Mapping species assemblages of tropical forests at different hierarchical levels based on multivariate regression trees

Background: Vegetation distribution maps are of great significance for nature protection and management. In diverse tropical forests, accurate spatial mapping of vegetation types is challenging;the high species diversity and abundance of rare species challenge classification concepts, while remote sensing signals may not vary systematically with species composition, complicating the technical capability for delineating vegetation types in the landscape.Methods: We used a combination of field-based compositional data and their relations to environmental variables to predict the distribution of forest types in the Wuzhishan National Natural Reserve(WNNR), Hainan Island,China, using multivariate regression trees(MRT). The MRT was based on arboreal vegetation composition in 132plots of 20 m×20 m with a regular spacing of 1 km. Apart from the MRT, non-metric multidimensional scaling(NMDS) was used to evaluate vegetation-environment relationships.Results: The MRT model worked best when using 14 key environmental variables including topography, climate,latitude and soil, although the difference with the simpler model including only topographical variables was small. The full model classified the 132 plots into 3 vegetation types, 6 formation groups, 20 formations and 65associations at different hierarchical syntaxonomic levels. This model was the basis for forest vegetation maps for the WNNR. MRT and NMDS showed that elevation was the main driving force for the distribution of vegetation types and formation groups. Climate, latitude, and soil(especially available P), together with topographic variables, all influenced the distribution of formations and associations.Conclusions: While elevation determines forest-type distributions, lower-level syntaxonomic forest classes respond to the topographic diversity typical for mountains. Apart from providing the first detailed forest vegetation map for any part of WNNR, we show how, in spite of limitations, MRT with existing environmental data can be a useful method for mapping diverse and remote tropical forests.

Groundwater level prediction of landslide based on classification and regression tree

According to groundwater level monitoring data of Shuping landslide in the Three Gorges Reservoir area, based on the response relationship between influential factors such as rainfall and reservoir level and the change of groundwater level, the influential factors of groundwater level were selected. Then the classification and regression tree（CART） model was constructed by the subset and used to predict the groundwater level. Through the verification, the predictive results of the test sample were consistent with the actually measured values, and the mean absolute error and relative error is 0.28 m and 1.15%respectively. To compare the support vector machine（SVM） model constructed using the same set of factors, the mean absolute error and relative error of predicted results is 1.53 m and 6.11% respectively. It is indicated that CART model has not only better fitting and generalization ability, but also strong advantages in the analysis of landslide groundwater dynamic characteristics and the screening of important variables. It is an effective method for prediction of ground water level in landslides.

Artificial Neural Network （ANN） and Regression Tree （CART） applications for the indirect estimation of unsaturated soil shear strength parameters

The shear strength parameters of soil （cohesion and angle of internal friction） are quite essential in solving many civil engineering problems. In order to determine these parameters, laboratory tests are used. The main objective of this work is to evaluate the potential of Artificial Neural Network （ANN） and Regression Tree （CART） techniques for the indirect estimation of these parameters. Four different models, considering different combinations of 6 inputs, such as gravel %, sand %, silt %, clay %, dry density, and plasticity index, were investigated to evaluate the degree of their effects on the prediction of shear parameters. A performance evaluation was carried out using Correlation Coefficient and Root Mean Squared Error measures. It was observed that for the prediction of friction angle, the performance of both the techniques is about the same. However, for the prediction of cohesion, the ANN technique performs better than the CART technique. It was further observed that the model considering all of the 6 input soil parameters is the most appropriate model for the prediction of shear parameters. Also, connection weight and bias analyses of the best neural network （i.e., 6/2/2） were attempted using Connec- tion Weight, Garson, and proposed Weight-bias approaches to characterize the influence of input variables on shear strength parameters. It was observed that the Connection Weight Approach provides the best overall methodology for accurately quantifying variable importance, and should be favored over the other approaches examined in this study.

Sustainable Intensification and Large-scale Operation of Cultivated Land Use at the Farmers’ Scale:A Case Study of Shandong Province,China

Sustainable intensification of cultivated land use(SICLU) and large-scale operations(LSO) are widely acknowledged strategies for enhancing agricultural performance.However,the existing literature has faced challenges in precisely defining SICLU and constructing comprehensive indicators,which has hindered the exploration of factors influencing LSO within the SICLU framework.To address this gap,we integrated self-efficacy theory into the design of an index framework for evaluating SICLU.We subsequently employed econometric models to analyze the significant factors that impact LSO.Our findings reveal that SICLU can be divided into four key dimensions:intensive management,efficient output,resource conservation,and ecological environment optimization.Furthermore,it is crucial to incorporate belief-based cognitive factors into the index system,as farmers’ understanding of fertilizer and pesticide application significantly influences their willingness to engage in LSO.Moreover,we identify grain market turnover as the most influential factor in promoting LSO,with single-factor contribution rates reaching 70.9% for cultivated land transfer willingness and 62.5% for the total planting areas.Interestingly,unlike irrigation and agricultural machinery inputs,increased labor inputs correspond to larger planting areas for farmers.This trend may be attributed to reduced labor availability because of rural labor migration,whereas the reduction in irrigation and agricultural input is contingent on innovations in production practices and the transfer of cultivated land management rights.Importantly,SICLU dynamically influences LSO,with each index related to SICLU having an optimal range that fosters LSO.These insights offer valuable guidance for policymakers,emphasizing farmers as their central focus,with the adjustment of input and output factors as a means to achieve LSO as the ultimate goal.In conclusion,we propose research avenues for further enriching the SICLU framework to ensure that it aligns with the specific characteristics of regional agricultural development.

Measuring Causal Effect with ARDL-BART: A Macroeconomic Application

Modeling dynamic systems with linear parametric models usually suffer limitation which affects forecasting performance and policy implications. This paper advances a non-parametric autoregressive distributed lag model that employs a Bayesian additive regression tree (BART). The performance of the BART model is compared with selection models like Lasso, Elastic Net, and Bayesian networks in simulation experiments with linear and non-linear data generating processes (DGP), and on US macroeconomic time series data. The results show that the BART model is quite competitive against the linear parametric methods when the DGP is linear, and outperforms the competing methods when the DGP is non-linear. The empirical results suggest that the BART estimators are generally more efficient than the traditional linear methods when modeling and forecasting macroeconomic time series.

Simulation of reservoir outflows using regression tree and support vector machine

Water stored in reservoirs has a lot of crucial function,including generating hydropower,supporting water supply,and relieving lasting droughts.During floods,water deliveries from reservoirs must be acceptable,so as to guarantee that the gross volume of water is at a safe level and any release from reservoirs will not trigger flooding downstream.This study aims to develop a well-versed assessment method for managing reservoirs and pre-releasing water outflows by using the machine learning technology.As a new and exciting AI area,this technology is regarded as the most valuable,time-saving,supervised and cost-effective approach.In this study,two data-driven forecasting models,i.e.,Regression Tree(RT)and Support Vector Machine(SVM),were employed for approximately 30 years’hydrological records,so as to simulate reservoir outflows.The SVM and RT models were applied to the data,accurately predicting the fluctuations in the water outflows of a Bhakra reservoir.Different input combinations were used to determine the most effective release.For cross-validation,the number of folds varied.It is found that quadratic SVM for 10 folds with seven different parameters would give the minimum RMSE,maximum R2,and minimum MAE;therefore,it can be considered as the best model for the dataset used in this study.

Modelling the dead fuel moisture content in a grassland of Ergun City,China

The dead fuel moisture content(DFMC)is the key driver leading to fire occurrence.Accurately estimating the DFMC could help identify locations facing fire risks,prioritise areas for fire monitoring,and facilitate timely deployment of fire-suppression resources.In this study,the DFMC and environmental variables,including air temperature,relative humidity,wind speed,solar radiation,rainfall,atmospheric pressure,soil temperature,and soil humidity,were simultaneously measured in a grassland of Ergun City,Inner Mongolia Autonomous Region of China in 2021.We chose three regression models,i.e.,random forest(RF)model,extreme gradient boosting(XGB)model,and boosted regression tree(BRT)model,to model the seasonal DFMC according to the data collected.To ensure accuracy,we added time-lag variables of 3 d to the models.The results showed that the RF model had the best fitting effect with an R2value of 0.847 and a prediction accuracy with a mean absolute error score of 4.764%among the three models.The accuracies of the models in spring and autumn were higher than those in the other two seasons.In addition,different seasons had different key influencing factors,and the degree of influence of these factors on the DFMC changed with time lags.Moreover,time-lag variables within 44 h clearly improved the fitting effect and prediction accuracy,indicating that environmental conditions within approximately 48 h greatly influence the DFMC.This study highlights the importance of considering 48 h time-lagged variables when predicting the DFMC of grassland fuels and mapping grassland fire risks based on the DFMC to help locate high-priority areas for grassland fire monitoring and prevention.

Spatial pattern of impervious surfaces and their impacts on land surface temperature in Beijing,China

Land surface temperature （LST）, which is heavily influenced by urban surface structures, is a significant parameter in urban environmental analysis. This study examined the effect impervious surfaces （IS） spatial patterns have on LST in Beijing, China. A classification and regression tree model （CART） was adopted to estimate IS as a continuous variable using Landsat images from two seasons combined with QuickBird. LST was retrieved from the Landsat Thematic Mapper （TM） image to examine the relationships between IS and LST. The results revealed that CART was capable of consistently predicting LST with acceptable accuracy （correlation coefficient of 0.94 and the average error of 8.59%）. Spatial patterns of IS exhibited changing gradients across the various urban-rural transects, with LST values showing a concentric shape that increased as you moved from the outskirts towards the downtown areas. Transect analysis also indicated that the changes in both IS and LST patterns were similar at various resolution levels, which suggests a distinct linear relationship between them. Results of correlation analysis further showed that IS tended to be positively correlated with LST, and that the correlation coefficients increased from 0.807 to 0.925 with increases in IS pixel size. The findings identified in this study provide a theoretical basis for improving urban planning efforts to lessen urban temperatures and thus dampen urban heat island effects.

Development of ensemble learning models to evaluate the strength of coal-grout materials

In the loose and fractured coal seam with particularly low uniaxial compressive strength(UCS),driving a roadway is extremely difficult as roof falling and wall spalling occur frequently.To address this issue,the jet grouting(JG)technique(high-pressure grout mixed with coal particles)was first introduced in this study to improve the self-supporting ability of coal mass.To evaluate the strength of the jet-grouted coal-grout composite(JG composite),the UCS evolution patterns were analyzed by preparing 405 specimens combining the influential variables of grout types,curing time,and coal to grout(C/G)ratio.Furthermore,the relationships between UCS and these influencing variables were modeled using ensemble learning methods i.e.gradient boosted regression tree(GBRT)and random forest(RF)with their hyperparameters tuned by the particle swarm optimization(PSO).The results showed that the chemical grout composite has higher short-term strength,while the cement grout composite can achieve more stable strength in the long term.The PSO-GBRT and PSO-RF models can both achieve high prediction accuracy.Also,the variable importance analysis demonstrated that the grout type and curing time should be considered carefully.This study provides a robust intelligent model for predicting UCS of JG composites,which boosts JG design in the field.

Ladle Furnace Temperature Prediction Model Based on Large-scale Data With Random Forest

In ladle furnace, the prediction of the liquid steel temperature is always a hot topic for the researchers. The most of the existing temperature prediction models use small sample set. Today, the precision of them can not satisfy practical production. Fortunately, the large sample set is accumulated from the practical production process. However, a large sample set makes it difficult to build a liquid steel temperature model. To deal with the issue, the random forest method is preferred in this paper, which is a powerful regression method with low complexity and can be designed very quickly. It is with the parallel ensemble structure,uses sample subsets,and employs a simple learning algorithm of sub-models. Then, the random forest method is applied to establish a temperature model by using the data sampled from the production process. The experiments show that the random forest temperature model is more precise than other temperature models.

Optimization of environmental variables in habitat suitability modeling for mantis shrimp Oratosquilla oratoria in the Haizhou Bay and adjacent waters

Habitat suitability index(HSI)models have been widely used to analyze the relationship between species abundance and environmental factors,and ultimately inform management of marine species.The response of species abundance to each environmental variable is different and habitat requirements may change over life history stages and seasons.Therefore,it is necessary to determine the optimal combination of environmental variables in HSI modelling.In this study,generalized additive models(GAMs)were used to determine which environmental variables to be included in the HSI models.Significant variables were retained and weighted in the HSI model according to their relative contribution(%)to the total deviation explained by the boosted regression tree(BRT).The HSI models were applied to evaluate the habitat suitability of mantis shrimp Oratosquilla oratoria in the Haizhou Bay and adjacent areas in 2011 and 2013–2017.Ontogenetic and seasonal variations in HSI models of mantis shrimp were also examined.Among the four models(non-optimized model,BRT informed HSI model,GAM informed HSI model,and both BRT and GAM informed HSI model),both BRT and GAM informed HSI model showed the best performance.Four environmental variables(bottom temperature,depth,distance offshore and sediment type)were selected in the HSI models for four groups(spring-juvenile,spring-adult,falljuvenile and fall-adult)of mantis shrimp.The distribution of habitat suitability showed similar patterns between juveniles and adults,but obvious seasonal variations were observed.This study suggests that the process of optimizing environmental variables in HSI models improves the performance of HSI models,and this optimization strategy could be extended to other marine organisms to enhance the understanding of the habitat suitability of target species.

Spatial variation and prediction of forest biomass in a heterogeneous landscape

Large areas assessments of forest bioinass distribution are a challenge in heterogeneous landscapes, where variations in tree growth and species composition occur over short distances. In this study, we use statistical and geospatial modeling on densely sample.d forest biomass data to analyze the relative importance of ecological and physiographic variables as determinants of spatial variation of forest biomass in the environmentally heterogeneous region of the Big Sur, California. We estimated biomass in 280 forest Plots （one plot per 2.85 km2） and meas- ured an array of ecological （vegetation community type, distance to edge, amount of surrounding non-forest vegetation, soil properties, fire history） and physiographic drivers （elevation, potential soil moisture and solar radiation, proximity to the coast） of tree growth at each plot location. Our geostatistical analyses revealed that biomass distribution is spatially structured and autocorrelated up to 3.1 kin. Regression tree （RT） models showed that both physiographic and ecological factors influenced bio- mass distribution. Across randomly selected sample densities （sample size 112 to 280）, ecological effects of vegetation community type and distance to forest edge, and physiographic effects of elevation, potential soil moisture and solar radiation were the most consistent predictors of biomass. Topographic moisture index and potential solar radiation had apositive effect on biomass, indicating the importance of topographically- mediated energy and moisture on plant growth and biomass accumula- tion. RT model explained 35% of the variation in biomass and spatially autocorrelated variation were retained in regession residuals. Regression kriging model, developed from RT combined with kriging of regression residuals, was used to map biomass across the Big Sur. This study dem- onstrates how statistical and geospatial modeling can be used to dis- criminate the relative importance of physiographic and ecologic effects on forest biomass and develop spatial models to predict and map biomass distribution across a heterogeneous landscape.

Machine learning-based constitutive models for cement-grouted coal specimens under shearing

Cement-based grouting has been widely used in mining engineering;its constitutive law has not been comprehensively studied.In this study,a novel constitutive law of cement-grouted coal specimens(CGCS)was developed using hybrid machine learning(ML)algorithms.Shear tests were performed on CGCS for the analysis of stress-strain curves and the preparation of the dataset.To maintain the interpretation of the trained ML models,regression tree(RT)was used as the main technique.The effect of maximum RT depth(Maxdepth)on its performance was studied,and the hyperparameters of RT were tuned using the genetic algorithm(GA).The RT performance was also compared with ensemble learning techniques.The optimum correlation coefficient on the training set was determined as 0.835,0.946,0.981,and 0.985 for RT models with Maxdepth=3,5,7,and 9,respectively.The overall correlation coefficient was over 0.9 when the Maxdepth≥5,indicating that the constitutive law of CGCS can be well described.However,the failure type of CGCS could not be captured using the trained RT models.Random forest was found to be the optimum algorithm for the constitutive modeling of CGCS,while RT with the Maxdepth=3 performed the worst.

Variations in the biomass of Eucalyptus plantations at a regional scale in Southern China

We quantified deviations in regional forest biomass from simple extrapolation of plot data by the biomass expansion factor method（BEF） versus estimates obtained from a local biomass model,based on large-scale empirical field inventory sampling data.The sources and relative contributions of deviations between the two models were analyzed by the boosted regression trees method.Relative to the local model,BEF overestimated accumulative biomass by 22.12%.The predominant sources of the total deviation （70.94%） were stand-structure variables.Stand age and diameter at breast height are the major factors.Compared with biotic variables,abiotic variables had a smaller overall contribution （29.06%）,with elevation and soil depth being the most important among the examined abiotic factors.Large deviations in regional forest biomass and carbon stock estimates are likely to be obtained with BEF relative to estimates based on local data.To minimize deviations,stand age and elevation should be included in regional forest-biomass estimation.

Time Delay Identification in Dynamical Systems Based on Interpretable Machine Learning

The existence of time delay in complex industrial processes or dynamical systems is a common phenomenon and is a difficult problem to deal with in industrial control systems,as well as in the textile field.Accurate identification of the time delay can greatly improve the efficiency of the design of industrial process control systems.The time delay identification methods based on mathematical modeling require prior knowledge of the structural information of the model,especially for nonlinear systems.The neural network-based identification method can predict the time delay of the system,but cannot accurately obtain the specific parameters of the time delay.Benefit from the interpretability of machine learning,a novel method for delay identification based on an interpretable regression decision tree is proposed.Utilizing the self-explanatory analysis of the decision tree model,the parameters with the highest feature importance are obtained to identify the time delay of the system.Excellent results are gained by the simulation data of linear and nonlinear control systems,and the time delay of the systems can be accurately identified.

Patterns of forest composition and their long term environmental drivers in the tropical dry forest transition zone of southern Africa

Background： Tropical dry forests cover less than 13 % of the world＇s tropical forests and their area and biodiversity are declining. In southern Africa, the major threat is increasing population pressure, while drought caused by climate change is a potential threat in the drier transition zones to shrub land. Monitoring climate change impacts in these transition zones is difficult as there is inadequate information on forest composition to allow disentanglement from other environmental drivers. Methods： This study combined historical and modern forest inventories covering an area of 21,000 km2 in a transition zone in Namibia and Angola to distinguish late succession tree communities, to understand their dependence on site factors, and to detect trends in the forest composition over the last 40 years. Results： The woodlands were dominated by six tree species that represented 84 % of the total basal area and can be referred to as Bdikioea - Pterocarpus woodlands. A boosted regression tree analysis revealed that late succession tree communities are primarily determined by climate and topography. The Schinziophyton rautanenfi and Baikiaea plurijuga communities are common on slightly inclined dune or valley slopes and had the highest basal area （5.5 - 6.2 m^2 ha＆-1）. The Burkea africana - Guibourtia coleosperma and Pterocarpus angolensis - Diafium englerianum communities are typical for the sandy plateaux and have a higher proportion of smaller stems caused by a higher fire frequency. A decrease in overall basal area or a trend of increasing domination by the more drought and cold resilient B. africana community was not confirmed by the historical data, but there were significant decreases in basal area for Ochna pulchra and the valuable fruit tree D. englerianum. Conclusions： The slope communities are more sheltered from fire, frost and drought but are more susceptible to human expansion. The community with the important timber tree P. angolensis can best withstand high fire frequency but shows signs of a higher vulnerability to climate change. Conservation and climate adaptation strategies should include protection of the slope communities through refuges. Follow-up studies are needed on short term dynamics, especially near the edges of the transition zone towards shrub land.

Combining otolith elemental signatures with multivariate analytical models to verify the migratory pattern of Japanese Spanish mackerel(Scomberomorus niphonius) in the southern Yellow Sea

Japanese Spanish mackerel,Scomberomorus niphonius,is a commercially important,highly migratory species that is widely distributed throughout the northwestern Pacific region.However,its life history and migratory patterns are only partially understood.This study used otolith chemistry to investigate the migratory pattern of S.niphonius in the southern Yellow Sea,an important fishing ground.Transverse sections of otoliths from 15 age-1 spawning or spent individuals,comprising up to one complete migration cycle,were analyzed from the core to the margin by using laser ablation inductively coupled plasma mass spectrometry.The ratios of the element to Ca were integrated with microstructural analysis to produce age-related elemental profiles.Combining multielemental analysis of otolith composition with multivariate analytical models,we quantified structural changes in otolith chemistry profiles.Results revealed there were diverse changing patterns of otolith chemistry profiles for detected elements and the elements of Na,Mg,Sr and Ba were important for the chronological signal.Five clusters were identified through chronological clustering,representing the five life stages from the early stage to the spawning stage.Variation of Ba:Ca ratio was most informative,showing a step-decreasing pattern in the first four stages and a rebound in the spawning stage.These results support the hypothesized migratory pattern of S.niphonius:hatching and spending their early life in the coastal sandy ridges system of the southern Yellow Sea,migrating northeastward and offshore for feeding during juvenile stage,aggregating in early October and migrating outward to the Jeju Island for wintering,and returning to the coastal waters for spawning.This study demonstrated the value of life-history related otolith chemistry profiles combined with multivariate analytical models was a means to verify the migration patterns of S.niphonius at regional scales with potential application in fisheries assessment and management.

Forecasting Flowering and Maturity Times of Barley Using Six Machine Learning Algorithms

Interior Alaska has a short growing season of 110 d.The knowledge of timings of crop flowering and maturity will provide the information for the agricultural decision making.In this study,six machine learning algorithms,namely Linear Discriminant Analysis(LDA),Support Vector Machines(SVMs),k-nearest neighbor(kNN),Naïve Bayes(NB),Recursive Partitioning and Regression Trees(RPART),and Random Forest(RF),were selected to forecast the timings of barley flowering and maturity based on the Alaska Crop Datasets and climate data from 1991 to 2016 in Fairbanks,Alaska.Among 32 models fit to forecast flowering time,two from LDA,12 from SVMs,four from NB,three from RF outperformed models from other algorithms with the highest accuracy.Models from kNN performed worst to forecast flowering time.Among 32 models fit to forecast maturity time,two models from LDA outperformed the models from other algorithms.Models from kNN and RPART performed worst to forecast maturity time.Models from machine learning methods also provided a variable importance explanation.In this study,four out of six algorithms gave the same variable importance order.Sowing date was the most important variable to forecast flowering but less important variable to forecast maturity.The daily maximum temperature may be more important than daily minimum temperature to fit flowering models while daily minimum temperature may be more important than daily maximum temperature to fit maturity models.The results indicate that models from machine learning provide a promising technique in forecasting the timings of flowering and maturity of barley.

Delay recovery model for high-speed trains with compressed train dwell time and running time

Modeling the application of train operation adjustment actions to recover from delays is of great importance to supporting the decision-making of dispatchers.In this study,the effects of two train operation adjustment actions on train delay recovery were explored using train operation records from scheduled and actual train timetables.First,the modeling data were sorted to extract the possible influencing factors under two typical train operation adjustment actions,namely the compression of the train dwell time at stations and the compression of the train running time in sections.Stepwise regression methods were then employed to determine the importance of the influencing factors corresponding to the train delay recovery time,namely the delay time,the scheduled supplement time,the running interval,the occurrence time,and the place where the delay occurred,under the two train operation adjustment actions.Finally,the gradient-boosted regression tree(GBRT)algorithm was applied to construct a delay recovery model to predict the delay recovery effects of the train operation adjustment actions.A comparison of the prediction results of the GBRT model with those of a random forest model confirmed the better performance of the GBRT prediction model.

The Derivation of Nutrient Criteria for the Adjacent Waters of Yellow River Estuary in China

Ecological protection and high-quality development of the Yellow River basin are becoming part of the national strategy in recent years.The Yellow River Estuary has been seriously affected by human activities.Especially,it has been severely polluted by the nitrogen and phosphorus from land sources,which have caused serious eutrophication and harmful algal blooms.Nutrient criteria,however,was not developed for the Yellow River Estuary,which hindered nutrient management measures and eutrophication risk assessment in this key ecological function zone of China.Based on field data during 2004-2019,we adopted the frequency distribution method,correlation analysis,Linear Regression Model(LRM),Classification and Regression Tree(CART)and Nonparametric Changepoint Analysis(nCPA)methods to establish the nutrient criteria for the adjacent waters of Yellow River Estuary.The water quality criteria of dissolved inorganic nitrogen(DIN)and soluble reactive phosphorus(SRP)are recommended as 244.0μg L^(−1) and 22.4μg L^(−1),respectively.It is hoped that the results will provide scientific basis for the formulation of nutrient standards in this important estuary of China.