Temporal and spatial patterns of remotely sensed litterfall in tropical and subtropical forests of Taiwan

Abstract

AbstractLitterfall is important for returning nutrients and carbon to the forest floor, and microbes decompose the litterfall to release CO2 into the atmosphere. Litterfall is a pivotal component in the forest biogeochemical cycle, which is sensitive to climate variability and plant physiology. In this study, we combined field litterfall estimates and time series (2001–2011) climate (the Moderate Resolution Imaging Spectroradiometer (MODIS) land surface temperature (LST) and Tropical Rainfall Measuring Mission (TRMM) precipitations) and green vegetation (MODIS photosynthetically active vegetation cover (PV)) variables to estimate regional annual litterfall in tropical/subtropical forests in Taiwan. We found that time series MODIS LST‐ and PV‐derived metrics, the annual accumulated MODIS LST, and coefficient of variation of PV, respectively, but not the TRMM precipitation variables were salient factors for the estimation (r2 = 0.548 and p < 0.001). The mean (±standard deviation) annual litterfall was 5.1 ± 1.2 Mg ha−1 yr−1 during the observation period. The temporal dynamics of the litterfall revealed that typhoons and consecutive drought events might affect the litterfall temporal variation. Overall, the annual litterfall decreased along the elevation gradient, which may reflect a change in the vegetation type. The northeast and northwest facing slopes yielded the highest amount of annual litterfall (≥5.9 Mg ha−1 yr−1), which was in contrast with the southern aspect (5.1 Mg ha−1 yr−1). This variation may be associated with the dryness of the microclimate influenced by solar radiation. This study demonstrates the feasibility of utilizing time series MODIS LST and PV data to predict large‐scale field litterfall, which may facilitate large‐scale monitoring of biogeochemical cycles in forest ecosystems.