and terrestrial systems will be a major leap forward for the human dimensions of global change, providing the capability to model human systems as an integral part of the global earth system.
Project Components
There are three model development efforts required to dynamically couple the terrestrial and economic models: (1) Disaggregation of the economic model and development of satellite accounts to link economic results to a spatially explicit physical land data base to drive land use change in the terrestrial model, (2) Development and testing of crop model components of the terrestrial model that capture the effects of changing climate, CO2, ozone, soils and the nitrogen cycle to simulate changes in potential productivity of crops, (3) Modeling of the economic adjustments that can moderate changes in potential productivity, estimating functions and parameters that can be represented in the global economic model. Once this dynamic coupling is complete, investigation of the importance of interactions will benefit greatly from previous developments of the MIT Integrated Global Systems Model. In particular, the MIT IGSM already models the economic drivers of multiple greenhouse gases and pollutants, atmospheric chemistry interactions, and atmosphere-ocean interactions, including an existing NSF biocomplexity grant to elaborate the terrestrial and climate feedbacks. Thus, the investigations of future scenarios can be assured of consistency between scenarios of economic growth and emissions as they drive atmospheric composition and climate and the economic modeling scenarios that assess the economic implications of these changes in agriculture and forests. The pre-existing components of the MIT IGSM will make it possible to investigate how these agricultural effects of changing climate and atmospheric composition feedback on greenhouse gas and other emissions, atmospheric composition and climate. The mechanisms by which those feedbacks will occur include changes in overall economic activity and thus demand for energy and agricultural products, but likely more importantly the changed demand for food and the spatial pattern of production driven by the spatially varying changes in crop productivity. Further, economic research has also shown that the economic cost of climate policy in terms of lost welfare and consumption depends on how such policies interact with pre-existing distortions in the economy. Existing tax and subsidy distortions are represented in the EPPA model, and so it will possible to consider these rich set of interactions as well. The energy and agriculture sectors are highly distorted by subsidies and taxes, and thus this should provide a rich avenue for investigation.
A fourth component of the project is simulation and investigation of uncertainties. A fundamental question in climate change impact analysis is the predictability of climate at the spatial and temporal scales required. This link between coarsely resolved climate model output and impact analysis is typically described as a problem of downscaling. As often practiced, however, downscaling is a somewhat mechanical process of adjusting coarse grid scale climate data to a finer grid or site. Such downscaling introduces additional uncertainties that are not well characterized, and often overlooked. At the global level progress has been made in using optimal fingerprint methods to characterize uncertainty. We will investigate of the applicability of such techniques for regional climate forecasts. While there are many problems in trying to apply the methods used for global predictions to local and regional prediction the possibility exists, we believe, that fine scale changes in precipitation, for example, or other climate variables could easily overwhelm an average change. We would thus characterize this component of the work as high risk, but potentially high reward. We expect the result of this investigation to at least provide one or more numerical downscaling approaches, well-informed by the expertise of climate scientists, that are in any case needed for our simulations. By broadening the investigation, the ultimate goal is to actually characterize uncertainty or understand the limits of predictability. While the chance of reaching this goal appears small within the time frame of the grant, if we can develop avenues for further investigation toward that goal it will be a big step forward. To be meaningful, such investigation is best conducted in the context of an effort to estimate a specific set of impacts where it is critical to resolve climate change at fine spatial and temporal levels, as is the case of agriculture. In addition, we will conduct more conventional uncertainty and sensitivity analyses using multiple existing climate scenarios and analysis of uncertainty in future emissions.