Quantifying patterns of tropical forest aboveground net primary productivity (ANPP) is critical for understanding the global carbon cycle. Field measurements of ANPP are generally made in plots that are small compared to the landscapes they represent. The small scale of ANPP measurements limits our ability to understand drivers of ANPP variation over large scales relevant for global carbon fluxes and Earth System Models. Positive global correlations exist between ANPP, total leaf area, and aboveground carbon density (ACD) because over large climatic gradients increased leaf area reflects higher light absorption and water availability and supports more plant tissue. However, it is not clear whether this relationship is valid when climatic gradients are held constant. We used lidar-derived models to estimate total leaf area, ANPP, and aboveground carbon density (ACD) over 770 ha of old-growth tropical forest at La Selva, Costa Rica. We used these data to interrogate whether positive correlations between leaf area, ANPP, and ACD exist at the landscape-scale in a tropical forest. Additionally, we used a Random Forest model to evaluate whether topographic metrics (slope, elevation, catchment area, and aspect) explain variation in ANPP, total leaf area, and ACD.
Results/Conclusions
The landscape-level mean ANPP, predicted from vertical leaf-area profiles, across 1539 0.5 ha samples was 13.5 Mg ha-1 yr-1, which agrees closely with the plot-based estimate of 13.9 Mg C ha-1 yr-1. We found a positive relationship between ACD and total leaf area (r2 = 0.52, P < 0.001), but significant negative correlations between ANPP and both total leaf area (r2 = 0.36, P < 0.001) and ACD (r2 = 0.08, P < 0.001). Despite the global positive relationship between leaf area and ANPP, our results demonstrate there can be a negative relationship between leaf area and ANPP when climate is held constant. Given that absorbed photosynthetically active radiation increases with leaf area, these results imply that it is not the quantity of absorbed light that drives ANPP variation, but the efficiency with which light drives net production. Topographic metrics explained only 9% of ANPP variation, despite previous studies have found that species distributions and soil nutrient concentrations vary with topography at La Selva. All topographic metrics (slope, elevation, catchment area, and aspect) were approximately equally important for explaining ANPP variation.