A Multimodel Ensemble-based Kalman Filter for the Retrieval of Soil Moisture Profiles

With the combination of three land surface models （LSMs） and the ensemble Kalman filter （EnKF）, a multimodel EnKF is proposed in which the multimodel background superensemble error covariance matrix is estimated by two different algorithms： the Simple Model Average （SMA） and the Weighted Average Method （WAM）. The two algorithms are tested and compared in terms of their abilities to retrieve the true soil moisture profile by respectively assimilating both synthetically-generated and actual near-surface soil moisture measurements. The results from the synthetic experiment show that the performances of the SMA and WAM algorithms were quite different. The SMA algorithm did not help to improve the estimates of soil moisture at the deep layers, although its performance was not the worst when compared with the results from the single-model EnKF. On the contrary, the results from the WAM algorithm were better than those from any single-model EnKF. The tested results from assimilating the field measurements show that the performance of the two multimodel EnKF algorithms was very stable compared with the single-model EnKF. Although comparisons could only be made at three shallow layers, on average, the performance of the WAM algorithm was still slightly better than that of the SMA algorithm. As a result, the WAM algorithm should be adopted to approximate the multimodel background superensemble error covariance and hence used to estimate soil moisture states at the relatively deep layers.

Prediction model of moisture content of dead fine fuel in forest plantations on Maoer Mountain,Northeast China

Preventing and suppressing forest fires is one of the main tasks of forestry agencies to reduce resource loss and requires a thorough understanding of the importance of factors affecting their occurrence.This study was carried out in forest plantations on Maoer Mountain in order to develop models for predicting the moisture content of dead fine fuel using meteorological and soil variables.Models by Nelson(Can J For Res 14:597-600,1984)and Van Wagner and Pickett(Can For Service 33,1985)describing the equilibrium moisture content as a function of relative humidity and temperature were evaluated.A random forest and generalized additive models were built to select the most important meteorological variables affecting fuel moisture content.Nelson’s(Can J For Res 14:597-600,1984)model was accurate for Pinus koraiensis,Pinus sylvestris,Larix gmelinii and mixed Larix gmelinii—Ulmus propinqua fuels.The random forest model showed that temperature and relative humidity were the most important factors affecting fuel moisture content.The generalized additive regression model showed that temperature,relative humidity and rain were the main drivers affecting fuel moisture content.In addition to the combined effects of temperature,rainfall and relative humidity,solar radiation or wind speed were also significant on some sites.In P.koraiensis and P.sylvestris plantations,where soil parameters were measured,rain,soil moisture and temperature were the main factors of fuel moisture content.The accuracies of the random forest model and generalized additive model were similar,however,the random forest model was more accurate but underestimated the effect of rain on fuel moisture.

Influence of Vegetation Cover on the Oh Soil Moisture Retrieval Model: A Case Study of the Malinda Wetland, Tanzania

Soil moisture is an important parameter that drives agriculture, climate and hydrological systems. In addition, retrieval of soil moisture is important in the analysis as well as its influence on these systems. Radar imagery is best suited for this retrieval due to its all-weather capability and independence from solar irradiation. Soil moisture retrieval was done for the Malinda Wetland, Tanzania, during two time periods, March and September 2013. The aim of this paper was to analyze soil moisture retrieval performance when vegetation contribution is taken into account. Backscatter values were obtained from TerraSAR-X Spotlight mode imagery taken in March and September 2013. The backscatter values recorded by SAR imagery are influenced by vegetation, soil roughness and soil moisture. Thus, in order to obtain the backscatter due to soil moisture, the roughness and vegetation contribution are determined and decoupled from total backscatter. The roughness parameters were obtained from a Digital Surface Model (DSM) from Unmanned Aerial Vehicle (UAV) photographs whereas the vegetation parameter was obtained by inverting the Water Cloud Model (WCM). Lastly, soil moisture was retrieved using the Oh Model. The coefficient of correlation between the observed and retrieved was 0.39 for the month of March and 0.65 in the month of August. When the vegetation contribution was considered, the r2 for March was 0.64 and that in August was 0.74. The results revealed that accounting for vegetation improved soil moisture retrieval.

Evaluating effects of Dielectric Models on the surface soil moisture retrieval in the Qinghai-Tibet Plateau

Based on the measurement of L-band ground-based microwave radiometer(ELBARA-III type)in the Qinghai-Tibet Plateau and theτ-ωradiative transfer model,this research evaluated the effects of four soil dielectric models,i.e.,Wang-Schmugge,Mironov,Dobson,and Four-phase,on the L-band microwave brightness temperature simulation and soil moisture retrieval.The results show that with the same vegetation and roughness parameterization scheme,the four soil dielectric models display obvious differences in microwave brightness temperature simulation.When the soil moisture is less than 0.23 m3/m3,the simulated microwave brightness temperature in Wang-Schmugge model is significantly different from that of the other three models,with maximum differences of horizontal polarization and vertical polarization reaching 8.0 K and 4.4 K,respectively;when the soil moisture is greater than 0.23 m3/m3,the simulated microwave brightness temperature of Four-phase significantly exceeds that of the other three models;when the soil moisture is saturated,maximum differences in simulated microwave brightness temperature with horizontal polarization and vertical polarization are 6.1 K and 4.8 K respectively,and the four soil dielectric models are more variable in the microwave brightness temperature simulation with horizontal polarization than that with vertical polarization.As for the soil moisture retrieval based on the four dielectric models,the comparison study shows that,under the condition of horizontal polarization,Wang-Schmugge model can reduce the degree of retrieved soil moisture underestimating the observed soil moisture more effectively than other parameterization schemes,while under the condition of vertical polarization,the Mironov model can reduce the degree of retrieved soil moisture overestimating the observed soil moisture.Finally,based on the Wang-Schmugge model and FengYun-3C observation data,the spatial distribution of soil moisture in the study area is retrieved.

Enhancing Deep Learning Soil Moisture Forecasting Models by Integrating Physics-based Models

Accurate soil moisture(SM)prediction is critical for understanding hydrological processes.Physics-based(PB)models exhibit large uncertainties in SM predictions arising from uncertain parameterizations and insufficient representation of land-surface processes.In addition to PB models,deep learning(DL)models have been widely used in SM predictions recently.However,few pure DL models have notably high success rates due to lacking physical information.Thus,we developed hybrid models to effectively integrate the outputs of PB models into DL models to improve SM predictions.To this end,we first developed a hybrid model based on the attention mechanism to take advantage of PB models at each forecast time scale(attention model).We further built an ensemble model that combined the advantages of different hybrid schemes(ensemble model).We utilized SM forecasts from the Global Forecast System to enhance the convolutional long short-term memory(ConvLSTM)model for 1–16 days of SM predictions.The performances of the proposed hybrid models were investigated and compared with two existing hybrid models.The results showed that the attention model could leverage benefits of PB models and achieved the best predictability of drought events among the different hybrid models.Moreover,the ensemble model performed best among all hybrid models at all forecast time scales and different soil conditions.It is highlighted that the ensemble model outperformed the pure DL model over 79.5%of in situ stations for 16-day predictions.These findings suggest that our proposed hybrid models can adequately exploit the benefits of PB model outputs to aid DL models in making SM predictions.

Forecast and Analysis of Soil Moisture Based on SIMPEL Model

The forecast of soil moisture lays foundation for water management in farmlands. The change of soil moisture is influenced by multiple meteorological fac- tors. It becomes much significant for improvement of agricultural production and ef- fective use of water to explore the rule of water dynamic at small scale, spatially or temporally. In the research, water dynamic in soil horizons at 0-40 cm in winter wheat belts was simulated by SIMPLE model as per water balance principle. Fur- thermore, ETp in fields was computed according to Haude method （DVWK stan- dards）; retained amount of water in fields and wilting coefficient were calculated based on soil parameters with SPAW （Soil-Plant-Air-Water）. The simulated results of SIMPLE model showed that the correlation of measured and simulated water con- tent in soils was 0.95 and relative error averaged lower than 3.1%, suggesting that the model would make a more precise estimation of water content in root zone in the area.

A Case Study of the Improvement of Soil Moisture Initialization in IAP-PSSCA

A prediction system is employed to investigate the potential use of a soil moisture initialization scheme in seasonal precipitation prediction through a case study of severe floods in 1998. The results show that driving the model with reasonable initial soil moisture distribution is helpful for precipitation prediction, and the initialization scheme is easy to use in operational prediction.

The Evolution of Organic Carbon in Chinese Mollisol Under Different Farming Systems: Validation and Prediction by Using Century Model

Soil organic carbon （SOC） is an important indicator of soil degradation process. In this study, the long-term SOC evolution in Chinese mollisol farmland was simulated and predicted by validating, analyzing, processing and assorting concerning data, based on clarifying parameters of Century model need, combined with best use of recorded data of field management, observed data of long-term experiments, climate, soil, and biology, and achieved results from Hailun Agro-Ecological Experimental Station, Chinese Academy of Sciences. The results were showed as follows： Before reclamation, SOC content was around 58.00 g kg^-1, SOC content dropped quickly in early years, and then decreased slowly after reclamation. SOC content was around 34.00 g kg^-1 with a yearly average rate of 8.91‰ decrease before long-term experiments was established. After a long-term experiment, SOC would change under different farming systems. Shift farming system changed as follows： By 20-year model simulation, SOC content decreased from 34.03 to 30.19 g kg^-1, with a yearly average rate of 5.97‰; by 100-year model simulation, SOC content decreased to 24.31 g kg^-1, with a yearly average rate of 3.36‰. Organic farming system changed as follows： By 20-year model simulation, SOC content decreased slowly from 34.03 to 33.39 g kg^-1, with a yearly average rate of 0.95‰, 5‰ less than that of shift farming system; by 100-year model simulation, SOC content decreased to 32.21 g kg^-1, with a yearly average rate of 0.55‰. ＂Petroleum＂ farming system changed as follows： By 20-year model simulation, SOC content decreased from 34.03 to 32.88 g kg^-1, with a yearly average rate of 1.72‰, much more than that of organic farming system; by 100-year model simulation, SOC content decreased to 30.89 g kg^-1, with a yearly average rate of 0.96‰. Combined ＂petroleum＂-organic farming system changed as follows： By 20-year model simulation, SOC content was increased slightly; by 100-year model simulation, SOC content increased from 34.03 to 34.41g kg^-1, with a yearly average rate of 0.11‰. The above results provided an optimal way for maintaining SOC in Chinese mollisol farmland： To increase, as much as possible within agro-ecosystem, soil organic matter returns such as crop stubble, crop litter, crop straw or stalk, and manure, besides applying chemical nitrogen and phosphorous, which increased system productivity and maintained SOC content as well. Also, the results provided a valuable methodology both for a study of CO2 sequestration capacity and for a target fertility determination in Chinese mollisol.

Evaluating the Capabilities of Soil Enthalpy, Soil Moisture and Soil Temperature in Predicting Seasonal Precipitation

Soil enthalpy （H） contains the combined effects of both soil moisture （w） and soil temperature （T） in the land surface hydrothermal process. In this study, the sensitivities of H to w and T are investigated using the multi-linear regression method. Results indicate that T generally makes positive contributions to H, while w exhibits different （positive or negative） impacts due to soil ice effects. For example, w negatively contributes to H if soil contains more ice; however, after soil ice melts, w exerts positive contributions. In particular, due to lower w interannual variabilities in the deep soil layer （i.e., the fifth layer）, H is more sensitive to T than to w. Moreover, to compare the potential capabilities of H, w and T in precipitation （P） prediction, the Huanghe-Huaihe Basin （HHB） and Southeast China （SEC）, with similar sensitivities of H to w and T, are selected. Analyses show that, despite similar spatial distributions of H-P and T-P correlation coefficients, the former values are always higher than the latter ones. Furthermore, H provides the most effective signals for P prediction over HHB and SEC, i.e., a significant leading correlation between May H and early summer （June） P. In summary, H, which integrates the effects of T and w as an independent variable, has greater capabilities in monitoring land surface heating and improving seasonal P prediction relative to individual land surface factors （e.g., T and w）.

Assimilation of ASAR Data with a Hydrologic and Semi-empirical Backscattering Coupled Model to Estimate Soil Moisture

The most promising approach for studying soil moisture is the assimilation of observation data and computational modeling. However, there is much uncertainty in the assimilation process, which affects the assimilation results. This research developed a one-dimensional soil moisture assimilation scheme based on the Ensemble Kalman Filter (EnKF) and Genetic Algorithm (GA). A two-dimensional hydrologic model-Distributed Hydrology-Soil-Vegetation Model (DHSVM) was coupled with a semi-empirical backscattering model (Oh). The Advanced Synthetic Aperture Radar (ASAR) data were assimilated with this coupled model and the field observation data were used to validate this scheme in the soil moisture assimilation experiment. In order to improve the assimilation results, a cost function was set up based on the distance between the simulated backscattering coefficient from the coupled model and the observed backscattering coefficient from ASAR. The EnKF and GA were used to re-initialize and re-parameterize the simulation process, respectively. The assimilation results were compared with the free-run simulations from hydrologic model and the field observation data. The results obtained indicate that this assimilation scheme is practical and it can improve the accuracy of soil moisture estimation significantly.

Probabilistic modeling of soil moisture dynamics in a revegetated desert area

Soil moisture is the key link between land hydrological and ecological processes which plays an important role in the terrestrial water cycle. As extreme weather events have increased in recent years, the stochastic simulation of soil moisture has gradually become the focus of ecohydrology research. Based on continuous monitoring of soil moisture data from 2008 to 2011, and histor- ical precipitation data from 199l to 2011, combined with the Rodriguez-Iturbe soil moisture dynamic stochastic model, soil mois- ture dynamics and its probability density fimction in a revegetated desert area was simulated. Results show that annual soil mois- ture dynamic changes of the revegetated desert area during the growing season complied with rainfall distribution; soil moisture probability presents a single-peak distribution in the plant rhizosphere layer （0-60 cm）. The peak width in the 20 cm topsoil was wider than in other soils, and the distribution presented the strong fluctuations and multiple aggregates. The peak widths of 40 cm and 60 cm soil moisture probability distribution were small, which are in accordance with simulated results of the Rodri- guez-lturbe model. This confrms that the Rodriguez-Imrbe model has good applicability and can well simulate the statistical characteristics of soil moisture in an arid revegetated desert area.

Resilience, reliability and risk analyses of maize, sorghum and sunflower in rain-fed systems using a soil moisture modeling approach

This paper is aimed at examining the applicability of methods for resilience, reliability and risk analyses of rain-fed agricultural systems from modeled continuous soil moisture availability in rain-fed crop lands. The methodology involves integration of soil and climatic data in a simple soil moisture accounting model to assess soil moisture availability, and a risk used as indicator of sustainability of rain-fed agricultural systems. It is also attempted to demonstrate the role of soil moisture modeling in risk analysis and agricultural water management in a semi-arid region in Limpopo Basin where rain-fed agriculture is practiced. For this purpose, a daily-time step soil moisture accounting model is employed to simulate daily soil moisture, evaporation, surface runoff, and deep percolation using 40 years (1961-2000) of agroclimatic data, and cropping cycle data of maize, sorghum and sunflower. Using a sustainability criterion on crop water requirement and soil moisture availability, we determined resilience, risk and reliability as a quantitative measure of sustainability of rain-fed agriculture of these three crops. These soil moisture simulations and the sustainability criteria revealed further confirmation of the relative sensitivity to drought of these crops. Generally it is found that the risk of failure is relatively low for sorghum and relatively high for maize and sunflower in the two sites with some differences of severity of failure owing to the slightly different agroclimatic settings.

Predicting and validating the load-settlement behavior of large-scale geosynthetic-reinforced soil abutments using hybrid intelligent modeling

Settlement prediction of geosynthetic-reinforced soil(GRS)abutments under service loading conditions is an arduous and challenging task for practicing geotechnical/civil engineers.Hence,in this paper,a novel hybrid artificial intelligence(AI)-based model was developed by the combination of artificial neural network(ANN)and Harris hawks’optimisation(HHO),that is,ANN-HHO,to predict the settlement of the GRS abutments.Five other robust intelligent models such as support vector regression(SVR),Gaussian process regression(GPR),relevance vector machine(RVM),sequential minimal optimisation regression(SMOR),and least-median square regression(LMSR)were constructed and compared to the ANN-HHO model.The predictive strength,relalibility and robustness of the model were evaluated based on rigorous statistical testing,ranking criteria,multi-criteria approach,uncertainity analysis and sensitivity analysis(SA).Moreover,the predictive veracity of the model was also substantiated against several large-scale independent experimental studies on GRS abutments reported in the scientific literature.The acquired findings demonstrated that the ANN-HHO model predicted the settlement of GRS abutments with reasonable accuracy and yielded superior performance in comparison to counterpart models.Therefore,it becomes one of predictive tools employed by geotechnical/civil engineers in preliminary decision-making when investigating the in-service performance of GRS abutments.Finally,the model has been converted into a simple mathematical formulation for easy hand calculations,and it is proved cost-effective and less time-consuming in comparison to experimental tests and numerical simulations.

‘Initial’ Soil Moisture Effects on the Climate in China——A Regional Climate Model Study

In this study,the effects of ‘initial’ soil moisture(SM) in arid and semi-arid Northwestern China on subsequent climate were investigated with a regional climate model. Besides the control simulations(denoted as CTL),a series of sensitivity experiments were conducted,including the DRY and WET experiments,in which the simulated ‘initial’ SM over the region 30 –50°N,75 –105°E was only 5% and 50%,and up to 150% and 200% of the simulated value in the CTL,respectively. The results show that SM change can modify the subsequent climate in not only the SM-change region proper but also the far downstream regions in Eastern and even Northeastern China. The SM-change effects are generally more prominent in the WET than in the DRY experiments. After the SM is initially increased,the SM in the SM-change region is always higher than that in the CTL,the latent(sensible) heat flux there increases(decreases),and the surface air temperature decreases. Spatially,the most prominent changes in the WET experiments are surface air temperature decrease,geopotential height decrease and corresponding abnormal changes of cyclonic wind vectors at the mid-upper troposphere levels. Generally opposite effects exist in the DRY experiments but with much weaker intensity. In addition,the differences between the results obtained from the two sets of sensitivity experiments and those of the CTL are not always consistent with the variation of the initial SM. Being different from the variation of temperature,the rainfall modifications caused by initial SM change are not so distinct and in fact they show some common features in the WET and DRY experiments. This might imply that SM is only one of the factors that impact the subsequent climate,and its effect is involved in complex processes within the atmosphere,which needs further investigation.

Surface Soil Moisture Simulation for a Typical Torrential Event with a Modified Noah LSM Coupling to the NWP Model

Surface soil moisture has great impact on both meso-and microscale atmospheric processes,especially on severe local convection processes and on the dynamics of short-lived torrential rains.To promote the performance of the land surface model (LSM) in surface soil moisture simulations,a hybrid hydrologic runoff parameterization scheme based upon the essential modeling theories of the Xin'anjiang model and Topography based hydrological Model (TOPMODEL) was developed in preference to the simple water balance model (SWB) in the Noah LSM.Using a strategy for coupling and integrating this modified Noah LSM to the Global/Regional Assimilation and Prediction System (GRAPES) analogous to that used with the standard Noah LSM,a simulation of atmosphere-land surface interactions for a torrential event during 2007 in Shandong was attempted.The results suggested that the surface,10-cm depth soil moisture simulated by GRAPES using the modified hydrologic approach agrees well with the observations.Improvements from the simulated results were found,especially over eastern Shandong.The simulated results,compared with the products of the Advanced Microwave Scanning Radiometer-Earth Observing System (AMSR-E) soil moisture datasets,indicated a consistent spatial pattern over all of China.The temporal variation of surface soil moisture was validated with the data at an observation station,also demonstrated that GRAPES with modified Noah LSM exhibits a more reasonable response to precipitation events,even though biases and systematic trends may still exist.

Spatio-Temporal Prediction of Root Zone Soil Moisture Using Multivariate Relevance Vector Machines

Root zone soil moisture at one and two meter depths are forecasted four days into the future. In this article, we propose a new multivariate output prediction approach to root zone soil moisture assessment using learning machine models. These models are known for their robustness, efficiency, and sparseness;they provide a statistically sound approach to solving the inverse problem and thus to building statistical models. The multivariate relevance vector machine (MVRVM) is used to build a model that forecasts soil moisture states based upon current soil moisture and soil temperature conditions. The methodology combines the data at different depths from 5 cm to 50 cm, the largest of which corresponds to the depth at which the soil moisture sensors are generally operational, to produce soil moisture predictions at larger depths. The MVRVM test results for soil moisture predictions at 1 m and 2 m depth on the 4th day are excellent with RMSE = 0.0131 m3/m3 for 1 m;and RMSE = 0.0015 m3/m3 for 2 m forecasted values. The statistics of predictions for 4th day (CoE = 0.87 for 1 m and CoE = 0.96 for 2 m) indicate good model generalization capability and computations show good agreement with actual measurements with R2 = 0.88 and R2 = 0.97 for 1 m and 2 m depths, respectively. The MVRVM produces good results for all four days. Bootstrapping is used to check over/under-fitting and uncertainty in model estimates.

Analysis of Variation Characters and Prediction Model of Soil Temperature in Solar Greenhouse

[Objective] The aim was to study the soil temperature changes and its forecast model in greenhouse by solar heat. [Method] Annual and daily variation characters of soil temperature were analyzed in this paper by using the observation data of air temperature out of solar greenhouse and different layers soil temperature in it. The soil temperature (daily maximum, daily minimum and daily mean) forecasting models were also studied. Simulation and test were conducted to the forecast model of soil temperature in the greenhouse. [Result] The annual changes and daily changes of soil temperature of each layer in the greenhouse were in single peak curve. The lower layer temperature changes were smaller than the upper layer. The soil temperature of each layer within the greenhouse was closely related to the relevance of same type temperature outside the greenhouse of the day. Taking the average daily temperature, daily maximum temperature and daily lowest temperature of the day and the day before as forecast factors, soil temperature forecast model of different layer of same type within greenhouse was constructed. The simulation outcome of average daily temperature of each layer within the greenhouse was better than the simulation outcome of highest temperature of corresponding layer, worse than the simulation of lowest temperature of corresponding layer. The highest temperature of lower soil and daily temperature of soil were better than the upper layer. The simulated soil temperature was much more close to the observation when the observation was during 15-30 ℃. In other interval, it was lower than the observation. [Conclusion] The study offered theoretical reference for the growth environment of sunlight greenhouse plantation.

On the Application of Probabilistic Hydrometeorological Simulation of Soil Moisture across Different Stations in India

An application of a proposed hydrometeorological approach for probabilistic simulation of soil moisture is carried out. The time series of in-situ soil moisture and meteorological variables at monthly scale from a few monitoring stations having different soil-hydrologic properties across India are utilized. Preliminary investigation with both precipitation and near-surface air-tempera- ture as meteorological variables to establish that the strength of association between soil moisture and precipitation is more significant as compared to that between soil moisture and temperature. Precipitation-based probabilistic estimation of soil moisture using the proposed hydrometeorological approach is tested with in-situ observed soil moisture, CPC model output and with soil moisture data of the Climate Change Initiative (CCI) project. The parameter of the developed model is linked to the soil-hydrologic characteristics through Hydrologic Soil Group (HSG) classification. Higher values of model parameter (dependence parameter (θ) for the selected copula) correspond to HSG A and B having higher soil porosity, whereas, lower values correspond to HSG B and C having lower soil porosity.

The Soil Moisture and Net Primary Production Affected by CO_2 and Climate Change Using a Coupled Model

In this paper, a coupled model was used to estimate the responses of soil moisture and net primary production of vegetation(NPP) to increasing atmospheric CO2 concentration and climate change. The analysis uses three experiments simulated by the second-generation Earth System Model(CanESM2) of the Canadian Centre for Climate Modelling and Analysis(CCCma), which are part of the phase 5 of the Coupled Model Intercomparison Project(CMIP5). The authors focus on the magnitude and evolution of responses in soil moisture and NPP using simulations modeled by CanESM, in which the individual effects of increasing CO2 concentration and climate change and their combined effect are separately accounted for. When considering only the single effect of climate change, the soil moisture and NPP have a linear trend of 0.03 kg m–2 yr–1 and –0.14 gC m–2 yr–2, respectively. However, such a reduction in the global NPP results from the decrease of NPP at lower latitudes and in the Southern Hemisphere, although increased NPP has been shown in high northern latitudes. The largest negative trend is located in the Amazon basin at –1.79 gC m–2 yr–2. For the individual effect of increasing CO2 concentration, both soil moisture and NPP show increases, with an elevated linear trend of 0.02 kg m–2 yr–1 and 0.84 gC m–2 yr–2, respectively. Most regions show an increasing NPP, except Alaska. For the combined effect of increasing atmospheric CO2 and climate change, the increased soil moisture and NPP exhibit a linear trend of 0.04 kg m–2 yr–1 and 0.83 gC m–2 yr–2 at a global scale. In the Amazon basin, the higher reduction in soil moisture is illustrated by the model, with a linear trend of –0.39 kg m–2 yr–1, for the combined effect. Such a change in soil moisture is caused by a weakened Walker circulation simulated by this coupled model, compared with the single effect of increasing CO2 concentration(experiment M2), and a consequence of the reduction in NPP is also shown in this area, with a linear trend of-0.16 gC m-2 yr-2.

Prediction Model of Soil Nutrients Loss Based on Artificial Neural Network

On the basis of Artificial Neural Network theory, a back propagation neural network with one middle layer is building in this paper, and its algorithms is also given, Using this BP network model, study the case of Malian-River basin. The results by calculating show that the solution based on BP algorithms are consis- tent with those based multiple - variables linear regression model. They also indicate that BP model in this paper is reasonable and BP algorithms are feasible.