The global terrestrial carbon cycle

There is great uncertainty with regard to the future role of the terrestrial biosphere in the global carbon cycle. The uncertainty arises from both an inadequate understanding of current pools and fluxes as well as the potential effects of rising atmospheric concentrations of CO₂ on natural ecosystems. Despite these limitations, a number of studies have estimated current and future patterns of terrestrial carbon storage. Future estimates focus on the effects of a climate change associated with a doubled atmospheric concentration of CO₂. Available models for examining the dynamics of terrestrial carbon storage and the potential role of forest management and landuse practices on carbon conservation and sequestration are discussed.     

The response of terrestrial c storage to climate change: Modeling C dynamics at varying temporal and spatial scales

The response of terrestrial C storage to GCM derived climate change scenarios was investigated over a range of temporal and spatial scales. The potential changes in the global distribution of major ecosystem complexes were examined by combining changes in land cover and corresponding soil type with C storage estimates for each of the ecosystem types. All scenarios consistently showed an overall increase in global carbon storage. On a more regional basis, the potential impacts of climate change on the structure, composition and biomass dynamics of major forest types within the North American Boreal zone were investigated using individual based stand models. Biomass fluxes were found to be dependent on the tree species, site and GCM parameters. A method to simulate corresponding changes in intra- and interannual patterns of CO₂ flux by combining a gap model with an ecosystem model which incorporates photosynthesis, respiration (both canopy and decomposer) and transpiration is demonstrated.     

Temporal dynamics of iron-rich, tropical soil organic carbon pools after land-use change from forest to sugarcane

Background, aim, and scope  

Land-use change can significantly influence carbon (C) storage and fluxes in terrestrial ecosystems. Soil–plant systems can act as sinks or sources of atmospheric CO₂ depending on formation and decomposition rates of soil organic matter. Therefore, changes in tropical soil C pools could have significant impacts on the global C cycle. This study aims to evaluate the impacts of long-term sugarcane cultivation on soil aggregation and organic matter, and to quantify temporal dynamics of soil organic matter in cultivated sugarcane plantation soils previously under a tropical natural secondary forest.     

Aspects of the interaction between vegetation and soil under global change

Responses of terrestrial ecosystems to a world undergoing a change in atmospheric CO₂ concentration presents a formidable challenge to terrestrial ecosystem scientists. Strong relationships among climate, atmosphere, soils and biota at many different temporal and spatial scales make the understanding and prediction of changes in net ecosystem production (NEP) at a global scale difficult. Global C cycle models have implicitly attempted to account for some of this complexity by adapting lower pool sizes and smaller flux rates representing large regions and long temporal averages than values appropriate for a small area. However, it is becoming increasingly evident that terrestrial ecosystems may be experiencing a strong transient forcing as a result of increasing levels of atmospheric CO₂ that will require a finer temporal and spatial representation of terrestrial systems than the parameters for current global C cycle models allow. To adequately represent terrestrial systems in the global C cycle it is necessary to explicitly model the response of terrestrial systems to primary environmental factors. While considerable progress has been made experimentally and conceptually in aspects of photosynthetic responses, and gross and net primary production, the application of this understanding to NEP at individual sites is not well developed. This is an essential step in determining effects of plant physiological responses on the global C cycle. We use a forest stand succession model to explore the effects of several possible plant responses to elevated atmospheric CO₂ concentration. These simulations show that ecosystem C storage can be increased by increases in individual tree growth rate, reduced transpiration, or increases in fine root production commensurate with experimental observations.     

Boreal forests and tundra

The circumpolar boreal biomes coverca. 2 10⁹ ha of the northern hemisphere and containca. 800 Pg C in biomass, detritus, soil, and peat C pools. Current estimates indicate that the biomes are presently a net C sink of 0.54 Pg C yr^?1. Biomass, detritus and soil of forest ecosystems (includingca. 419 Pg peat) containca. 709 Pg C and sequester an estimated 0.7 Pg C yr^?1. Tundra and polar regions store 60–100 Pg C and may recently have become a net source of 0.17 Pg C yr^?1. Forest product C pools, including landfill C derived from forest biomass, store less than 3 Pg C but increase by 0.06 Pg C yr^?1. The mechanisms responsible for the present boreal forest net sink are believed to be continuing responses to past changes in the environment, notably recovery from the little ice-age, changes in forest disturbance regimes, and in some regions, nutrient inputs from air pollution. Even in the absence of climate change, the C sink strength will likely be reduced and the biome could switch to a C source. The transient response of terrestrial C storage to climate change over the next century will likely be accompanied by large C exchanges with the atmosphere, although the long-term (equilibrium) changes in terrestrial C storage in future vegetation complexes remains uncertain. This transient response results from the interaction of many (often non-linear) processes whose impacts on future C cycles remain poorly quantified. Only a small part of the boreal biome is directly affected by forest management and options for mitigating climate change impacts on C storage are therefore limited but the potential for accelerating the atmospheric C release are high.     

Vegetation redistribution: A possible biosphere source of CO2 during climatic change

A new biogeographic model, MAPSS, predicts changes in vegetation leaf area index (LAI), site water balance and run off, as well as changes in Biome boundaries. Potential scenarios of equilibrium vegetation redistribution under 2 × CO₂ climate from five different General Circulation Models (GCMs) are presented. In general, large spatial shifts in temperate and boreal vegetation are predicted under the different scenarios; while, tropical vegetation boundaries are predicted (with one exception) to experience minor distribution contractions. Maps of predicted changes in forest LAI imply drought-induced losses of biomass over most forested regions, even in the tropics. Regional patterns of forest decline and dieback are surprisingly consistent among the five GCM scenarios, given the general lack of consistency in predicted changes in regional precipitation patterns. Two factors contribute to the consistency among the GCMs of the regional ecological impacts of climatic change: 1) regional, temperature-induced increases in potential evapotranspiration (PET) tend to more than offset regional increases in precipitation; and, 2) the unchanging background interplay between the general circulation and the continental margins and mountain ranges produces a fairly stable pattern of regionally specific sensitivity to climatic change. Two areas exhibiting among the greatest sensitivity to drought-induced forest decline are eastern North America and eastern Europe to western Russia. Drought-induced vegetation decline (losses of LAI), predicted under all GCM scenarios, will release CO₂ to the atmosphere; while, expansion of forests at high latitudes will sequester CO₂. The imbalance in these two rate processes could produce a large, transient pulse of CO₂ to the atmosphere.     

Natural sinks of CO2: Technical synthesis from the palmas Del Mar Workshop

Natural CO₂ sinks in terrestrial and marine environments are important components of the global carbon cycle, yet the sign and magnitudes of key fluxes among them are unknown. The results of the Palmas Del Mar Workshop — Natural Sinks of CO₂ presented in this special issue and its companion hardbound volume of Water, Air, & Soil Pollution, provide a synthesis of current research on the carbon cycle, CO₂ sinks and associated processes and fluxes, and critical research needs to assess the potential role of forest and land-use management in carbon sequestration. The papers in this volume present data, observations, and model simulations that demonstrate: 1) the existence of natural CO₂ sinks that could mitigate a significant amount of CO₂ emissions from fossilfuel combustion; 2) probable, human-caused imbalances in C exchanges among vegetation, soils, and the atmosphere; 3) enhanced C storage in vegetation in response to excess atmospheric CO₂; 4) strong interactions among carbon, nutrient and hydrological cycles; and 5) an excess of carbon production over consumption in several, large managed forests. Although it appears unlikely that the search for the “missing” C sink required to balance the C budget will end in the open ocean, new estimates of C storage in mangrove wood and peat, suggest that coastal ecosystems have the capacity to store significant amounts of carbon in vegetation and sediments. Convincing analyses are also presented indicating the technical and economical feasibility of managing existing lands to sequester additional carbon. Long-term field studies of CO₂ fertilization effects and carbon cycling by plants and soils in geographically important systems, native forests, and coastal ecosystems will go a long way toward meeting the research needs identified at the workshop.     

VEGETATION RESPONSE TO GLOBAL WARMING: THE ROLE OF HYSTERESIS EFFECT

A new approach is proposed for the evaluation of vegetation equilibrium response to global warming. The approach considers the dependence of the position of biome boundaries as significantly multi-valued function of climatic conditions; the reason for the multiplicity may be partly due to capacity of vegetation to change its environment. This result in hysteresis manifestations (threshold and irreversibility effects) in response to climate change. Matthews' global vegetation data set and IIASA climatic data base were used to reconstruct the domains of different biomes in a space of climatic factors (biotemperature and average precipitation). Based on the overlap of these domains, the maps of biomes' potential extent are calculated for present climate and for two scenarios of global warming (GISS and GFDL). These results imply a significant role for hysteresis phenomena in the global vegetation pattern. Maps of vegetation changes under two climate scenarios calculated with the help of a new algorithm to account for hysteresis indicate much less change than equivalent maps obtained by other equilibrium approaches under the two climate change scenarios. Changes are predicted for 20% of terrestrial area. A relatively small increase of forest and decrease of nonforest vegetation area predicted by both scenarios.     

Interactions between physical and biotic factors influence CO2 flux in Antarctic dry valley soils

Soil carbon dioxide (CO₂) flux is an integrative measure of ecosystem functioning representing both biotic and physical controls over carbon (C) balance. In the McMurdo Dry Valleys of Antarctica, soil CO₂ fluxes (approximately −0.1-0.15 μmol m⁻² s⁻¹) are generally low, and negative fluxes (uptake of CO₂) are sometimes observed. A combination of biological respiration and physical mechanisms, driven by temperature and mediated by soil moisture and mineralogy, determine CO₂ flux and, therefore, soil organic C balance. The physical factors important to CO₂ flux are being altered with climate variability in many ecosystems including arid forms such as the Antarctic terrestrial ecosystems, making it critical to understand how climate factors interact with biotic drivers to control soil CO₂ fluxes and C balances. We measured soil CO₂ flux in experimental field manipulations, microcosm incubations and across natural environmental gradients of soil moisture to estimate biotic soil respiration and abiotic sources of CO₂ flux in soils over a range of physical and biotic conditions. We determined that temperature fluctuations were the most important factor influencing diel variation in CO₂ flux. Variation within these diel CO₂ cycles was explained by differences in soil moisture. Increased temperature (as opposed to temperature fluctuations) had little or no effect on CO₂ flux if moisture was not also increased. We conclude that CO₂ flux in dry valley soils is driven primarily by physical factors such as soil temperature and moisture, indicating that future climate change may alter the dry valley soil C cycle. Negative CO₂ fluxes in arid soils have recently been identified as potential net C sinks. We demonstrate the potential for arid polar soils to take up CO₂, driven largely by abiotic factors associated with climate change. The low levels of CO₂ absorption into soils we observed may not constitute a significant sink of atmospheric CO₂, but will influence the interpretation of CO₂ flux for the dry valley soil C cycle and possibly other arid environments where biotic controls over C cycling are secondary to physical drivers.     

Temperature sensitivity increases with soil organic carbon recalcitrance along an elevational gradient in the Wuyi Mountains, China

No consensus exists regarding soil organic carbon (SOC) lability and the temperature sensitivity of its decomposition. This lack of clear understanding limits the accuracy in predicting the long-term impacts of climate change on soil carbon (C) storage. In this study, we determined the temperature responses of labile and recalcitrant organic carbon (LOC vs. ROC) by comparing the time required to decompose a given amount of C at different incubation temperatures along an elevational gradient in the Wuyi Mountains in southeastern China. Results showed that the temperature sensitivity increased with increasing SOC recalcitrance (Q_10-labile = 1.39 ± 0.04 vs. Q_{10-recalcitrant} = 3.94 ± 0.30). Q_10-labile and Q_{10-recalcitrant} values significantly increased with increasing soil depth. The effect of elevational vegetation change was significant for Q_{10-recalcitrant} but not for Q_10-labile, though they increased along the elevational gradient. The response of ROC pools to changes in temperature would accelerate the soil-stored C losses in the Wuyi Mountains. Kinetic theory suggested that SOC decomposition was both temperature- and quality-dependent due to an increased temperature. This would promote more CO₂ release from recalcitrant soil organic matter (SOM) in cold regions, resulting in a greater positive feedback to global climate change than previously expected. Moreover, the response of ROC to changes in temperature will determine the magnitude of the positive feedback due to its large storage in soils.     

Soil Carbon Dynamics Under Changing Climate——A Research Transition from Absolute to Relative Roles of Inorganic Nitrogen Pools and Associated Microbial Processes: A Review

It is globally accepted that soil carbon (C) dynamics are at the core of interlinked environmental problems,deteriorating soil quality and changing climate.Its management remains a complex enigma for the scientific community due to its intricate relationship with soil nitrogen (N) availability and moisture-temperature interactions.This article reviews the management aspects of soil C dynamics in light of recent advances,particularly in relation to the availability of inorganic N pools and associated microbial processes under changing climate.Globally,drastic alterations in soil C dynamics under changing land use and management practices have been primarily attributed to the variation in soil N availability,resulting in a higher decomposition rate and a considerable decline in soil organic C (SOC) levels due to increased soil CO2 emissions,degraded soil quality,and increased atmospheric CO2 concentrations,leading to climate warming.Predicted climate warming is proposed to enhance SOC decomposition,which may further increase soil N availability,leading to higher soil CO2 efflux.However,a literature survey revealed that soil may also act as a potential C sink,if we could manage soil inorganic N pools and link microbial processes properly.Studies also indicated that the relative,rather than the absolute,availability of inorganic N pools might be of key importance under changing climate,as these N pools are variably affected by moisture-temperature interactions,and they have variable impacts on SOC turnover.Therefore,multi-factorial studies are required to understand how the relative availability of inorganic N pools and associated microbial processes may determine SOC dynamics for improved soil C management.     

Forest responses to CO2 enrichment and climate warming

Two of the major uncertainties in forecasting future terrestrial sources and sinks of CO₂ are the CO₂-enhanced growth response of forests and soil warming effects on net CO₂ efflux from forests. Carbon dioxide enrichment of tree seedlings over time periods less than 1 yr has generally resulted in enhanced rates of photosynthesis, decreased respiration, and increased growth, with minor increases in leaf area and small changes in C allocation. Exposure of woody species to elevated CO₂ over several years has shown that high rates of photosynthesis may be sustained, but net C accumulation may not necessarily increase if CO₂ release from soil respiration increases. The impact of the 25% rise in atmospheric CO₂ with industrialization has been examined in tree ring chronologies from a range of species and locations. In contrast to the seedling tree results, there is no convincing evidence for CO₂-enhanced stem growth of mature trees during the last several decades. However, if mature trees show a preferential root growth response to CO₂ enrichment, the gain in root mass for an oak-hickory forest in eastern Tennessee is estimated to be only 9% over the last 40 years. Root data bases are inadequate for detecting such an effect. A very small shift in ecosystem nutrients from soil to vegetation could support CO₂-enhanced growth. Climate warming and the accompanying increase in mean soil temperature could have a greater effect than CO₂ enrichment on terrestrial sources and sinks of CO₂. Soil respiration and N mineralization have been shown to increase with soil temperature. If plant growth increases with increased N availability, and more C is fixed in growth than is released by soil respiration, then a negative feedback on climate warming will occur. If warming results in a net increase in CO₂ efflux from forests, then a positive feedback will follow. A 2 to 4°C increase in soil temperature could increase CO₂ efflux from soil by 15 to 32% in eastern deciduous forests. Quantifying C budget responses of forests to future global change scenarios will be speculative until mature tree responses to CO₂ enrichment and the effects of temperature on terrestrial sources and sinks of CO₂ can be determined.     

Can barley (Hordeum vulgare L. s.l.) adapt to fast climate changes? A controlled selection experiment

The projected future climate will affect the global agricultural production negatively, however, to keep abreast of the expected increase in global population, the agricultural production must increase. Therefore, to safeguard the future crop yield and quality, the adaptive potential of crops to environmental change needs to be explored in order to select the most productive genotypes. Presently, it is unknown whether cereal crops like spring barley can adapt to climate stressors over relatively few generations. To evaluate if strong selection pressures could change the performance of barley to environmental stress, we conducted a selection experiment over five plant generations (G0–G4) in three scenarios, where atmospheric [CO₂] and temperature were increased as single factors and in combination. The treatments represented the expected environmental characteristics in Northern Europe around year 2075 [700 ppm CO₂, 22/17 °C (day/night)] as well as a control mimicking present day conditions (390 ppm CO₂, 19/12 °C). Two different barley accessions, a modern cultivar and an old landrace, were evaluated in terms of yield and biomass production. In all treatments representing future environmental scenarios, the G4-generation of selected plants did not improve its reproductive output compared to the G0-generation, as G4 produced less seeds and had a lower yield than unselected plants. These results indicate that barley might not respond positively to rapid and strong selection by elevated [CO₂] and temperature, contrary to previous results from oilseed rape. The two barley accessions analyzed presented almost the same response pattern in a given treatment, though the modern cultivar had the highest yield in the climate scenarios, while the landrace was superior in yield under present day climate conditions.     

Global forest systems: An uncertain response to atmospheric pollutants and global climate change?

Forest systems cover more than 4.1×10⁹ ha of the Earth's land area. The future response and feedbacks of forest systems to atmospheric pollutants and projected climate change may be significant. Boreal, temperate and tropical forest systems play a prominent role in carbon (C), nitrogen (N) and sulfur (S) biogeochemical cycles at regional and global scales. The timing and magnitude of future changes in forest systems will depend on environmental factors such as a changing global climate, an accumulation of CO₂ in the atmosphere, and increase global mineralization of nutrients such as N and S. The interactive effects of all these factors on the world's forest regions are complex and not intuitively obvious and are likely to differ among geographic regions. Although the potential effects of some atmospheric pollutants on forest systems have been observed or simulated, large uncertainty exists in our ability to project future forest distribution, composition and productivity under transient or nontransient global climate change scenarios. The potential to manage and adapt forests to future global environmental conditions varies widely among nations. Mitigation practices, such as liming or fertilization to ameliorate excess NO_x or SO_x or forest management to sequester CO₂ are now being applied in selected nations worldwide.The U.S. Government's right to a non-exclusive, royalty free licence in and to any copyright is acknowledged.     

Quantifying feedback processes in the response of the terrestrial carbon cycle to global change: The modeling approach of image-2

The terrestrial biosphere component of the Integrated Model to Assess the Greenhouse Effect (IMAGE 2) uses changes in land cover to compute dynamically the C fluxes between the terrestrial biosphere and the atmosphere. The model explores the potential impact of feedback processes incorporated in the model, which are the enhancement of plant growth (CO₂ fertilization) and a more efficient use of water under increased CO₂ concentrations in the atmosphere; the temperature response of photosynthesis and respiration of plants; the temperature and soil water response of decomposition processes; and the climate-induced changes in vegetation and agricultural patterns with the consequent changes in land cover. In this paper we discuss the implementation and operation of the different feedback processes in the IMAGE 2 model. Results are shown for each process separately as well as the combined processes. The aim of this paper is to quantify the importance of these feedback processes geographically. The main results are that vegetation shifts due to climatic change and increased water use efficiency, CO₂ fertilization decreases net C emissions, while changed decomposition rates strongly increase C emissions to the atmosphere. Changes in the global balance between photosynthesis and respiration make little net difference. With the IPPC business-as-usual scenario the terrestrial biosphere continues to emit C into the atmosphere. This behavior is governed by changes in land-use, caused by a multitude of anthropogenic processes.     

Agricultural sources and sinks of carbon

Most existing agricultural lands have been in production for sufficiently long periods that C inputs and outputs are nearly balanced and they are neither a major source nor sink of atmospheric C. As population increases, food requirements and the need for more crop land increase accordingly. An annual conversion of previously uncultivated lands up to 1.5 × 10⁷ hectares may be expected. It is this new agricultural land which suffers the greatest losses of C during and subsequent to its conversion. The primary focus for analysis of future C fluxes in agroecosystems needs to be on current changes in land use and management as well as on direct effects of CO₂ and climate change. A valid assessment of C pools and fluxes in agroecosystems requires a global soils data base and comprehensive information on land use and management practices. A comprehensive effort to assemble and analyze this information is urgently needed.     

On the potential CO2 release from tundra soils in a changing climate

About 30% of the carbon in terrestrial ecosystems is stored in northern wetlands and boreal forest regions. Prevailing cold and wet soil conditions have largely been responsible for this carbon accumulation. It has been suggested that a warmer and drier climate in these regions might increase the decomposition rate and, hence, release more CO₂ to the atmosphere than at present. This study reports on the spatial variability and temperature dependence of the potential carbon release after incubating highly organic soils from the European Arctic and Siberia at different temperatures. We found that the decay potential, measured as CO₂ production in laboratory experiments, differed strongly within and among sites, particularly at higher soil temperatures. Furthermore, both the decay potential and its temperature response decreased significantly with depth in the soil, presumably because the older soils at deeper layers contained higher proportions of recalcitrant carbon than the younger soil organic matter at the surface. These results have implications for global models of potential feedbacks on climate change inferred from changes in the carbon balance of northern wetlands and tundra. Firstly, because the decay potential of the organic matter varies locally as well as regionally, predictions of how the tundra carbon balance may change will be unreliable if these are based on measurements at a few sites only. Secondly, any increase in CO₂ production may be transitional as both the carbon flux and its temperature sensitivity decrease when the most easily degradable organic material near the soil surface has decomposed. Consequently, it is crucial to account for transient responses and regional differences in the models of potential feedbacks on climate change from changed carbon cycling in northern terrestrial ecosystems.     

Dynamics of labile and recalcitrant soil carbon pools in a sorghum free-air CO2 enrichment (FACE) agroecosystem

Experimentation with dynamics of soil carbon pools as affected by elevated CO₂ can better define the ability of terrestrial ecosystems to sequester global carbon. In the present study, 6 N HCl hydrolysis and stable-carbon isotopic analysis (δ¹³C) were used to investigate labile and recalcitrant soil carbon pools and the translocation among these pools of sorghum residues isotopically labeled in the 1998-1999 Arizona Maricopa free air CO₂ enrichment (FACE) experiment, in which elevated CO₂ (FACE: 560 μmol mol⁻¹) and ambient CO₂ (Control: 360 μmol mol⁻¹) interact with water-adequate (wet) and water-deficient (dry) treatments. We found that on average 53% of the final soil organic carbon (SOC) in the FACE plot was in the recalcitrant carbon pool and 47% in the labile pool, whereas in the Control plot 46% and 54% of carbon were in recalcitrant and labile pools, respectively, indicating that elevated CO₂ transferred more SOC into the slow-decay carbon pool. Also, isotopic mixing models revealed that increased new sorghum residue input to the recalcitrant pool mainly accounts for this change, especially for the upper soil horizon (0-30 cm) where new carbon in recalcitrant soil pools of FACE wet and dry treatments was 1.7 and 2.8 times as large as that in respective Control recalcitrant pools. Similarly, old C in the recalcitrant pool under elevated CO₂ was higher than that under ambient CO₂, indicating that elevated CO₂ reduces the decay of the old C in recalcitrant pool. Mean residence time (MRT) of bulk soil carbon at the depth of 0-30 cm was significantly longer in FACE plot than Control plot by the averages of 12 and 13 yr under the dry and wet conditions, respectively. The MRT was positively correlated to the ratio of carbon content in the recalcitrant pool to total SOC and negatively correlated to the ratio of carbon content in the labile pool to total SOC. Influence of water alone on the bulk SOC or the labile and recalcitrant pools was not significant. However, water stress interacting with CO₂ enhanced the shift of the carbon from labile pool to recalcitrant pool. Our results imply that terrestrial agroecosystems may play a critical role in sequestrating atmospheric CO₂ and mitigating harmful CO₂ under future atmospheric conditions.     

Quantifying greenhouse gas emissions from soils: Scientific basis and modeling approach

Abstract

Global climate change is one of the most important issues of contemporary environmental safety. A scientific consensus is forming that the emissions of greenhouse gases, including carbon dioxide, nitrous oxide and methane, from anthropogenic activities may play a key role in elevating the global temperatures. Quantifying soil greenhouse gas emissions is an essential task for understanding the atmospheric impacts of anthropogenic activities in terrestrial ecosystems. In most soils, production or consumption of the three major greenhouse gases is regulated by interactions among soil redox potential, carbon source and electron acceptors. Two classical formulas, the Nernst equation and the Michaelis–Menten equation, describe the microorganism-mediated redox reactions from aspects of thermodynamics and reaction kinetics, respectively. The two equations are functions of a series of environmental factors (e.g. temperature, moisture, pH, Eh) that are regulated by a few ecological drivers, such as climate, soil properties, vegetation and anthropogenic activity. Given the complexity of greenhouse gas production in soils, process-based models are required to interpret, integrate and predict the intricate relationships among the gas emissions, the environmental factors and the ecological drivers. This paper reviews the scientific basis underlying the modeling of greenhouse gas emissions from terrestrial soils. A case study is reported to demonstrate how a biogeochemical model can be used to predict the impacts of alternative management practices on greenhouse gas emissions from rice paddies.