Environmental changes in biodiversity hotspot ecosystems of
South Ecuador: RESPonse and feedback effECTs (FOR2730)

Abstract:

Understanding the response of net nitrogen (N) mineralization to climate and land-use change is important to predict the effect of environmental change on the biodiversity and nutrient supply of the tropical montane forest in Ecuador. Slow mineralization or microbial immobilization may limit the availability of N, although organic N stocks in soils are high. To determine the roles of land use and climate for N mineralization in the mineral topsoil, field incubations over 31 days were conducted with the help of soil-filled PVC cylinders in forest and pasture soils along a land-use and elevation gradient from 1000-3000 m above sea level (a.s.l.). The experiment made use of a threefold replicated, full-factorial design with two land-use types and three elevations. The incubation cylinders were closed both at the bottom and the top and therefore only permitted horizontal water and element fluxes through lateral slits. Start and end concentration of NH4-N and NO3-N were determined in 1 M KCl extracts to calculate net N mineralization as the sum of ammonification and nitrification rates. Ammonification was significantly higher under pasture than under forest, except at 3000 m a.s.l., where the highest ammonification rates were detected on the forest plots. Nitrification was, in contrast, significantly higher under forest than under pasture, with highest nitrification rates at 2000 m a.s.l. Although the climate was wetter and cooler at higher elevations, mean N mineralization in the mineral soil on the forest sites significantly increased with elevation. On pastures, N mineralization was not significantly related with elevation, which may be explained by soil management and farming intensity. The slope of the regression line of δ13C values on soil organic C concentrations in 10-cm soil layers revealed a close relationship with the N mineralization rates, indicating that the short-term field incubations reflected the long-term C mineralization regime. The shift to higher δ13C values with increasing depth of the soil profile was related to N turnover and could thus serve as predictor of N mineralization rates on the forest sites, but not on the pastures, where the vertical distribution of the δ13C values was altered by the input of organic matter from C4 grasses after land-use change. Furthermore, the relationship between the slope of the regression line of the enrichment of δ13C values on soil depth and N mineralization measured by in-situ incubation depended on the parent material. The topsoils developed from granodiorite showed lower N mineralization rates than those developed from phyllite and meta-sandstone. Different soil texture, rooting depth, nutrient availability, and different organic layers may provide possible explanations for the observed differences.

Abstract:

Dissolved organic matter (DOM) contributes to forest C cycling. We assessed temporal variability, sources, and transformations of DOM during four years in a tropical montane forest with the help of stable C isotope ratios (δ13C values). We measured δ13C values of DOM in rainfall (RF), throughfall (TF), stemflow (SF), litter leachate (LL), soil solutions at the 0.15 and 0.30 m depths (SS15, SS30), and streamflow (ST)with TOC-IRMS. The δ13C values of DOM did not vary seasonally. We detected an event with a high δ13C value likely attributable to black carbon from local pasture fires. The mean δ13C values of DOM outside the event decreased in the order, RF (−26.0 ± 1.3‰) > TF (−28.7 ± 0.3‰) > SF (−29.2 ± 0.2‰) > LL (−29.6 ± 0.2‰) because of increasing leaching of C-isotopically light compounds. The higher δ13C values of DOM in SS15 (−27.8 ± 1.0‰), SS30 (−27.6 ± 1.1‰), and ST (−27.9 ± 1.1‰) than in the above-ground solutions suggested that roots and root exudates are major belowground DOM sources. Although in DOM the C/N ratios correlated with the δ13C values when all solutions were considered, this was not the case for SS15, SS30, and ST alone. Thus, the δ13C values of DOM provide an additional tool to assess the sources and turnover of DOM.

Abstract:

Tropical regions are currently undergoing remarkable rates of land use change accompanied by altered litter inputs to soil. In vast areas of Southern Ecuador forests are clear cut and converted for use as cattle pastures. Frequently these pasture sites are invaded by bracken fern, when bracken becomes dominant pasture productivity decreases and the sites are abandoned. In the present study implications of invasive bracken on soil biogeochemical properties were investigated. Soil samples (0–5 cm) were taken from an active pasture with Setaria sphacelata as predominant grass and from an abandoned pasture overgrown by bracken. Grass (C4 plant) and bracken (C3 plant) litter, differing in C:N ratio (33 and 77, respectively) and lignin content (Klason-lignin: 18% and 45%, respectively), were incubated in soils of their corresponding sites and vice versa for 28 days at 22 C. Unamended microcosms containing only the respective soil or litter were taken as controls. During incubation the amount of CO2 and its d13C-signature were determined at different time intervals. Additionally, the soil microbial community structure (PLFA-analysis) aswell as the concentrations of KCl-extractable C and N were monitored. The comparison between the control soils of active and abandoned pasture sites showed that the massive displacement of Setaria-grass by bracken after pasture abandonment was characterized by decreased pH values accompanied by decreased amounts of readily available organic carbon and nitrogen, a lower microbial biomass and decreased activity as well as a higher relative abundance of actinomycetes. The d13C-signature of CO2 indicated a preferential mineralization of grass-derived organic carbon in pasture control soils. In soils amended with grass litter the mineralization of soil organic matter was retarded (negative priming effect) and also a preferential utilization of easily available organic substances derived from the grass litter was evident. Compared to the other treatments, the pasture soil amended with grass litter showed an opposite shift in the microbial community structure towards a lower relative abundance of fungi. After addition of bracken litter to the abandoned pasture soil a positive priming effect seemed to be supported by an N limitation at the end of incubation. This was accompanied by an increase in the ratio of Gram-positive to Gram-negative bacterial PLFA marker. The differences in litter quality between grass and bracken are important triggers of changes in soil biogeochemical and soil microbial properties after land use conversion.