Terrestrial ecosystems are important global carbon sinks, and their functioning modulates the variability of the carbon exchange between the land surface and the atmosphere at annual to climate time scales. Vegetation cover also regulates the physical properties of land surface (e.g. albedo and roughness) and the surface energy partitioning of net radiation between sensible and latent heat, affecting water and energy exchange with the atmosphere. Anthropogenic Land-Use/Land-Cover Change (LULCC) modifies these land properties and the carbon cycle. The terrestrial response to this forcing can positively or negatively feedback to the climate system, and thus magnify or reduce the initial perturbation [1
Model Intercomparison studies, such as the Coupled Climate Carbon Cycle Model Intercomparison Project (C4MIP) and the Land-Use and Climate, Identification of Robust Impacts (LUCID) [5
]) systematically evaluate biogeochemical and Biogeophysical feedbacks. Within the C4MIP framework, Friedlingstein [8
] projected a global net positive carbon-climate feedback on surface temperature and atmospheric carbon dioxide (CO2
)-levels; but large uncertainty persists due to the complex processes involved in this feedback [9
]. An important control variable is the effect of increases in temperature on either ecosystem productivity or respiration, and the relative sensitivity of these processes to ambient temperature increases governs the sign of the terrestrial branch of the carbon-climate feedback.
Biogeophysical feedbacks between LULCC and climate have a generally slightly damping effect on the global temperature increase due to dominant increases in albedo in areas where forest is replaced by low vegetation, and snow shading plays a strong role [12
]. However, these albedo changes and their associated feedback mechanisms have a strong spatial structure, leading to possibly large feedback responses at the regional scale [13
]. But again models assessing regional Biogeophysical impacts of anthropogenic LULCC show large divergence between them [5
The uncertainty associated with feedbacks between LULCC and the climate system has triggered considerable research, both at the global and at the regional-scale. While the global mean Biogeophysical response to LULCC is relatively small, it is potentially an important driver of climate change in regions with intensive LULCC [6
]. In some regions (particularly in mid- and high- latitudes), the LULCC-induced cooling effect due to albedo changes is of similar magnitude but of opposite sign compared to the warming induced by the increasing CO2
. In other regions (i.e. the tropics) LULCC can amplify the CO2
induced warming by promoting sensible heat release and reduce evaporative cooling [14
The comprehensive modeling studies supporting the periodic assessments of the Intergovernmental Panel on Climate Change (IPCC) demonstrate that in various regions, future global warming is associated with increasing temperature variability and more frequent extreme events, such as heat waves and droughts [15
]. Biogeophysical feedbacks, especially those related to soil moisture, play an important role in the duration and frequency of these events. Widespread vegetation responses to weather extremes can systematically modulate the carbon uptake by terrestrial ecosystems [16
] that can be noticeable in the global carbon cycle. The short-term vegetation response to adverse extreme climatological conditions such as heat waves and droughts, and its interaction with the governing climate system, are thus worthwhile to explore.
The wide variety of processes, responses and feedbacks involved with land use change and the range of spatial and temporal scales at which these operate complicate systematic exploration of the relative importance of processes and effects. Processes can counteract, reinforce, or conditionally affect each other at different spatial and temporal scales. A careful examination of the net result of the balance of processes requires a modeling framework in which they are represented realistically, and where the mutual sensitivities are well imposed. Analysis of (multi-)model experiments in which these feedbacks are addressed such as [5
] does not always lead to firm conclusions on the overall sign of the responses and feedbacks. In order to increase our understanding of these complex interactions even more system components in these modeling systems need to be considered, or additional detail to the processes already implemented needs to be added, which introduces a new set of dimensions and degrees of freedom to analyze. This process of deepening our understanding of the complex climate-vegetation system is aided by reflecting on a limited perspective of the whole system; a conceptual, simplified picture of it. Such a “conceptual framework” underlies many studies dealing with complex physical climate systems e.g. [16
], and helps to identify the major feedbacks that are worth exploring further in advanced coupled modeling studies.
In this paper we discuss such a conceptual framework that depicts the major interactions between LULCC (i.e., deforestation and irrigation), vegetation and a changing climate at the regional scale. The framework by design simplifies existing relationships, and is limited in scope by not considering all external and internal processes and feedbacks that potentially affect the functioning of the regional vegetation-climate system. But this simplification serves the purpose of careful experimental design, and in teaching new experts in the field.
We shortly review the main feedbacks that play a role at the regional scale and introduce the conceptual framework in Terrestrial ecosystems – climate feedbacks in a conceptual framework. “Regional” is used here to denote the spatial scale at which systematic interactions between vegetation and climate conditions can be expected, and is loosely specified to be in the order of 250,000 km2
(500 × 500 km) or more. We illustrate this framework for three regions where strong feedbacks take place in which LULCC plays a role (Applying the conceptual framework at the regional scale). Quantitative implications of the feedbacks based on literature are provided. The selection of these regions does not intend to be complete or formally justified; it rather serves the purpose of demonstrating the conceptual framework. We will discuss the necessary experimental model design needed to explore these regional scale feedbacks (Implications for experimental design), which entered a recent review of possible modeling strategies for feedback exploration by van Vuuren [19
]. Finally, conclusions are given (Summary and conclusions).