Climate change due to greenhouse gas emissions is predicted to raise the mean global temperature by 1.0-3.5°C in the next 50-100 years. The direct and indirect effects of this potential increase in temperature on terrestrial ecosystems and ecosystem processes are likely to be complex and highly varied in time and space. The Global Change and Terrestrial Ecosystems core project of the International Geosphere-Biosphere Programme has recently launched a Network of Ecosystem Warming Studies, the goals of which are to integrate and foster research on ecosystem-level effects of rising temperature. In this paper, we use meta-analysis to synthesize data on the response of soil respiration, net N mineralization, and aboveground plant productivity to experimental ecosystem warming at 32 research sites representing four broadly defined biomes, including high (latitude or altitude) tundra, low tundra, grassland, and forest. Warming methods included electrical heat-resistance ground cables, greenhouses, vented and unvented field chambers, overhead infrared lamps, and passive night-time warming. Although results from individual sites showed considerable variation in response to warming, results from the meta-analysis showed that, across all sites and years, 2-9 years of experimental warming in the range 0.3-6.0°C significantly increased soil respiration rates by 20% (with a 95% confidence interval of 18-22%), net N mineralization rates by 46% (with a 95% confidence interval of 30-64%), and plant productivity by 19% (with a 95% confidence interval of 15-23%). The response of soil respiration to warming was generally larger in forested ecosystems compared to low tundra and grassland ecosystems, and the response of plant productivity was generally larger in low tundra ecosystems than in forest and grassland ecosystems. With the exception of aboveground plant productivity, which showed a greater positive response to warming in colder ecosystems, the magnitude of the response of these three processes to experimental warming was not generally significantly related to the geographic, climatic, or environmental variables evaluated in this analysis. This underscores the need to understand the relative importance of specific factors (such as temperature, moisture, site quality, vegetation type, successional status, land-use history, etc.) at different spatial and temporal scales, and suggests that we should be cautious in "scaling up" responses from the plot and site level to the landscape and biome level. Overall, ecosystem-warming experiments are shown to provide valuable insights on the response of terrestrial ecosystems to elevated temperature.

We examined the spatial distribution of soil nutrients in desert ecosystems of the southwestern United States to test the hypothesis that the invasion of semiarid grasslands by desert shrubs is associated with the development of "islands of fertility" under shrubs. In grasslands of the Chihuahuan Desert of New Mexico, 35—76% of the variation in soil N was found at distances <20 cm, which may be due to local accumulations of soil N under Bouteloua eriopoda, a perennial bunchgrass. The remaining variance is found over distances extending to 7 m, which is unlikely to be related to nutrient cycling by grasses. In adjacent shrublands, in which Larrea tridentata has replaced these grasses over the last century, soil N is more concentrated under shrubs and autocorrelated over distances extending 1.0—3.0 m, similar to mean shrub size and reflecting local nutrient cycling by shrubs. A similar pattern was seen in the shrublands of the Mojave Desert of California. Soil PO_4, Cl, SO_4, and K also accumulate under desert shrubs, whereas Rb, Na, Li, Ca, Mg, and Sr are usually more concentrated in the intershrub spaces. Changes in the distribution of soil properties may be a useful index of desertification in arid and semiarid grasslands worldwide.

In a range of plant species from three Californian vegetation types, we examined the widely used "normalized difference vegetation index" (NDVI) and "simple ratio" (SR) as indicators of canopy structure, light absorption, and photosynthetic activity. These indices, which are derived from canopy reflectance in the red and near‐infrared wavebands, highlighted phenological differences between evergreen and deciduous canopies. They were poor indicators of total canopy biomass due to the varying abundance of non‐green standing biomass in these vegetation types. However, in sparse canopies (leaf area index [LAI]°0‐2), NDVI was a sensitive indicator of canopy structure and chemical content (green biomass, green leaf area index, chlorophyll content, and foliar nitrogen content). At higher canopy green LAI values (>2; typical of dense shrubs and trees), NDVI was relatively insensitive to changes in canopy structure. Compared to SR, NDVI was better correlated with indicators of canopy structure and chemical content, but was equivalent to the logarithm of SR. In agreement with theoretical expectations, both NDVI and SR exhibited near‐linear correlations with fractional PAR intercepted by green leaves over a wide range of canopy densities. Maximum daily photosynthetic rates were positively correlated with NDVI and SR in annual grassland and semideciduous shrubs where canopy development and photosynthetic activity were in synchrony. The indices were also correlated with peak springtime canopy photosynthetic rates in evergreens. However, over most of the year, these indices were poor predictors of photosynthetic performance in evergreen species due to seasonal reductions in photosynthetic radiation‐use efficiency that occurred without substantial declines in canopy greenness. Our results support the use of these vegetation indices as remote indicators of PAR absorption, and thus potential photosynthetic activity, even in heterogeneous landscapes. To provide accurate estimates of vegetation‐atmosphere gas fluxes, remote NDVI and SR measurements need to be coupled with careful estimates of canopy photosynthetic radiation‐use efficiency.

Whether climate change will turn cold biomes from large long-term carbon sinks into sources is hotly debated because of the great potential for ecosystem-mediated feedbacks to global climate. Critical are the direction, magnitude and generality of climate responses of plant litter decomposition. Here, we present the first quantitative analysis of the major climate-change-related drivers of litter decomposition rates in cold northern biomes worldwide. Leaf litters collected from the predominant species in 33 global change manipulation experiments in circum-arctic-alpine ecosystems were incubated simultaneously in two contrasting arctic life zones. We demonstrate that longer-term, large-scale changes to leaf litter decomposition will be driven primarily by both direct warming effects and concomitant shifts in plant growth form composition, with a much smaller role for changes in litter quality within species. Specifically, the ongoing warming-induced expansion of shrubs with recalcitrant leaf litter across cold biomes would constitute a negative feedback to global warming. Depending on the strength of other (previously reported) positive feedbacks of shrub expansion on soil carbon turnover, this may partly counteract direct warming enhancement of litter decomposition.

1 Macrolichens are important for the functioning and biodiversity of cold northern ecosystems and their reindeer-based cultures and economies. 2 We hypothesized that, in climatically milder parts of the Arctic, where ecosystems have relatively dense plant canopies, climate warming and/or increased nutrient availability leads to decline in macrolichen abundance as a function of increased abundance of vascular plants. In more open high-arctic or arctic-alpine plant communities such a relationship should be absent. To test this, we synthesized cross-continental arctic vegetation data from ecosystem manipulation experiments simulating mostly warming and increased nutrient availability, and compared these with similar data from natural environmental gradients. 3 Regressions between abundance or biomass of macrolichens and vascular plants were consistently negative across the subarctic and mid-arctic experimental studies. Such a pattern did not emerge in the coldest high-arctic or arctic-alpine sites. The slopes of the negative regressions increased across 10 sites as the climate became milder (as indicated by a simple climatic index) or the vegetation denser (greater site above-ground biomass). 4 Seven natural vegetation gradients in the lower-altitude sub-and mid-arctic zone confirmed the patterns seen in the experimental studies, showing consistent negative relationships between abundance of macrolichens and vascular plants. 5 We conclude that the data supported the hypothesis. Macrolichens in climatically milder arctic ecosystems may decline if and where global changes cause vascular plants to increase in abundance. 6 However, a refining of our findings is needed, for instance by integrating other abiotic and biotic effects such as reindeer grazing feedback on the balance between vascular plants and lichens.

Models represent our primary method for integration of small-scale, processlevel phenomena into a comprehensive description of forest-stand or ecosystem function. They also represent a key method for testing hypotheses about the response of forest ecosystems to multiple changing environmental conditions. This paper describes the evaluation of 13 stand-level models varying in their spatial, mechanistic, and temporal complexity for their ability to capture intra-and interannual components of the water and carbon cycle for an upland, oak-dominated forest of eastern Tennessee. Comparisons between model simulations and observations were conducted for hourly, daily, and annual time steps. Data for the comparisons were obtained from a wide range of methods including: eddy covariance, sapflow, chamber-based soil respiration, biometric estimates of stand-level net primary production and growth, and soil water content by time or frequency domain reflectometry. Response surfaces of carbon and water flux as a function of environmental drivers, and a variety of goodness-of-fit statistics (bias, absolute bias, and model efficiency) were used to judge model performance.A single model did not consistently perform the best at all time steps or for all variables considered. Intermodel comparisons showed good agreement for water cycle fluxes, but considerable disagreement among models for predicted carbon fluxes. The mean of all model outputs, however, was nearly always the best fit to the observations. Not surprisingly, models missing key forest components or processes, such as roots or modeled soil water content, were unable to provide accurate predictions of ecosystem responses to short-term drought phenomenon. Nevertheless, an inability to correctly capture short-term physiological processes under drought was not necessarily an indicator of poor annual water and carbon budget simulations. This is possible because droughts in the subject ecosystem were of short duration and therefore had a small cumulative impact. Models using hourly time steps and detailed mechanistic processes, and having a realistic spatial representation of the forest ecosystem provided the best predictions of observed data. Predictive ability of all models deteriorated under drought conditions, suggesting that further work is needed to evaluate and improve ecosystem model performance under unusual conditions, such as drought, that are a common focus of environmental change discussions.

Abstract. We provide an assessment of the global sources of NH, in the atmosphere, which indicates an annual flux of about 75 Tg of N as NH,. The emissions from land are dominated by the release of NH, during the hydrolysis of urea from the urine of domestic animals (32 TgN/yr) and by emanations from soils in unmanaged' ecosystems (10 TgN/yr) and from fertilized agricultural soils (9 TgN/yr). Emissions from the sea surface may approach 13 TgN/yr. The total annual source of NH, is in reasonable agreement with estimates of global NH: deposition from the atmosphere, the major fate of atmospheric NH,. As an alkaline atmospheric species, NH, emitted to the atmosphere each year can neutralize only about 32% of the annual production of H+ in the atmosphere from natural and anthropogenic sources.

Summary• Arbuscular mycorrhizal fungi (AMF) are known to promote plant growth when phosphorus is limiting, but the role of AMF in plant growth under nitrogen (N) limiting conditions is unclear.• Here, we manipulated N (control vs inorganic and organic forms) and AMF species (control vs four AMF species) for five old-field perennials grown individually in a glasshouse under N-limiting conditions.• We found that AMF were at best neutral and that some AMF species depressed growth for some plant species (significant plant-fungus interaction). Native plant species growth was strongly depressed by all but one AMF species; exotic plant species were less sensitive to AMF. We found no evidence of plant N preferences. Both natives and exotics were able to acquire more N with N addition, but only exotics grew more with added N.• Our results suggest that AMF do not promote plant N acquisition at low N supply, and our results are consistent with other research showing that AMF can act as a parasitic carbon drain when phosphorus availability is relatively high.