Local spatial modeling of white-tailed deer distribution

Complex spatial heterogeneity of ecological systems is difficult to capture and interpret using global models alone. For this reason, recent attention has been paid to local spatial modeling techniques. We used one local modeling approach, geographically weighted regression (GWR), to investigate the effects of local spatial heterogeneity on multivariate relationships of white-tailed deer distribution using land cover patch metrics and climate factors. The results of these analyses quantify differences in the contributions of model parameters to estimates of deer density over space. A GWR model with local kernel bandwidth was compared to a GWR model with global kernel bandwidth and an ordinary least-squares regression (OLS) model with the same parameters to evaluate their relative abilities in modeling deer distributions. The results indicated that the GWR models predicted deer density better than the traditional ordinary least-squares model and also provided useful information regarding local environmental processes affecting deer distribution. GWR model comparisons showed that the local kernel bandwidth GWR model was more realistic than the global kernel bandwidth GWR model, as the latter exaggerated local spatial variation. The parameter estimates and model statistics (e.g., model R²) of the GWR models were mapped using geographic information systems (GIS) to illustrate local spatial variation in the regression relationship and to identify causes of large-scale model misspecifications and low estimation efficiencies.     

The effect of species response form on species distribution model prediction and inference

Ecological theory and current evidence support the validity of various species response curves according to a variety of environmental gradients. Various methods have been developed for building species distribution models but it is not well known how these methods perform under various assumptions about the form of the underlying species response. It is also not well known how spatial correlation in species occurrence affects model performance. These effects were investigated by applying an environmental envelope method (BIOCLIM) and three regression-based methods: logistic regression (LR), generalized additive modelling (GAM), and classification and regression tree (CART) to simulated species occurrence data. Each simulated species was constructed as a sum of responses with varying weights. Three basic species response curves were assumed: Gaussian (bell-shaped), Beta (skew) and linear. The two non-linear responses conform to standard ecological niche theory. All three responses were applied in turn to three simulated environmental variables, each with varying degrees of spatial autocorrelation. GAM produced the most consistent model performance over all forms of simulated species response. BIOCLIM and CART were inclined to underrate the performance of variables with a linear response. BIOCLIM was less sensitive to data density. LR was susceptible to model misspecification. The use of a linear function in LR underestimated the performance of variables with non-linear species response and contributed to increased spatial autocorrelation in model residuals. Omission of important environmental variables with non-linear species response also contributed to increased spatial autocorrelation in model residuals. Adding a spatial autocovariate term to the LR model (autologistic model) reduced the spatial autocorrelation and improved model performance, but did not correct the misidentification of the dominant environmental determinant. This is to be expected since the autologistic approach was designed primarily for prediction and not for inference. Given that various forms of species response to environmental determinants arise commonly in nature: (1) higher order functions should always be tested when applying LR in modelling species distribution; (2) spatial autocorrelation in species distribution model residuals can indicate that environmental determinants with non-linear response are missing from the model; and (3) deficiencies in LR model performance due to model misspecification can be addressed by adding a spatial autocovariate to the model, but care should be taken when interpreting the coefficients of the model parameters.     

Predictability of marine nematode biodiversity

In this paper, we investigated: (1) the predictability of different aspects of biodiversity, (2) the effect of spatial autocorrelation on the predictability and (3) the environmental variables affecting the biodiversity of free-living marine nematodes on the Belgian Continental Shelf. An extensive historical database of free-living marine nematodes was employed to model different aspects of biodiversity: species richness, evenness, and taxonomic diversity. Artificial neural networks (ANNs), often considered as “black boxes”, were applied as a modeling tool. Three methods were used to reveal these “black boxes” and to identify the contributions of each environmental variable to the diversity indices. Since spatial autocorrelation is known to introduce bias in spatial analyses, Moran's I was used to test the spatial dependency of the diversity indices and the residuals of the model. The best predictions were made for evenness. Although species richness was quite accurately predicted as well, the residuals indicated a lack of performance of the model. Pure taxonomic diversity shows high spatial variability and is difficult to model. The biodiversity indices show a strong spatial dependency, opposed to the residuals of the models, indicating that the environmental variables explain the spatial variability of the diversity indices adequately. The most important environmental variables structuring evenness are clay and sand fraction, and the minimum annual total suspended matter. Species richness is also affected by the intensity of sand extraction and the amount of gravel of the sea bed.     

Robust ridge regression for estimating the effects of correlated gene expressions on phenotypic traits

Statistical packages such as edgeR and DESeq are intended to detect genes that are relevant to phenotypic traits and diseases. A few studies have also modeled the relationships between gene expressions and traits. In the presence of multicollinearity and outliers, which are unavoidable in genetic data, the robust ridge regression estimator can be applied with the trait value as the response variable and the gene expressions as explanatory variables. In some simulation scenarios, the robust ridge estimator is resistant to outliers and less susceptible to multicollinearity than the ordinary least-squares (OLS) estimator. This study investigated the reliability of the robust ridge estimator, in a scenario where the explanatory variables have tail-dependence and negative binomial distributions, by comparing its performance to that of OLS using vine copula to model the tail-dependence among gene expressions. The robust ridge estimator and OLS were both applied to an ecological dataset. First, statistical analysis was used to compare RNA sequencing data between two treatments; then, 15 differentially expressed genes were selected. Next, the regression parameter estimates of robust ridge and OLS for the effects of the 15 contigs (explanatory variables) on trait values (response variables) were compared. Robust ridge regression was found to detect fewer positive and negative slopes than OLS regression. These results indicate that robust ridge regression can be successfully applied for RNA sequencing analysis to estimate the effect of trait-associated genes using real data, and holds great promise as a tool for modeling the association between RNA expression and phenotypic traits.     

Predictive vegetation modeling for conservation: impact of error propagation from digital elevation data.

The effect of digital elevation model (DEM) error on environmental variables, and subsequently on predictive habitat models, has not been explored. Based on an error analysis of a DEM, multiple error realizations of the DEM were created and used to develop both direct and indirect environmental variables for input to predictive habitat models. The study explores the effects of DEM error and the resultant uncertainty of results on typical steps in the modeling procedure for prediction of vegetation species presence/absence. Results indicate that all of these steps and results, including the statistical significance of environmental variables, shapes of species response curves in generalized additive models (GAMs), stepwise model selection, coefficients and standard errors for generalized linear models (GLMs), prediction accuracy (Cohen's kappa and AUC), and spatial extent of predictions, were greatly affected by this type of error. Error in the DEM can affect the reliability of interpretations of model results and level of accuracy in predictions, as well as the spatial extent of the predictions. We suggest that the sensitivity of DEM-derived environmental variables to error in the DEM should be considered before including them in the modeling processes.     

Species distribution models and ecological theory: A critical assessment and some possible new approaches

Given the importance of knowledge of species distribution for conservation and climate change management, continuous and progressive evaluation of the statistical models predicting species distributions is necessary. Current models are evaluated in terms of ecological theory used, the data model accepted and the statistical methods applied. Focus is restricted to Generalised Linear Models (GLM) and Generalised Additive Models (GAM). Certain currently unused regression methods are reviewed for their possible application to species modelling.A review of recent papers suggests that ecological theory is rarely explicitly considered. Current theory and results support species responses to environmental variables to be unimodal and often skewed though process-based theory is often lacking. Many studies fail to test for unimodal or skewed responses and straight-line relationships are often fitted without justification.Data resolution (size of sampling unit) determines the nature of the environmental niche models that can be fitted. A synthesis of differing ecophysiological ideas and the use of biophysical processes models could improve the selection of predictor variables. A better conceptual framework is needed for selecting variables.Comparison of statistical methods is difficult. Predictive success is insufficient and a test of ecological realism is also needed. Evaluation of methods needs artificial data, as there is no knowledge about the true relationships between variables for field data. However, use of artificial data is limited by lack of comprehensive theory.Three potentially new methods are reviewed. Quantile regression (QR) has potential and a strong theoretical justification in Liebig's law of the minimum. Structural equation modelling (SEM) has an appealing conceptual framework for testing causality but has problems with curvilinear relationships. Geographically weighted regression (GWR) intended to examine spatial non-stationarity of ecological processes requires further evaluation before being used.Synthesis and applications: explicit theory needs to be incorporated into species response models used in conservation. For example, testing for unimodal skewed responses should be a routine procedure. Clear statements of the ecological theory used, the nature of the data model and sufficient details of the statistical method are needed for current models to be evaluated. New statistical methods need to be evaluated for compatibility with ecological theory before use in applied ecology. Some recent work with artificial data suggests the combination of ecological knowledge and statistical skill is more important than the precise statistical method used. The potential exists for a synthesis of current species modelling approaches based on their differing ecological insights not their methodology.     

Comparative performance of generalized additive models and multivariate adaptive regression splines for statistical modelling of species distributions

Two statistical modelling techniques, generalized additive models (GAM) and multivariate adaptive regression splines (MARS), were used to analyse relationships between the distributions of 15 freshwater fish species and their environment. GAM and MARS models were fitted individually for each species, and a MARS multiresponse model was fitted in which the distributions of all species were analysed simultaneously. Model performance was evaluated using changes in deviance in the fitted models and the area under the receiver operating characteristic curve (ROC), calculated using a bootstrap assessment procedure that simulates predictive performance for independent data. Results indicate little difference between the performance of GAM and MARS models, even when MARS models included interaction terms between predictor variables. Results from MARS models are much more easily incorporated into other analyses than those from GAM models. The strong performance of a MARS multiresponse model, particularly for species of low prevalence, suggests that it may have distinct advantages for the analysis of large datasets. Its identification of a parsimonious set of environmental correlates of community composition, coupled with its ability to robustly model species distributions in relation to those variables, can be seen as converging strongly with the purposes of traditional ordination techniques.     

Comparison and ranking of different modelling techniques for prediction of site index in Mediterranean mountain forests

Forestry science has a long tradition of studying the relationship between stand productivity and abiotic and biotic site characteristics, such as climate, topography, soil and vegetation. Many of the early site quality modelling studies related site index to environmental variables using basic statistical methods such as linear regression. Because most ecological variables show a typical non-linear course and a non-constant variance distribution, a large fraction of the variation remained unexplained by these linear models. More recently, the development of more advanced non-parametric and machine learning methods provided opportunities to overcome these limitations. Nevertheless, these methods also have drawbacks. Due to their increasing complexity they are not only more difficult to implement and interpret, but also more vulnerable to overfitting. Especially in a context of regionalisation, this may prove to be problematic. Although many non-parametric and machine learning methods are increasingly used in applications related to forest site quality assessment, their predictive performance has only been assessed for a limited number of methods and ecosystems.In this study, five different modelling techniques are compared and evaluated, i.e. multiple linear regression (MLR), classification and regression trees (CART), boosted regression trees (BRT), generalized additive models (GAM), and artificial neural networks (ANN). Each method is used to model site index of homogeneous stands of three important tree species of the Taurus Mountains (Turkey): Pinus brutia, Pinus nigra and Cedrus libani. Site index is related to soil, vegetation and topographical variables, which are available for 167 sample plots covering all important environmental gradients in the research area. The five techniques are compared in a multi-criteria decision analysis in which different model performance measures, ecological interpretability and user-friendliness are considered as criteria.When combining these criteria, in most cases GAM is found to outperform all other techniques for modelling site index for the three species. BRT is a good alternative in case the ecological interpretability of the technique is of higher importance. When user-friendliness is more important MLR and CART are the preferred alternatives. Despite its good predictive performance, ANN is penalized for its complex, non-transparent models and big training effort.     

基于GWR的中国地级城市SO2年均质量浓度模拟

中国城市空气污染问题已经引起广泛关注。目前相关研究很多,但是以空间位置为拟合参数,对空气质量进行回归模拟的研究较少。以2010年中国地级以上城市SO2年均质量浓度为因变量,分别应用普通线性回归和地理加权回归（GWR）模型模拟SO2年均质量浓度,其中地理加权回归方法考虑了空间位置的影响并以此作为回归参数。回归的自变量指标体系包括气象要素（多年平均温度、光照、降水）、植被覆盖（NDVI）、地形要素（坡度、坡向、起伏度）、人为因素（GDP、能源消费）几个方面。由于各指标之间存在较强的相关性,用主成分分析方法计算得到温度、日照、降水、NDVI表征的气象植被综合指标,高程、坡度、起伏度表征的地形综合指标,和GDP、能源消费表征的人为因素综合指标。用3个综合指标值作为自变量进行回归模拟。普通回归结果较差,其r^2为0.11,矫正的r^2为0.10;GWR模型模拟结果相对较好,其拟合优度显著提高,r^2为0.66,矫正的r^2为0.47。因此,地理加权回归适合进行此类拟合,普通线性回归不适合。通过对比地理加权回归模拟的各个城市的拟合优度,发现年均质量浓度数值较高的地区拟合效果较差,这些地区主要集中在中国华北和南部部分地区。与基于机理的模型相比,GWR 模型和其各具优缺点,GWR 的优势主要表现在数据及其格式化要求低,计算机软硬件条件要求低,运算速度快等。     

New approaches to modelling fish-habitat relationships

Ecologists often develop models that describe the relationship between faunal communities and their habitat. Coral reef fishes have been the focus of numerous such studies, which have used a wide range of statistical tools to answer an equally wide range of questions. Here, we apply a series of both conventional statistical techniques (linear and generalized additive regression models) and novel machine-learning techniques (the support vector machine and three ensemble techniques used with regression trees) to predict fish species richness, biomass, and diversity from a range of habitat variables. We compare the techniques in terms of their predictive performance, and we compare a subset of the models in terms of the influence each habitat variable has for the predictions. Prediction errors are estimated by cross-validation, and variable importance is assessed using permutations of individual variable values. For predictions of species richness and diversity the tree-based models generally and the random forest model specifically are superior (produce the lowest errors). These model types are all able to model both nonlinear and interaction effects. The linear model, unable to model either effect type, performs the worst (produces the highest errors). For predictions of biomass, the generalized additive model is superior, and the support vector machine performs the worst. Depth range, the difference between maximum and minimum water depth at a given site, is identified as the most important variable in the majority of models predicting the three fish community variables. However, variable importance is highly dependent upon model type, which leads to questions regarding the interpretation of variable importance and its proper use as an indicator of causality. The representation of ecological relationships by tree-based ensemble learners will improve predictive performance, and provide a new avenue for exploring ecological relationships, both statistical and causal.     

The evaluation strip: A new and robust method for plotting predicted responses from species distribution models

Increasing use is being made in conservation management of statistical models that couple extensive collections of species and environmental data to make predictions of the geographic distributions of species. While the relationships fitted between a species and its environment are relatively transparent for many of these modeling techniques, others are more ‘black box’ in character, only producing geographic predictions and providing minimal or untraditional summaries of the fitted relationships on which these predictions are based. This in turn prevents robust evaluation of the ecological sensibility of such models, a necessary process if model predictions are to be treated with confidence. Here we propose a new but simple method for visualizing modeled responses that can be implemented with any modeling method, and demonstrate its application using five common methods applied to the prediction of an Australian tree species. This is achieved by insetting an “evaluation strip” into the spatial data layers, which, after predictions have been made, can be clipped out and used for creating plots of the modelled responses. We present findings of the application strip for algorithms GLMs, GAMs, CLIM, DOMAIN and MARS. Evaluation strips can be constructed to investigate either uni-variate responses, or the simultaneous variation in predicted values in relation to two variables. The latter option is particularly useful for evaluating responses in models that allow the fitting of complex interaction terms.     

Environmental filtering and land-use history drive patterns in biomass accumulation in a mediterranean-type landscape

Aboveground biomass (AGB) reflects multiple and often undetermined ecological and land-use processes, yet detailed landscape-level studies of AGB are uncommon due to the difficulty in making consistent measurements at ecologically relevant scales. Working in a protected mediterranean-type landscape (Jasper Ridge Biological Preserve, California, USA), we combined field measurements with remotely sensed data from the Carnegie Airborne Observatory's light detection and ranging (lidar) system to create a detailed AGB map. We then developed a predictive model using a maximum of 56 explanatory variables derived from geologic and historic-ownership maps, a digital elevation model, and geographic coordinates to evaluate possible controls over currently observed AGB patterns. We tested both ordinary least-squares regression (OLS) and autoregressive approaches. OLS explained 44% of the variation in AGB, and simultaneous autoregression with a 100-m neighborhood improved the fit to an r2 = 0.72, while reducing the number of significant predictor variables from 27 variables in the OLS model to 11 variables in the autoregressive model. We also compared the results from these approaches to a more typical field-derived data set; we randomly sampled 5% of the data 1000 times and used the same OLS approach each time. Environmental filters including incident solar radiation, substrate type, and topographic position were significant predictors of AGB in all models. Past ownership was a minor but significant predictor, despite the long history of conservation at the site. The weak predictive power of these environmental variables, and the significant improvement when spatial autocorrelation was incorporated, highlight the importance of land-use history, disturbance regime, and population dynamics as controllers of AGB.     

Modeling and Predicting Species Occurrence Using Broad-Scale Environmental Variables: an Example with Butterflies of the Great Basin

Abstract: If occurrence of individual species can be modeled as a function of easily quantified environmental variables (e.g., derived from a geographic information system [GIS]) and the predictions of these models are demonstrably successful, then the scientific foundation for management planning will be strengthened. We used Bayesian logistic regression to develop predictive models for resident butterflies in the central Great Basin of western North America. Species inventory data and values for 14 environmental variables from 49 locations (segments of canyons) in the Toquima Range ( Nevada, U.S.A.) were used to build the models. Squares of the environmental variables were also used to accommodate possibly nonmonotonic responses. We obtained statistically significant models for 36 of 56 (64%) resident species of butterflies. The models explained 8–72% of the deviance in occurrence of those species. Each of the independent variables was significant in at least one model, and squared versions of five variables contributed to models. Elevation was included in more than half of the models. Models included one to four variables; only one variable was significant in about half the models. We conducted preliminary tests of two of our models by using an existing set of data on the occurrence of butterflies in the neighboring Toiyabe Range. We compared conventional logistic classification with posterior probability distributions derived from Bayesian modeling. For the latter, we restricted our predictions to locations with a high ( 70%) probability of predicted presence or absence. We will perform further tests after conducting inventories at new locations in the Toquima Range and nearby Shoshone Mountains, for which we have computed environmental variables by using remotely acquired topographic data, digital-terrain and microclimatic models, and GIS computation.     

Fine-scale habitat modeling of a top marine predator: do prey data improve predictive capacity?

Predators and prey assort themselves relative to each other, the availability of resources and refuges, and the temporal and spatial scale of their interaction. Predictive models of predator distributions often rely on these relationships by incorporating data on environmental variability and prey availability to determine predator habitat selection patterns. This approach to predictive modeling holds true in marine systems where observations of predators are logistically difficult, emphasizing the need for accurate models. In this paper, we ask whether including prey distribution data in fine-scale predictive models of bottlenose dolphin (Tursiops truncatus) habitat selection in Florida Bay, Florida, U.S.A., improves predictive capacity. Environmental characteristics are often used as predictor variables in habitat models of top marine predators with the assumption that they act as proxies of prey distribution. We examine the validity of this assumption by comparing the response of dolphin distribution and fish catch rates to the same environmental variables. Next, the predictive capacities of four models, with and without prey distribution data, are tested to determine whether dolphin habitat selection can be predicted without recourse to describing the distribution of their prey. The final analysis determines the accuracy of predictive maps of dolphin distribution produced by modeling areas of high fish catch based on significant environmental characteristics. We use spatial analysis and independent data sets to train and test the models. Our results indicate that, due to high habitat heterogeneity and the spatial variability of prey patches, fine-scale models of dolphin habitat selection in coastal habitats will be more successful if environmental variables are used as predictor variables of predator distributions rather than relying on prey data as explanatory variables. However, predictive modeling of prey distribution as the response variable based on environmental variability did produce high predictive performance of dolphin habitat selection, particularly foraging habitat.     

Modelling the productivity of naturalised pasture in the North Island,New Zealand: a decision tree approach

Decision tree, one of the data mining methods, has been widely used as a modelling approach and has shown better predictive ability than traditional approaches (e.g. regression). However, very little is known from the literature about how the decision tree performs in predicting pasture productivity. In this study, decision tree models were developed to investigate and predict the annual and seasonal productivity of naturalised hill-pasture in the North Island, New Zealand, and were compared with regression models with respect to model fit, validation and predictive accuracy. The results indicated that the decision tree models for annual and seasonal pasture productivity all had a smaller average squared error (ASE) and a higher percentage of correctly predicted cases than the corresponding regression models. The decision tree model for annual pasture productivity had an ASE which was only half of that of the regression model, and correctly predicted 90% of the cases in the model validation which was 10.8 percentage points higher than that of the regression model. Furthermore, the decision tree models for annual and seasonal pasture productivity also clearly revealed the relative importance of environmental and management variables in influencing pasture productivity, and the interaction among these variables. Spring rainfall was the most significant factor influencing annual pasture productivity, while hill slope was the most significant factor influencing spring and winter pasture productivity, and annual P fertiliser input and autumn rainfall were the most significant factors influencing summer and autumn pasture productivity. One limitation of using the decision tree to predict pasture productivity was that it did not generate a continuous prediction, and thus could not detect the influence of small changes in environmental and management variables on pasture productivity.     

Incorporating spatial autocorrelation into cellular automata model: An application to the dynamics of Chinese tamarisk (Tamarix chinensis Lour.)

Spatial autocorrelation (SAC) is frequently encountered in most spatial data in ecology. Cellular automata (CA) models have been widely used to simulate complex spatial phenomena. However, little has been done to examine the impact of incorporating SAC into CA models. Using image-derived maps of Chinese tamarisk (Tamarix chinensis Lour.), CA models based on ordinary logistic regression (OLCA model) and autologistic regression (ALCA model) were developed to simulate landscape dynamics of T. chinensis. In this study, significant positive SAC was detected in residuals of ordinary logistic models, whereas non-significant SAC was found in autologistic models. All autologistic models obtained lower Akaike's information criterion corrected for small sample size (AICc) values than the best ordinary logistic models. Although the performance of ALCA models only satisfied the minimum requirement, ALCA models showed considerable improvement upon OLCA models. Our results suggested that the incorporation of the autocovariate term not only accounted for SAC in model residuals but also provided more accurate estimates of regression coefficients. The study also found that the neglect of SAC might affect the statistical inference on underlying mechanisms driving landscape changes and obtain false ecological conclusions and management recommendations. The ALCA model is statistically sound when coping with spatially structured data, and the adoption of the ALCA model in future landscape transition simulations may provide more precise probability maps on landscape transition, better model performance and more reasonable mechanisms that are responsible for landscape changes.     

Changes in distribution patterns of weed species richness in agricultural lands on Qinghai-Tibet Plateau under climate change北大核心CSCD

农田杂草是阻碍农业生产的主要因素之一.明确农田杂草丰富度分布格局对农业生产管理具有重要意义.以青藏高原农田杂草为研究对象,利用物种分布模型探讨基于县域尺度的农田杂草物种丰富度分布格局及其未来(2050s)的变化,利用逐步回归筛选影响物种丰富度的环境因子,基于传统最小二乘法(OLS)和地理加权回归模型(GWR)分析环境因子对农田杂草物种丰富度的影响,并对两种分析方法进行比较.结果显示:(1)分布在青藏高原的农田主要杂草有51科284种,其中59种单子叶杂草、222种双子叶杂草、135种一年生杂草和149种多年生杂草.青藏高原农田杂草物种丰富度呈由西向东递增的变化规律,物种丰富度中心(丰富度值为167-194)主要集中在一江两河、河湟谷地和川西北等地区;(2)全球气候变化背景下,未来(2050s)青藏高原农田杂草物种丰富度整体呈由东南向西北方向增加的趋势,其中SSP1-2.6情境下最多增加43种,SSP5-8.5情境下最多增加49种;(3)GWR模型优于OLS,其结果表明青藏高原农田杂草物种丰富度的主要驱动因子是最冷季平均温、太阳辐射和最干月降水量,上述变量对杂草丰富度的影响存在明显的空间差异性,其中最冷季平均温由南向北逐渐从负向影响转变为正向影响.太阳辐射整体在青藏高原东部边缘等地区对农田杂草丰富度起正向的影响,在藏东南、青藏高原北部边缘等地区起负向的影响.最干月降水量对整个研究区域起负向影响,并表现出影响力由南向北逐步递增的趋势.上述结果表明青藏高原农田杂草物种丰富度调查不足,实际观测到的丰富度值明显低于当前气候下潜在的丰富度值,存在低估现象.当前气候背景下的农田杂草物种丰富度中心分布地区在未来仍是重点监管对象,且未来青藏高原部分地区作物可能面临新的杂草入侵风险.建议未来研究应注重于青藏高原粮食主产区农田杂草群落结构和功能调查、杂草和作物种间关系、耕地尺度上丰富度驱动因子分析等方面,为区域杂草管理和防治提供充分科学依据.(图6表2参53)     

Prediction of response of zooplankton biomass to climatic and oceanic changes

This paper examines the long-term variation in zooplankton biomass in response to climatic and oceanic changes, using a neural network as a nonlinear multivariate analysis method. Zooplankton data collected from 1951 to 1990 off the shore of northeastern Japan were analyzed. We considered patterns of the Kuroshio and the Oyashio, sea surface temperature, and meteorological parameters as environmental factors that affect zooplankton biomass. Back propagation neural networks were trained to generate mapping functions between environmental variables and zooplankton biomass. The performance of the network models was tested by varying the numbers of input and hidden units. Changes in zooplankton biomass could be predicted from environmental conditions. The neural network yielded predictions with smaller errors than those of predictions determined by linear multiple regression. The sensitivity analysis of networks was used to extract predictive knowledge. The air pressure, sea surface temperature, and some indices of atmospheric circulation were the primary factors for predictions. The patterns of the Kuroshio and the Oyashio demonstrated different effects among sea areas.