Patterns and driving factors of litter decomposition across Chinese terrestrial ecosystems

Litter decomposition is a vital component of carbon and nutrients cycling and energy flow within terrestrial ecosystems. Understanding the litter decomposition rate with a temperature change of 1 degrees C (k(DD), degree days(-1)) is critical for quantifying the climate-carbon, nutrients, and energy cycle feedback and predicting the response of ecosystems to climate change. However, the spatial pattern of k(DD) is uncertain, given differences in litter quality and soil property responses to temperature. Therefore, our goal was to explore the spatial patterns and driving factors of the k(DD) with a synthesized dataset, which included 1370 individual studies from 253 publications and accumulated daily temperature of each site from 389 meteorological stations. The dataset covered forest, grassland, and cropland ecosystems. The k(DD) significantly decreased with increasing latitude and varied with climate variables, litter quality, soil properties, and experimental duration under different ecosystems. The averaged k(DD) in the forest, grassland, and cropland ecosystems were 1.30, 2.23, and 3.35 x 10(-4) degree days(-1), respectively. The k(DD) markedly decreased with the increase of experimental years. Climate variables [mean annual temperature (MAT) and precipitation (MAP)] accounted for 36.8% of the total variance in the k(DD), followed by soil properties (e.g., soil sand content and pH; 28.1%); experimental duration (23.0%); and litter properties (e.g., nitrogen and lignin; 12.1%). MAT and MAP had substantially indirect effects on k(DD) via regulating soil and litter properties. Lignin directly and negatively affected k(DD). Higher correlations between k(DD) and litter properties were maintained if the effects of climate, soil properties, and experimental duration were removed. Overall, our findings highlight that intrinsic (litter properties) and extrinsic (climatic and soil properties) variables directly and indirectly regulated the k(DD) at a regional scale, respectively, providing a framework for optimizing the ecosystem process model to global warming scenarios. Meantime, these results clarify the crucial role of environmental variables in regulating the k(DD), potentially affecting the terrestrial carbon, nutrients, and energy-climate feedback. (C) 2020 Elsevier Ltd. All rights reserved.