Search for: All records

Abstract Decomposition is the transformation of dead organic matter into its inorganic constituents. In most biomes, decomposition rates can be accurately predicted with simple mathematical models, but these models have long under‐predicted decomposition in globally extensive drylands.We posit that the exposed surface conditions characteristic of drylands make litter decomposition uniquely subject to microsite‐specific environmental controls and spatially variable microbial communities. As such, decomposition in dryland ecosystems—which are characterized by extremes in temporal heterogeneity of climate conditions and spatial heterogeneity of vegetation cover with corresponding microclimate variability—is a prime example of a macrosystems process that can be addressed by merging field data with new predictive process models operating across a hierarchical continuum of spatial scales and process resolutions.A macrosystems approach offers promise to reconcile model‐measurement discrepancies by integrating observations and experiments across multiple scales, from microsites (e.g. shrub sub‐canopy or intercanopy) to regions (e.g. across a 100s of km²study site with complex topography, precipitation and temperature) and ultimately to a continental perspective (e.g. North American drylands).Recent developments in technology and data availability position the scientific community to integrate laboratory, field, modelling and remote sensing approaches across a hierarchical range of scales to capture the spatiotemporal distribution of litter and environmental conditions needed to predict decay dynamics at the micro‐to‐macroscale. This multi‐scale approach promises a path forward to resolving a longstanding disconnect between measured data and modelled processes in dryland litter decomposition.Dryland litter decomposition presents an excellent case study for resolving spatially and temporally complex biogeochemical dynamics through a hierarchical, multidisciplinary macrosystems approach.We focus on dryland litter decomposition, but the hierarchical, multidisciplinary macrosystems approach we outline shows great potential for resolving other spatially and temporally complex biogeochemical processes across a wide range of ecosystems. Read the freePlain Language Summaryfor this article on the Journal blog. 

Our understanding of carbon and nutrient dynamics in globally vast and socioeconomically critical dryland ecosystems lags behind mesic systems. Litter decomposition models consistently underestimate measured decomposition in these regions. Both models and measurements largely represent spatially dominant intercanopy areas; however, little litter resides in these interspaces as transport vectors move litter to other microsites such as beneath plant canopies and buried in soil. Abiotic and biotic conditions differ among microsites, but few studies have characterized microsite impacts on decomposition. We collated data on microsites where litter accumulates. In microsites with sufficient available data, we used meta-analysis to test hypotheses on decomposition relative to litter in intercanopy spaces. Decomposition was lower under woody plant canopies than in intercanopy spaces. Buried litter decomposed faster than surface litter. There was no difference in decomposition between surface litter and litter suspended aboveground to simulate standing dead. All microsite contrasts had exceptions, suggesting that site-specific characteristics influence microclimate and subsequent decomposition. Extrapolation of decomposition rates to the landscape-level (using estimates of microsite-specific decomposition rates multiplied by litter pools), suggests that decomposition estimates based on intercanopy data alone underrepresent landscape-level decomposition. Thus, despite advances in the understanding of mechanistic decomposition drivers in drylands advancing, most studies are spatially unrepresentative analyses in intercanopy areas and this will underestimate decomposition at the landscape level. Expanding the ecological relevance of decomposition processes to be useful for predicting larger-scale carbon and nutrient dynamics requires improved characterization of dryland litter distribution, coupled with a mechanistic understanding of decomposition in microsites where litter accumulates.