Climate warming is one key issue of global change and plays an important role on carbon cycling in terrestrial ecosystems. This paper reviews the recent advance in our understanding of global warming and its impacts on terrestrial carbon cycling and underlying mechanisms. We also discuss the state-of-the art in ecosystem modeling and its applications to ecosystem assessment. Climate warming will influence terrestrial carbon cycling in several aspects: 1) net primary productivity (NPP) will decrease in low-latitude region, and increase in mid- and high-latitude zones, totally show an increase on the global scale; 2) soil respiration will increase at the initial stage and then keep relatively stable because of biotic adaptability; 3) plant carbon storage will increase in high-latitude region, and remain stable or even decrease in low-latitude zone, and show a slight increase on the global level; 4) the production and decomposition rate of litterfall will increase; 5) the decomposition rate of soil organic carbon will increase and thus decrease soil carbon stock, meanwhile, soil carbon stock will increase for more carbon input from plant litter. These two processes will trade off in certain degree, resulting in different results for varied ecosystems. On the global scale, soil carbon stock will show a decrease; 6) although the different performances of diverse ecosystems, the global terrestrial ecosystem acts as a weak carbon source. Biophysical, biogeographical and biogeochemical models were developed in the past decades for global change research. In the future research, there are an urgent need to address the interaction among climate warming and other factors including elevated CO₂, O₃, drought, fire disturbance. The big challenge we are facing is how to deal with the complexity with multi-factors and multi-scales by using experimental and modeling approaches.

Land use/cover change (LUCC) is one of the most concerned environmental problems by scientists, land managers and policy makers. LUCC can affect energy flow, biogeochemical and hydrological cycling in terrestrial ecosystems through altering land surface and species composition. Ecosystem carbon cycling responds differently to various LUCC types, showing a pattern of CO₂ release into the atmosphere when LUCC from a high-biomass forest to low-biomass grassland, cropland or urban area. Previous reports indicated that global terrestrial ecosystem released 2.21 Pg C (1 Pg C=10¹⁵ g C) per year induced by LUCC during the 1990s, which explains about 25% of the global C emission per year in the same period; and in the last two centuries, the released C from LUCC accounts for 50% of the C emission from fossil fuel combustion. The LUCC patterns are totally diversified for regions around the world, which cause obviously different C fluxes among them. The reports showed that LUCC in the tropics is a C source, while it is a C sink in the middle and high latitude regions in the northern hemisphere, which possibly explain a large part of the “missing carbon sink" in the terrestrial ecosystems. Currently, modeling is the most popular way to simulate LUCC-induced changes in ecosystem C cycling. The quantitative relationship between LUCC patterns and their related processes and ecosystem carbon cycling remains uncertain. This uncertainty causes great discrepancies in the estimation of terrestrial ecosystem CO₂ fluxes from land use/cover changes. In the near future, except for carrying on long-term experiments to determine these quantitative relationships, model development by integrating LUCC with vegetation dynamic model and ecosystem process model will be essential for making an accurate estimation of C fluxes induced by LUCC. Sound land management can greatly increase C storage in the terrestrial ecosystems during LUCC processes. However, the quantification of land management effects is not well-established yet and land management is thus not included in most simulation models of LUCC impacts, which needs more researches in the future.

Human activities have greatly changed the pathway and rate of nitrogen transferred from atmosphere to terrestrial ecosystems worldwide. Anthropogenic nitrogen enrichment, as a result of chronic nitrogen deposition, has caused a wide range of impacts on ecosystem structures and functions. In this paper, we summarize the main effects of anthropogenic nitrogen addition on terrestrial ecosystems as follows:

1) Increased nitrogen input may have influences on plant production and ecosystem carbon storage, and the direction and magnitude of responses are determined by initial nitrogen status of ecosystems (N-limited or N-saturated) and local properties of vegetation and soil;

2) Chronic nitrogen addition can also alter soil nitrogen cycling, decrease the capability of soil to retain N, even lead to soil acidification, depletion of base cation, and affect decomposition of SOC;

3) Both high rate of nitrogen deposition and chronic low nitrogen addition may accelerate losses of N-contained gas, but the magnitude of influences depends on initial status of ecosystem (N-limited or P-limited);

4) N enrichment will affect species richness in terrestrial ecosystems, plant communities' structures and dynamics. It may also contribute to forest expansion into grasslands, alter species composition and diversity of mycorrhizal fungi;

5) Continuous N input and the resulting changes in plant composition and physiological feature may have effects on consumption rate and population dynamics of herbivorous insects, and eventually change ecosystem trophic structure through food chain;

6) Since the influences of N addition, increased concentration of CO₂ and O₃ on ecosystem properties and processes are interdependent, it is difficult to distinguish each effect.

We also summarized the current status of researches on N deposition in China, and proposed the potential research activities and recommendations.

Air pollution is a key factor that threatens earth's ecosystems and sustainability. Tropospheric ozone continues to be of major concern in the field of air pollution effects on terrestrial ecosystem productivity. In this paper, we have reviewed recent progresses in the studies of ozone effects on terrestrial production processes (including plant photosynthesis, carbon allocation, plant growth and crop yield), nutrient cycling, and species composition. Increased evidence shows that ozone can reduce the capability of photosynthesis by influencing leaf area and stomatal conductance, and alters the pattern of carbon allocation between shoot and root. Previous studies have also shown that elevated ozone could result in a maximum loss of 30% crop yield or forest production. However, little is known about how elevated ozone affects decomposition, nutrient cycling, species composition and trophic dynamics. There is an urgent need to investigate the synergetic effects of ozone pollution and climate change on terrestrial ecosystems. To address such a complex and large-scale environmental problem, spatially-explicit process-based ecosystem models have been used extensively in recent years. To advance our understanding of how air pollution and climate change influence terrestrial ecosystem productivity, further studies are needed to address multifactor experiments in the field and enhance the interaction between field studies and modeling.

In recent years, there has been increasing concern about the impacts of drought stress on terrestrial ecosystem productivity and the carbon cycle in the context of global change. In this paper, we have reviewed recent progresses in understanding how drought stress affects terrestrial ecosystem processes and how ecosystems adapt to increasing drought stress. Drought stress could cause terrestrial ecosystems to act as a carbon source to the atmosphere by decreasing terrestrial gross primary productivity. Drought stress also results in a reduction of both autotrophic and heterotrophic respiration. Drought often associates with high rates of fire intensity, plant mortality and disease, which could lead to a large reduction of terrestrial ecosystem productivity. However, plant and ecosystem respond to drought dress in a complex way. There are three adaptation strategies that plants can live with a drought condition: 1) some plants adjust their growing season to avoid drought stress; 2) some other plants modify their internal mechanism to counter drought stress; 3) the other plants hold some physiological properties to tolerate drought stress. Experimental and modeling investigations of how ecosystems respond to drought and associated stresses are clearly needed in the future research.

There are complex interactions among climatic change, fire disturbance, and ecosystem productivity. Fire disturbance could change the biogeochemical properties of soil and ecosystem structure, nutrient cycle and distribution as well as atmospheric composition and then influence ecosystem productivity; additionally, fire-induced emission can change the atmospheric composition and hence impacts the climate system; climate warming could alter the properties of fuel load and increase the frequency of fire-weather, and then in turn, influences regional fire regime.

In order to evaluate the impact of global change on ecosystem productivity comprehensively, it is essential to fully understand this complex interaction.

In this paper, we have reviewed the mechanisms of the interactions among global change, fire disturbance, and ecosystem productivity. Four critical issues have been explored: 1) methods used for constructing time sequence of fire information, 2) interaction between fire disturbance and global climate change, 3) interaction between fire disturbance and ecosystem productivity, and 4) changing fire management strategy in response to global change.

The comprehensive understanding of the complex interaction will be an important basis for further study how ecosystem pattern and process respond global change.

Global warming, resulted from rising atmospheric greenhouse gases, has increased the Earth's surface temperature by 0.6 ℃ in the 20th century and will continue to increase it by approximately 1.4-5.8 ℃ in this century. The delivery of reactive forms of nitrogen to the environment through the sum of agricultural and industrial activities also exceeds that from natural processes. The ecological consequences arising from global warming and atmospheric N deposition have also become the very important issues of global change research. These changes not only impact the growth of aboveground vegetation and plant community structure, but also change belowground soil environment, and thus indirectly influence the microbial processes. Recent researches suggest that soil microbial responses to climate warming and atmospheric N deposition play an important role in the feedbacks of terrestrial ecosystems to climate change. Better understanding on the microbial responses to increasing temperature and N deposition is critical to predict the changes in terrestrial ecosystem C, N dynamics in the future.

From the viewpoints of microbial biomass, microbial activities and structure, litter decomposition, nutrient use and cycling, sequestration, retention and loss of nutrients, this article reviews recent advances research on microbial responses to climate warming and atmospheric N deposition. Consequence, taken from most of researches, shows that soil microbial community structure is more sensitive than soil microbial biomass and microbial activities to reflect global climatic change. Although much progress in research on impacts of climate warming and atmospheric N deposition on microbe has achieved, yet there are some questions unresolved: 1) amounts of short-term research cannot be used to predict the long-term influence of these global change drivers on microbe; 2) few researches on microbial turnover and interaction in nutrient cycling; 3) interactive effects among different global change drivers to soil microbe. Therefore, more efforts should be taken to study the long-term influence of these global change drivers on microbe and analyze the change in microbial process with soil environmental changes, and more attention should be paid to the microbial responses to global change drivers in natural ecosystems in future researches. In the near future, consequence of soil microbial response on global change will still be the key question that we should answer urgently.

Enrichment of atmospheric greenhouse gases resulted from human activities such as fossil fuel burning and deforestation has increased global mean temperature by 0.6 ℃ in the 20th century and is predicted to increase it by 1.4-5.8 ℃ in this century. The unprecedented global warming will have profound long-term impacts on terrestrial plants and ecosystems. Responses of terrestrial plants and ecosystems to global warming may feed back to climate change via ecosystem and global carbon cycling. As one of the major methodology in global change research, ecosystem warming studies can facilitate model projections on potential changes in terrestrial biomes in terms of parameterization and validation. The difficulty in comparing and integrating the results from various ecosystems manipulated with different warming facilities exacerbated the uncertainties of model prediction. In this paper, field facilities in simulating climate warming and their potential applications in different terrestrial biomes were discussed. Critical scientific questions that could be addressed by ecosystem warming studies were proposed.

Aims We studied life history characteristics and spatial distribution of the Betula platyphylla population in Dongling Mountain to explain: 1) population growth characteristics, and 2) how can biological and environmental factors affect population distribution and dynamics?

Methods B. platyphylla is distributed mainly on the eastern slope of Dongling Mountain peak, Nangou and Goushicao in Longmen National Forest Park. We sampled trees in 41 quadrats (10 m×10 m), and, in each quadrat, we chose 2 grids (5 m×5 m) for shrubs and 4 grids (1 m×1 m) for grasses. The diameter at breast height (DBH) of each tree was recorded, and we divided them into nine DBH classes: DBH Ⅰ, DBH<2.5 cm;DBH Ⅱ, 2.5 cm≤ DBH<7.5 cm;DBH Ⅲ, 7.5 cm≤ DBH<12.5 cm, etc. We determined the life table, age structure and survivorship curve of theB. platyphylla population to analyze population structure and dynamics. We divided trees into three groups based on the nine classes of DBH in order to analyze spatial patterns for different age classes: young trees, DBH Ⅰ-Ⅱ; adult trees, DBH Ⅲ-Ⅴ; and old trees, DBH Ⅶ-Ⅸ.

Important findings There is a peak of mortality at a DBH of about 20 cm (DBH Ⅴ) caused by intraspecific competition for space and light resource. The population is in a stable growth period at DBH Ⅵ, after which mortality increases gradually. The age structure of the population indicates a declining population, suggesting that B. platyphylla is replaced by other species in succession. The survivorship curve of B. platyphylla population generally matches a Deevey Type Ⅰ, although seedlings are rare and have a low survival rate due to high crown density. The spatial distribution patterns of the population differ among plots and growth periods. The distribution pattern tends to be clumped on the eastern slope of Dongling Mountain peak where interspecific competition is not intense, but tends to be random in Nangou and Goushicao where interspecific competition is intense. The spatial distribution in different growth periods is clumped or random. Clustering is more pronounced for adult trees than old trees as a result of biological and environmental factors. Populations tend to clump when interspecific competition is not severe in a wet, cold, sunny habitat in Dongling Mountain. Intraspecific competition also affects the clustering intensity of different growth periods.

Aims Many researches have shown that the interlaced roots of dwarf bamboo (Fargesia nitida) can influence the regeneration of trees and the growth of seedlings and saplings through competition for light, water and nutrients. We explore the effect of dwarf bamboo on the regeneration of Abies faxoniana on the forest edge and estimate the effect of dwarf bamboo on the dispersal and community development. We asked: 1) how will quantitative characteristics and population structure change under different densities of F. nitida, and 2) how will the density of F. nitida influence the growth and biomass allocation of A. faxoniana seedlings?

Methods We sampled five belt transects and analyzed 30 plots within each transect in Wolong Giant Panda Nature Reserve (30°51'41″ N,102°58'21″ E) during August and September 2005. The transects were selected based on different densities of F. nitida (i.e., distance to the F. nitida cluster): 2 m outside the cluster (Zone 1), 1 m outside the cluster (Zone 2), 1 m inside the cluster (Zone 3), 2 m inside the cluster (Zone 4) and 3 m inside the cluster (Zone 5). We measured the height, basal diameter and crown of each A. faxoniana seedling. We used five BD classes: Ⅰ (0< BD<0.3 cm), Ⅱ (0.3 cm≤BD<0.6 cm), Ⅲ (0.6 cm≤BD<0.9 cm), Ⅳ (0.9 cm≤BD<1.2 cm), Ⅴ (BD≥1.2 cm), and five size classes: Ⅰ (0<H<15 cm), Ⅱ (15 cm≤H<30 cm), Ⅲ (30 cm≤H<45 cm), Ⅳ (45 cm≤H<60 cm), and Ⅴ (H≥60 cm). The seedlings were divided into two groups, large (H>0.2 m) and small (H≤0.2 m), to explore the accumulation and allocation of biomass byA. faxoniana seedlings. We obtained data on biomass from regression equations based on a sample of 60 A. faxoniana seedlings.

Important findings Regeneration and growth of A. faxoniana seedlings were restrained in the F. nitida-dominated zones, and dispersal of the A. faxoniana population and development of an A. faxoniana-dominated community were affected. Closer to F. nitida clusters, the number of old seedlings and the efficiency of transformation (seedling number ratio of the next BD class to the preceding BD class) were reduced (the older the A. faxoniana seedlings, the smaller the efficiency of transformation); the peak of number of seedlings moved from larger (Ⅲ) to the smallest size (Ⅰ); the total biomass and allocation of biomass aboveground ofA. faxoniana seedlings decreased; and the presence of F. nitida restrained the height growth of small (Ⅰ)A. faxoniana seedlings (confirmed by regression of basal diameter and height). The expansion of crowns of bigger (Ⅲ) seedlings increased in moderate density (Zone 4) ofF. nitida.

Aims Pinus densata is an important forest species in the high mountains of the southeastern Tibetan Plateau. Previous investigations demonstrated that this pine originated through natural hybridization between P. yunnanensis and P. tabulaeformis. The mechanisms underlying this hybrid speciation and especially its adaptive evolution are poorly understood. Reproductive fitness plays a critical role in hybrid speciation; however, the fitness of P. densata in the high plateau environment has not been investigated.

Methods We investigated 13 cone and seed characters, related to reproductive potential of the species, from six representative populations distributed throughout its natural range. The 13 characters are cone length, number of scales per cone, number of fertile scales, cone scale density, fertile scale density, ratio of fertile scales, number of seeds per cone, seed length, length of seed wing, total seed length per cone, total length of seed wing per cone, seed productivity per cone and ovule abortion rate. Patterns of variation of these characters were analyzed using one-way ANOVA and correlated to geo-ecological factors of each population.

Important fingdings Characters, such as cone length, total number of scales in a cone and number of seeds per cone in P. densata were similar to that in P. tabulaeformis, P. yunnanensis and several other species of Pinus. The maximum mean value of seed productivity per cone was 74%. One-way ANOVA showed significant (p<0.01) differentiation in all 13 characters among the six populations. Correlation analysis between cone and seed characters and geo-ecological factors indicated that total number of cone scales, number of fertile scales, number of seeds per cone and ratio of fertile scales were negatively correlated with latitude and seed productivity was positively correlated with longitude and ecological gradient axes. All results suggested thatP. densata as a hybrid species is not inferior in reproductive fitness in the plateau environment. The patterns of geographic variations in cone and seed characters seem to be related to the genetic background and divergent ecological environments of the populations.

Aims Decreasing root biomass might be an effective way to increase crop production under drought stress. A winter wheat with root excision was chosen to determine whether 1) root biomass or R/S (root/shoot ratio) is reduced by root excision, 2) competitive ability is reduced by root excision and its link with R/S and 3) competitive ability influences crop production.

Methods The effects of root excision on root-to-shoot relations, competition ability and yield characters of winter wheat (Triticum aestivum cv Changwu135) were investigated using a deWit replacement series in a pot study. Root excision was conducted from tillering to jointing. Two culture methods (mono and mixed) were compared. The effect of drought was estimated from two groups of plants, one group maintained at 75% soil FWC and a second at about 50%, after root excision. Soil water was controlled by weight daily. At grain maturity, plants for each pot were harvested, divided into shoot and root, dried and weighed.

Important findings The R/S was significantly reduced after root excision in monocultures irrespective of water condition. Furthermore, spike weight and aerial biomass of wheat after root excision were reduced in mixed cultures, indicating that the competitive ability of wheat was reduced after root excision. There was positive correlation between competitive ability and R/S in crop varieties, i.e., the crop with the higher R/S had higher competitive ability. In monocultures, spike weight and yield of wheat after root excision were reduced with a complete water supply. However, the spike weight and yield of wheat were higher after root excision than without excision under drought conditions. This study indicated that the crop with higher competitive ability has a higher production capability with a complete water supply and that the crop could obtain higher yield with the reduction in individual competitive ability under limited water.

Aims Carbon density, net production and carbon stock were estimated using data from natural spruce forests of Northwest Subalpine Sichuan.

Methods We harvested biomass, litter and soil carbon and calculated net production by dividing biomass by forest age.

Important findings The mean biomass of spruce forest is 230.37×10³ kg·hm^-2, with the tree layer accounting for 212.77×10³ kg·hm^-2 (92.36%). The percentage of carbon density in tree organs are stem 57.85%, bark 47.12%, branch 51.22%, leaf 48.27%, and root 52.39%. The percentage of carbon density in different strata are shrub 49.91%, herb 46.34%, duff 43.21%, litter 39.44%, and soil 1.41%. Carbon density declines with increased soil depth. The carbon stock is 273.79×10³ kg·hm^-2, divided among the tree layer with 109.30×10³ kg·hm^-2 (39.92%), shrub 5.69×10³ kg·hm^-2 (2.08%), herb 1.26×10³ kg·hm^-2 (0.46%), duff 0.60×10³ kg·hm^-2 (0.22%), litter 0.83×10³ kg·hm^-2 (0.30%), and soil (0-100 cm) 156.11×10³ kg·hm^-2 (57.01%). Therefore, the carbon stocks are ordered: soil (0-100 cm) > tree layer > shrub > herb > litter > duff. Mean net production is 6 838.5 kg·hm ^-2·a^-1, with the tree layer accounting for 4 676 kg·hm^-2·a^-1 (68.38%). Mean annual carbon sequestration is 3 584.98 kg·hm^-2·a^-1, with the tree layer accounting for 2 552.99 kg·hm^-2·a^-1 (71.21%).

Aims The objective of this paper was to describe the phytoplankton community and dynamics of Meixi Reservoir, a small oligotrophic reservoir in the southern subtopics of China.

Methods Hydrological parameters, nutrition and phytoplankton were investigated at the center of the reservoir every two weeks in spring from January to June 2005. Nutrient concentrations were measured by BG3838-2002. Phytoplankton were identified, counted and measured with microscopes, and their biomass was calculated based on cell morphometrics.

Important findings The phytoplankton community was characterized by low species number, biomass and dominance by two dinoflagellate species, Ceratium hirundinella and Peridinium sp. In total, 42 taxa were identified from 24 samples. The community was distinctively different between early spring (January to March) and late spring (April to June). Only about 12 species were observed in each sampling in the early spring, but 21 species in the late spring. The total abundance of phytoplankton ranged from 31 to 273 cells·ml^-1, and the total biomass of phytoplankton ranged from 0.176 to 2.024 mg·L^-1. The average biomass of phytoplankton was higher in late spring than in early spring. Dinoflagellates had an advantage in competition with other species and were able to move vertically to get nutrients from near the bottom of the reservoir. In late spring,phytoplankton abundance and biomass increased significantly with water temperature, but decreased with rainfall, apparently as water transparence declined due to disturbance from precipitation. Therefore, water temperature was the main factor restricting the phytoplankton community, but was disturbed by rainfall. This study provided base data for phytoplankton communities in oligotrophic water bodies and was helpful to understand possible changes with cultural eutrophication.

Aims Water consumption of single trees can be estimated by measuring the sap flow rate in trunk sapwood. Previous studies of sap flow had problems researching temporal correlations between environmental factors and sap flow because there is a lag time between environment and sap flow due to stomatal regulation in the leaf and water capacitance in the inner tissue of the trunk. Although this method has been used extensively for forest trees, it has not been reported for Quercus acutissima.

Methods We used a micro-meteorological station and thermal diffusion probes to measure daily meteorological factors such as total solar radiation (R_s), net solar radiation (TBB), air humidity (RH_a), air temperature (TP_a), wind speed (W_s), soil temperature (TPs), soil relative humidity (RHs) and the diurnal course of sap flow at lower (1.3 m), mid (4.5 m) and upper (8.0 m) heights of 40 a Quercus acutissima trunk in May 2005. The research site was on a south-facing hillside of Tai Mountain at the Forestry Centre, Forestry Science Academy of Taishan. Weather factors were sampled at 30 s intervals and recorded as 10 min averages. Sap flow velocity (SFV) was recorded by a Delta-T data logger at 10 min intervals. The temporal response of SFV to climate forcing factors was investigated using cross-correlation analysis over a range of time lags from -100 min to +180 min.

Important findings Patterns of daily and diurnal SFV fluctuation were different at the three trunk heights. SFV in upper trunk sapwood changed quickly and peaked >0.002 cm·s ^-1. SFV in the lower trunk changed slowly and was no more than 0.001 cm·s^-1. SFV in the mid-trunk was intermediate. The main environmental factors correlated with SFV were TBB, TP_a, RH_a, although their effects were not similar to each other (TPs and RHs were not significantly correlated to SFV). TBB showed the strongest (positive) correlation with SFV. TP_a and RH_a had weaker correlations: positive for TP_a and negative for RH_a. Correlations ranged from 0.265 to 0.944 for TBB versus SFV, from 0.409 to 0.869 for TP_a versus SFV and from -0.406 to -0.159 for RH_a versus SFV. The correlation of sap flow and environmental factors indicated that there were lags between SFV and TBB, TP_a and RH_a. Upper, mid and lower trunk lag times were about 80, 20 and 30 min, respectively, for SFV versus TBB, 60, 130 and 110 min for SFV versus TP_a and 170, 160 and 90 min for SFV versus RH_a.

Aims The objective of this study is to determine the allocation pattern of caloric values in dominant species of the secondary forests that developed from deforestation of tropical rain forests in Xishuangbanna, Southwest China.

Methods We studied three 20 m × 20 m plots in each of four communities: Trema orientalis forest, Mallotus paniculatus forest, Macaranga denticulata forest and Millettia leptobotrya forest, which were 2, 4, 6 and >15 years old. We recorded the species name and DBH of all trees with a diameter >3 cm. Caloric values of 17 dominant tree species were determined using five sample trees of each dominant species: one small tree, three intermediate and one large tree. The sampled parts were leaves, branches, stems and roots. The caloric values of three replications for each sample part were measured with a SDACM-IIIa oxygen bomb calorimeter, with an error less than 100 J·g ^-1. Differences were tested by t-tests.

Important findings The mean caloric values of T. orientalis, Mallotus paniculatus, Macarange denticulata and Millettia leptobotrya forests were 19 182.11, 19 474.81, 19 551.38 and 19 445.95 J·g^-1, respectively. Generally speaking, the caloric values of the climax tree species were greater than those of the pioneer. Differences between leaves were significant, but differences between branches, stems, roots and the average were not significant. The caloric values of different parts were ranked as: leaves > stems or branches > roots at the average level, although T. orientalis, Vitex quinata and Aporusa yunnanensis showed lower caloric value in leaves. Results suggest that there was an increase in the utilization efficiency of energy with aging of forests. Ecosystems develop by systematically increasing their ability to convert incoming solar energy; therefore, the transformation efficiency of energy was higher in climax trees than the pioneers. We postulated that in the early succession, ecosystems increase the absorption of energy though biomass accumulation; therefore, pioneer trees show lower caloric values. Along with build-up of organic structure, however, ecosystems augment the fixation of the energy quality, and then caloric values can be enhanced per unit weight.

Aims Fine root turnover is a major pathway for carbon and nutrient cycling in forest ecosystems, but our understanding of fine root turnover is limited, because fine root dynamic processes associated with soil resource availability and climate factors are poorly understood. The objectives of this study were to: 1) examine patterns of fine root production and mortality in different seasons and soil depths in Larix gmelinii and Fraxinus mandshurica plantations, 2) analyze correlation of fine root production and mortality with environment factors such as air temperature, precipitation, soil temperature and available nitrogen, and 3) estimate fine root turnover.

Methods We installed 36 minirhizotron tubes in six mono-specific plots of each species in September 2003 in Maoershan Experiment Forest Station. Minirhizotron sampling was conducted every two weeks from April 2004 to April 2005. We calculated average fine root length, annual fine root length production and mortality using the image data of minirhizotrons and estimated fine root turnover using three approaches.

Important findings The average growth rate and mortality rate in L. gmelinii was markedly smaller than in F. mandshurica, and rates were highest in the surface soil and lowest in the deepest of four soil layers. Annual fine root production and mortality in F. mandshurica were significantly higher than in L. gmelinii and were highest in the surface layer. Fine root production in spring and summer accounted for 41.7% and 39.7% of total annual production in F. mandshurica and 24.0% and 51.2% in L. gmelinii. The majority of fine root mortality was in spring and summer for F. mandshurica and summer and autumn for L. gmelinii. Turnover rate was 3.1 a^-1 for L. gmelinii and 2.7 a^-1 for F. mandshurica. Multiple regression analysis indicated that climate and soil resource factors together explained 80% variation of fine root seasonal growth and 95% of seasonal mortality. This study showed that fine root production and mortality of L. gmelinii and F. mandshurica had different patterns in different seasons and at different soil depths, and air temperature, precipitation, soil temperature and soil available nitrogen controlled the dynamics of fine root production, mortality and turnover in both species.