Skip to main content
Dryad

Data from: Vegetation type controls root turnover in global grasslands

Data files

Aug 15, 2019 version files 53.59 KB

Abstract

Abstract: Aim: Root turnover is an important process determining carbon and nutrient cycling in terrestrial ecosystems. It is an established fact that root turnover is jointly regulated by climatic, edaphic, and biotic factors. However, the relative importance of these forces in determining the global patterns of root turnover time is far from clear. Location: Global. Time period: 1946–2017. Major taxa studied: Grasslands. Methods: We compiled a database of 141 sites with 433 observations on root turnover time and applied structural equation modelling (SEM) to investigate the relative contribution of climate, soil properties, and vegetation type to the observed variations in root turnover time. Results: Root turnover time was 3.1 years on average across the global grasslands and differed significantly among grassland types (tropical grassland & savanna, temperate grassland & meadow, alpine grassland & meadow, tundra, and desert). It decreased with mean annual temperature, mean annual precipitation, and Palmer Drought Severity Index but increased with soil organic carbon content, total nitrogen content, and carbon: nitrogen ratio. Soil bulk density and soil texture also significantly affected root turnover time, with clay content negatively correlating to root turnover time and explaining more variations than bulk density and sand content. The SEM showed that climatic factors had dominant effects on root turnover time when vegetation type was not considered. Vegetation type became the primary driver when it was included in the SEM. Main conclusions: Our results highlight that the influences of climatic and edaphic factors on root turnover time are predominantly manifested through vegetation type. The critical role of precipitation as revealed for the first time in this study challenges our current understanding of climate impacts on root turnover time. The findings necessitate accurate representation of vegetation type in Earth system models to predict root function dynamics under global change.