TL;DR
This paper presents a novel empirical method to estimate the maximum potential tropical tree cover (MPTC) under future climate conditions. The study predicts a significant reduction in MPTC across the tropics by the end of the 21st century, particularly in forest-savanna transition zones, highlighting a critical challenge for the world's largest carbon pool.
What Problem Does This Paper Address?
Tropical forests are under immense pressure from both climate change (warming and drought) and human land-use. While complex Dynamic Global Vegetation Models (DGVMs) attempt to predict these changes, they exhibit considerable variation in their projections of tropical forest dieback. This uncertainty stems from the diverse ways these models represent the intricate relationships between vegetation and climate. The authors aim to address this by developing a more quantitative and empirically grounded approach to estimate the potential tree cover that a given climate can sustain, thereby isolating the impact of climate change itself from other factors. This is crucial for understanding the fundamental climate limits on tropical forests.
Methods
The core of the paper's methodology lies in developing an empirical model to estimate the "maximum potential tree cover" (MPTC). This is achieved by:
- Relating Evapotranspiration (ET) to Tree Cover (TC): They propose a linear relationship between ET and TC, where ET is a function of both the presence of trees (represented by TC) and the underlying climate. This relationship is expressed as
ET = a × TC + b. - Defining the Climate Envelope: The climate envelope for tropical trees is defined by mean annual air temperature (T) and annual precipitation (P).
- Estimating MPTC: Under the assumption that in equilibrium, all precipitation (P) can be utilized for potential tree growth (neglecting runoff and other water storage), MPTC is calculated as the tree cover fraction that makes ET equal to P. This yields equations to determine MPTC based on P, T, and the empirically derived parameters 'a' and 'b'.
- Incorporating CO2 Effects: To account for the impact of rising atmospheric CO2 on plant transpiration (via reduced stomatal conductance and increased water use efficiency), a modification to the ET equation is introduced, leading to an estimate of future MPTC (MPTC').
- Statistical Modeling: Linear regression analyses are used to model MPTC and MPTC' as functions of P, T, and their product, using data from present-day and projected future climate scenarios.
Datasets
The study utilizes several key datasets:
- Satellite-derived Tree Cover Fraction (TC): From the MODIS tree cover data product (0.25 km resolution, 2000-2010).
- Evapotranspiration (ET): Multiyear mean annual ET from the MOD16 production (1 km resolution, 2000-2010).
- Precipitation (P) and Temperature (T):
- For present-day (2000-2009): Climate Research Unit (CRU) TS3.1 datasets (0.5° × 0.5° resolution).
- For future climate (2090-2099): Outputs from 19 Global Climate Models (GCMs) from the IPCC 4th Assessment Report under the SRES A2 scenario.
- DGVM Model Outputs: Tree cover fractions simulated by four DGVM ecosystem models (HYL, LPJ, ORC, and TRI) under SRES A2 for comparison.
Key Insights and Ideas
- Climate as a Primary Limiter: The study strongly emphasizes that even without human land-use change, climate alone imposes significant limitations on tropical tree cover.
- MPTC vs. Actual Tree Cover: The empirical MPTC is consistently higher than observed actual tree cover, indicating that factors beyond climate (e.g., soil fertility, fire, herbivory, human activity) currently limit tree cover in many areas.
- Sensitivity to Climate Variables: MPTC is shown to be sensitive to both temperature and precipitation, with complex interactions. For instance, the negative impact of rising temperature on MPTC is amplified at lower precipitation levels.
- Forest-Savanna Transition Zones are Vulnerable: The most significant projected reductions in MPTC occur in the transitional zones between forests and savannas, suggesting these areas are particularly susceptible to climate-induced biome shifts.
- CO2's Mitigating Role: While warming and drought are projected to reduce MPTC, rising CO2 concentrations are shown to have a partially mitigating effect by increasing water use efficiency. However, this effect is not enough to offset the negative impacts of climate change entirely.
Key Results
- Present-Day MPTC: In 92% of tropical climate bins, the estimated potential MPTC is higher than the observed actual tree cover. On average, observed tree cover is only 51% of the potential MPTC.
- Future MPTC Reduction: Under projected 21st-century climate change (2090-2099), MPTC is predicted to be reduced by 1% to 15% across the tropical band.
- Regional Projections:
- In South America and Africa, projections show high uncertainty, with some GCMs indicating expansion and others significant shrinkage of MPTC, particularly in the Amazon and Congo rainforests.
- Southeast Asia and Australia show less spread in projections, with generally small changes in MPTC.
- Specific Vulnerabilities: The study infers a high probability (75%) of at least 0.7% reduction in the Congo rainforest and a medium probability (50%) of at least 5.2% shrinkage in the eastern Amazon rainforest by the end of the century.
Limitations and Future Work
- Empirical Model Assumptions: The study acknowledges that its empirical approach relies on specific assumptions, such as the linear relationship between ET and TC, and the neglect of runoff.
- Exclusion of Radiation: Net radiation, a significant factor affecting ET, could not be included due to a lack of high-resolution global datasets and uncertainty in future radiation changes. This may lead to an overestimation of future ET demand and an underestimation of MPTC.
- Dynamic vs. Equilibrium Response: The model estimates equilibrium MPTC, whereas real-world vegetation changes can lag behind climate stabilization, as suggested by DGVM studies.
- Non-linearities and Disturbances: The empirical model does not fully capture non-linear responses or the complex interplay of factors like fire, grazing, and human intervention, which can drive rapid vegetation shifts, especially in transition zones.
Future work suggested includes incorporating high-resolution runoff and net radiation datasets, and further exploring the complex interactions between climate, vegetation, and disturbance regimes.
Summary updated