No abstract

We are investigating techniques for developing distributed and adaptive collections of agents that coordinate to retrieve, lter and fuse information relevant to the user, task and situation, as well as anticipate a user's information needs. In our system of agents, information gathering is seamlessly integrated with decision support. The task for which particular information is requested of the agents does not remain in the user's head but it is explicitly represented and supported through agent collaboration. In this paper we present the distributed system architecture, agent collaboration interactions, and a reusable set of software components for constructing agents. We call this reusable multi-agent computational infrastructure RETSINA Reusable Task Structure-based Intelligent Network Agents. It has three types of agents. Interface agents interact with the user receiving user speci cations and delivering results. They acquire, model, and utilize user preferences to guide system coordination in support of the user's tasks. Task agents help users perform tasks by formulating problem solving plans and carrying out these plans through querying and exchanging information with other software agents. Information agents provide intelligent access to a heterogeneous collection of information sources. We h a ve implemented this system framework and are developing collaborating agents in diverse complex real world tasks, such as organizational decision making the PLEIADES system, and nancial portfolio management the WARREN system.

Given that plant growth is often water‐limited in grasslands, it has been proposed that projected increases in precipitation could increase plant productivity and carbon sequestration. However, the existing evidence for this hypothesis comes primarily from observational studies along natural precipitation gradients or from short‐term manipulative experiments. It remains unclear whether long‐term increased precipitation persistently stimulates grassland productivity. In the world's largest remaining temperate grassland, we found that experimentally increased precipitation enhanced net primary production, soil‐available nitrogen and foliar nitrogen concentrations during the first six years, but it ceased to do so in the following four years, unless nitrogen was simultaneously added with water. The 15N enrichment of plant and soil nitrogen pools in later years indicates increased nitrogen losses, which exacerbated nitrogen limitation and ended the stimulation of productivity by increased precipitation. Changes in species abundance might have contributed little to the changes in water treatment effects. Our study demonstrates that the long‐term response of grassland ecosystems to increased precipitation will be mediated by nitrogen availability. Our results also point to a shift from co‐limitation by water and nitrogen early to perhaps limitation by nitrogen only later in this temperate grassland, highlighting significant variations in the type of resource limitation induced by climate change.

Afforestation of marginal agricultural land has been considered to be an effective measure to sequester atmospheric CO 2 . In this study, we adopted the volume-and mass-based methods to investigate the changes in soil organic C and total N stocks in 100 cm depth following afforestation of marginal agricultural land using a chronosequence of poplar (Populus euramericana cv. "N3016") stands in a semiarid region of Liaoning Province, Northeast China. Our results showed that soil organic C and total N concentrations in 45-60 cm layer increased gradually following afforestation of agricultural land, whereas in 60-100 cm layer, they declined initially, and then increased with stand development. Based on volume-and mass-based methods, such land-use change caused initial declines in soil organic C and total N stocks, and then increases between the stand ages of 10 and 20. Forest soils recovered to the initial soil organic C and N stocks found in agricultural land at age 15. However, the volume-based method would underestimate the absolute organic C and N stocks compared with the mass-based methods. Our results suggest that afforestation of marginal agricultural land has the potential to sequester atmospheric CO 2 in soils in semiarid regions. Stand age, soil sampling depth and the methods used to quantify organic C and N stocks should be considered for accurate assessments of changes in soil organic C and N stocks.

We compared soil moisture content, pH, total organic carbon (C org ), total nitrogen (TN), total phosphorus (TP) and inorganic N (NH 4concentrations, soil potential C and N mineralization rates, soil microbial biomass C (C mic ), soil metabolic quotient (qCO 2 ), soil microbial quotient (C mic /C org ) and soil enzyme (urease and invertase) activities in semiarid sandy soils under three types of land cover: grassland, Mongolian pine (Pinus sylvestris var. mongolica) plantation, and elm (Ulmus punila)-grass savanna in southeastern Keerqin, in northeast China. Soil C org , TN and TP concentrations (0-10, 10-20, 20-40 and 40-60 cm) were lower while soil C/N and C/P ratios were higher in the plantation than in grassland and savanna. The effects of land cover change on NH 4 + -N and NO 3 − -N concentrations, soil potential nitrification and C mineralization rates in the surface soil (0-10 cm) were dependent on sampling season; but soil potential N mineralization rates were not affected by land cover type and sampling season. The effects of land cover change on C mic and qCO 2 of surface soil were not significant; but C mic /C org were significantly affected by land cover change and sampling season. We also found that land cover change, sampling season and land cover type× sampling season interaction significantly influenced soil enzyme (urease and invertase) activities. Usually soil enzyme activities were lower in the pine plantations than in grassland and savanna. Our results suggest that land cover change markedly influenced soil chemical and biological properties in sandy soils in the semiarid region, and these effects vary with sampling season.

No abstract

To clarify responses of plant and soil carbon (C) and nitrogen (N) pools in grassland ecosystem to N addition, a field experiment was performed in a grassland in Keerqin Sandy Lands, Northeast China. We investigated vegetation composition and C and N pools of plant and soil (0-30 cm) after five consecutive years of N addition at a rate of 20 g N m -2 y -1 . Vegetation composition and species diversity responded dramatically to N addition, as dominance by C 4 perennials was replaced with C 3 annuals. Carbon in aboveground pool increased significantly (over two-fold), mainly due to the increase of the C in aboveground living plants and surface litter, which increased by 98 and 134%, respectively. Although soil C did not change significantly, the root C pool decreased in response to 5 years of N addition. The total ecosystem C pool was not significantly impacted by N addition because the large soil pool did not respond to N addition, and the increase in aboveground C was offset by the decrease in root C pool. Moreover, N addition significantly increased the aboveground N pool, but had no significant effects on belowground and total ecosystem N pools. Our results suggest that in the mid-term N addition alters the C and N partitioning in above-and belowground pools, but has no significant effects on total ecosystem C and N pools in these N-limited grasslands.

No abstract