Cross‐biome transplants of plant litter show decomposition models extend to a broader climatic range but lose predictability at the decadal time scale

• Published in: Global Change Biology
• Main Authors: CURRIE, W. S., HARMON, M. E., BURKE, I. C., HART, S. C., PARTON, W. J., SILVER, W.
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2010
• Subjects: Arctic; Tundra
• Online Access: http://dx.doi.org/10.1111/j.1365-2486.2009.02086.x; https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1111%2Fj.1365-2486.2009.02086.x; https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1365-2486.2009.02086.x

Abstract We analyzed results from 10‐year long field incubations of foliar and fine root litter from the Long‐term Intersite Decomposition Experiment Team (LIDET) study. We tested whether a variety of climate and litter quality variables could be used to develop regression models of decomposition parameters across wide ranges in litter quality and climate and whether these models changed over short to long time periods. Six genera of foliar and three genera of root litters were studied with a 10‐fold range in the ratio of acid unhydrolyzable fraction (AUF, or ‘lignin’) to N. Litter was incubated at 27 field sites across numerous terrestrial biomes including arctic and alpine tundra, temperate and tropical forests, grasslands and warm deserts. We used three separate mathematical models of first‐order (exponential) decomposition, emphasizing either the first year or the entire decade. One model included the proportion of relatively stable material as an asymptote. For short‐term (first‐year) decomposition, nonlinear regressions of exponential or power function form were obtained with r 2 values of 0.82 and 0.64 for foliar and fine‐root litter, respectively, across all biomes included. AUF and AUF : N ratio were the most explanative litter quality variables, while the combined temperature‐moisture terms AET (actual evapotranspiration) and CDI (climatic decomposition index) were best for climatic effects. Regressions contained some systematic bias for grasslands and arctic and boreal sites, but not for humid tropical forests or temperate deciduous and coniferous forests. The ability of the regression approach to fit climate‐driven decomposition models of the 10‐year field results was dramatically reduced from the ability to capture drivers of short‐term decomposition. Future work will require conceptual and methodological improvements to investigate processes controlling decadal‐scale litter decomposition, including the formation of a relatively stable fraction and its subsequent decomposition.