Search

Highlights • Precipitation exclusion induced a reduction of NPP in savanna. • Precipitation exclusion decreased NPP of perennials but elevated for the annuals. • NPP of savannas exhibited significant inter-annual variation. • Herb community had more resilience to precipitation exclusion than woody community in savanna. Abstract Declines in precipitation are expected to affect plant performance and ecosystem carbon uptake. The response of ecosystem productivity to declines in precipitation and potential underlying mechanisms have been well studied in many biomes; however, little is known about the role of declines in precipitation and the involved mechanisms in savanna ecosystems. In a 4-year field precipitation manipulation experiment, we simulated four levels of precipitation exclusion (control, 30%, 50% and 70%) to assess the effects of declines in precipitation on net primary productivity (NPP) in a savanna ecosystem in southwestern China. NPP was strongly correlated with soil water content during the experimental period. Precipitation exclusion significantly decreased the NPP of the entire vegetation including trees, shrubs, perennials and litterfall but significantly increased the NPP of annuals. Our results suggested that precipitation exclusion can reduce the productivity of savannas and that plant functional types differ in sensitivity to precipitation exclusion. These findings imply that future declines in precipitation in savanna regions may negatively impact carbon accumulation and may induce shifts in plant functional types to buffer the effects of declines in precipitation on productivity and stabilize ecosystem function in savannas. [ABSTRACT FROM AUTHOR]

The potential impact of drought on the carbon balance in plants has gained great attention. Non-structural carbohydrate (NSC) dynamics have been suggested as an important trait reflecting carbon balance under drought conditions. However, NSC dynamics under drought and the response mechanisms of NSC to drought remain unclear, especially in water-limited savanna ecosystems. A precipitation exclusion experiment was performed to simulate different drought intensities in a savanna ecosystem in Yuanjiang valley in southwestern China. Growth, total NSC concentration and diurnal change of NSC were determined for the leaves and non-photosynthetic organs of three dominant species (Lannea coromandelica, Polyalthia cerasoides and Heteropogon contortus) throughout the growing season. Drought significantly reduced the growth of all the three species. Total NSC concentration averaged ~8.1%, varying with species, organ and sampling period, and did not significantly decrease under drought stress. By contrast, the diurnal change of NSC in these three species increased under drought stress. These results indicate that these three dominant species did not undergo carbon limitation. Thus, relative change in NSC is a more sensitive and effective indicator than carbon reserves in evaluation of plant carbon balance. These findings provide new insights for the understanding of carbon balance and the mechanisms of carbon starvation. [ABSTRACT FROM AUTHOR]

Savanna ecosystems play a crucial role in global N 2 O emissions. However, our understanding of N 2 O emissions under limiting precipitation conditions is lacking. This study evaluates the effects of precipitation reduction on soil N 2 O fluxes from a woody savanna ecosystem in Yunnan Province, Southwest China. Precipitation exclusion shelters were installed above the tree canopy, and four total treatments were established as follows: a control (CK) and precipitation exclusions of 30% (PE3), 50% (PE5), and 70% (PE7). Two years (2015–2016) of N 2 O fluxes, soil temperature and soil water content data were collected. The N 2 O fluxes were generally low, ranging from 0.039 to 0.245 mg N m −2 day −1 , and they were strongly linked to precipitation events. Additionally, the N 2 O fluxes during the rainy season were significantly greater than those during the dry season. The maximum N 2 O flux was observed in August, and the minimum flux occurred in December. Precipitation exclusion had a significant negative influence on the N 2 O fluxes. The N 2 O emissions of CK, PE3, PE5, and PE7 were 0.20, 0.17, 0.13, and 0.12 kg N ha −1 yr −1 , respectively. With the exacerbation of precipitation exclusion, the decrease rate of precipitation exclusion on the N 2 O emissions increased over the entire year (eventually reaching 41.8% in PE7), but the decrease rate of precipitation exclusion on the soil N 2 O emission during the dry season was stronger than that during the rainy season. Additionally, the proportion of dry season N 2 O emissions to total annual emissions decreased (from 45% to 41%), and that of rainy season N 2 O emissions to total annual emissions increased (from 55% to 59%) over the year, whereas they exhibited a stable trend from PE5. The data show that the Yuanjiang savanna is a net source of N 2 O; precipitation reduction decreases the N 2 O emissions in the savanna regions, indicating that precipitation reduction can only slow the increase in the N 2 O concentration in the atmosphere and can therefore slow global warming. In addition, the N 2 O emissions during the dry season may play a significant role in total N 2 O emissions and be more sensitive to precipitation reduction than those during rainy season. These possibilities should be considered in future studies, especially in those ecosystems that experience substantial inter-annual climatic fluctuations. [ABSTRACT FROM AUTHOR]