Search

Plant litter derived from the living biomass is the primary source of soil carbon (C). However, the contribution and the mechanism of plant litter production to soil C remains uncertain in aeolian regions. In this study, the relationship between plant biomass, litter mass, and soil C content was analyzed after a spatial field investigation in Horqin sandy land, northeast China. Long-term monitoring of wind speed and plant litter production was also conducted in three habitats (mobile dune, semi-fixed dune, and fenced grassland). The soil C content was significantly affected by belowground biomass and belowground litter mass but not aboveground biomass and aboveground litter. The litter production in the desertified habitat (mobile dune and semi-fixed dune) significantly differed with that of the undesertified habitat (fixed dune and fenced grassland). In mobile dune, 38% of the aboveground primary production was lost, while there was a distinct enrichment of plant litter in semi-fixed dune and fenced grassland. Wind events (wind speeds of 7 and 8 m s−1) at the end of the growing season significantly affected litter production rate, and wind was a major driver for litter transportation between habitats. In summary, plant litter production can be affected by wind events, and this influence differs among habitats in degraded grassland, leading to spatial heterogeneity of soil C accumulation. These results can strengthen the understanding of the relationship between soil properties and plant litter production in aeolian regions and furthermore, can provide information for ecological restoration of different types of desertified land, especially for the soil C accumulation. Plant litter is the major source for soil organic matter accumulation. In degraded sandy grasslands in semi-arid regions, wind is a visible factor which transported the aboveground litter from regions with lower vegetation coverage (e.g. mobile dune or semi-fixed dune) to regions with higher vegetation coverage (e.g. fixed dune or fenced grassland), which means that, wind is the driving force for the source-sink relationships of plant litters. There remains no clear description of the influence of wind events on litter production and litter distribution. The contribution of plant litter production to soil C remains uncertain in aeolian regions. The schematic diagram strengthened the understanding of the relationship between soil properties and plant litter production, and providing important information for ecological restoration in semi-arid sandy grasslands. [Display omitted] [ABSTRACT FROM AUTHOR]

• The ecosystem carbon fluxes were affected by precipitation variation. • Drought weakened the capacity of the ecosystem carbon sink. • Drought decreased GEP, NEP and Reco. • Main driver factors of GEP and NEP were SWC, coverage and standing biomass. • Ecosystem was converted from carbon sink to carbon source with press-drought. Precipitation has an impact on both gross ecosystem productivity (GEP) and ecosystem respiration (Reco), which ultimately influences net ecosystem productivity (NEP). A positive NEP denotes an ecosystem functioning as a carbon sink; whereas, a negative NEP denotes an ecosystem functioning as a carbon source. Therefore, drought plays an important role in the carbon balance of an ecosystem. However, little is known about the point at which the ecosystem converts from a carbon sink to a carbon source in extreme droughts. Such knowledge is crucial for predicting terrestrial carbon cycling under climate change, and consequently, was the subject of this study. We imposed two types of drought treatments on desert-grassland: (1) press-drought, in which the quantity of natural precipitation was reduced by 66% from May to August; and (2) pulse-drought, in which the quantity of natural precipitation was reduced by 100% during June and July. Reco and NEP were measured and GEP was calculated and then regression analyses were employed to determine the point at which the carbon sink shifts to a carbon source. The regression equation of NEP (μmol·m−2·s−1) on Reco (μmol·m−2·s−1) took the form: NEP = 0.504 Reco – 0.086 and, consequently, when Reco equaled 0.171 μmol·m−2·s−1, there was zero change in the carbon sink and GEP also equaled 0.171 μmol·m−2·s−1. Below this value, the ecosystem functioned as a carbon source, whereas above this value, the ecosystem functioned as a carbon sink. Structural equation models (SEM) demonstrated that coverage, standing biomass, pH, soil water content and total soil carbon were the main driving factors on desert grassland ecosystem carbon fluxes. [ABSTRACT FROM AUTHOR]