Climatic changes as possible driver of vegetation browning in tropical forests

Alexander J. Winkler
Max-Planck-Institute for Meteorology (MPI-M), International Max-Planck Research School on Earth System Modeling, Hamburg, Germany

Ecosystems and climate interact through a complex web of processes involving a multitude of variables. Many of these variables can be measured with satellite sensors and processed to comprehensive datasets in space and time. In the Earth System Data Lab (ESDL) an extensive collection of these datasets are now only a mouse click away. Thus, this cloud-based computing interface provides the ideal environment to study the complex coupling of the climate-biosphere system.

Ecosystems are a hodgepodge of species that are highly adapted to their environment, interconnected and interdependent in a complex network. The cornerstones of these systems are plants that are able to store the incoming solar radiation in sugars, which serve as the primary food and energy source of the entire ecosystem. Thus, if long-term disturbances in the system, such as a change in climatic conditions or in the chemical composition of the ambient air, affect plant productivity, the entire ecosystem can be destabilized. In my ESDL project I analyse long-term changes in plant growth and in its control variables.

Changes in the abundance of living plants on the land surface can be assessed by satellite observations of the green leaf area. The green leaf area is best described by the leaf area index (LAI), which expresses the area of plant leaves per ground area. In strongly vegetated areas such as the tropical forest, LAI values of up 7 m2 green leaf area per 1 m2 ground area can be measured. In general, an increase in LAI is referred to as vegetation greening and browning in case of a decrease. Various Earth observation campaigns have shown that many of these ecosystems exhibit an increase in leaf area in recent decades. However, also strong browning clusters strike out, such as the tropical forests in the Congo Basin which, has suffered severe losses of leaf area according to the satellite data. We study these ecosystem-specific patterns of long-term changes in remote-sensing LAI which provide information on how the various ecosystems have developed over the period of the satellite era. To interpret these changes, we need to assess various Earth system variables to find out what is behind the observed changes in leaf area.

Regional maps of trends and temporal correlation of leaf area index (LAI) and dry season precipitation in Africa. Statistically significant trends in annual average LAI (a) and in dry season (June – August) precipitation (b) in Africa, and their 1-year lag temporal correlation coefficient (c, LAI is shifted forward by one year). The land cover map (d) displays the distribution of broad vegetation classes in Africa based on the International Geosphere-Biosphere Programme (IGBP) classification, aggregated in anthropogenic (Croplands) and seven natural vegetation classes (Tropical, Temperate, Savannas, Grasslands, and Shrublands). This analysis was performed using the Earth System Data Lab.

The two key modulators of plant growth are ambient temperature and water availability. With the increase of atmospheric CO2 both these environmental variables are prone to substantial changes, which can induce vegetation greening or browning. With the help of the ESDL we can not only easily analyse primary variables such as air temperature and precipitation, but also derive more sophisticated indicators, such as the aridity index or the vapour pressure deficit (VPD). VPD describes the difference, or deficit, between the amount of moisture in the air and the potential moisture the air could hold if it was saturated. An increase in VPD thus indicates an increase in atmospheric water demand, which could put plants into water stress. Another concept to assess changes in the dryness of a given climate is the aridity index which combines precipitation and evapotranspiration fluxes. As temperatures rise, aridity is expected to increase in large parts of the world with serious consequences for ecosystems, especially if they are already water-limited.

The analysis in the EDSL project showed that the spatial patterns of change of the different indicators and variables correspond well with the changes in the leaf area. Especially the tropical forests in the Congo Basin emerge as a prominent cluster of leaf area loss which coincides with a strong decrease in observed precipitation (Figure).

This analysis illustrates statistical correlations and only indicate that the long-term decline in water availability in the Congo Basin in recent decades could have been the cause of tropical forest browning. In theory, however, there could also be a causal relationship in the other direction. Besides climatic change, also plants themselves react to the increased supply of CO2 in the atmosphere, which they need for the process of photosynthesis, and reduce the gas exchange with the ambient air. As a result, they lose less water to the atmosphere, which could lead to a drying of the atmosphere. Recent studies have shown that this effect could be strong enough to reduce local convective precipitation.