共查询到20条相似文献，搜索用时 31 毫秒
1.
Yunlong Chen Xiujuan Shan Dingyong Zeng Harry Gorfine Yinfeng Xu Qiang Wu Tao Yang Xianshi Jin 《海洋学报(英文版)》2022,41(6):22-30
Acquiring a comprehensive and accurate understanding of habitat preference is essential for species conservation and fishery management, especially for mobile species that migrate seasonally. Presence and absence data from field surveys are recommended when available due to their high reliability. Using field survey data, we investigated seasonal habitat suitability requirements for Tanaka’s snailfish (Liparis tanakae) in the Bohai Sea and Yellow Sea (BSYS) via a machine-learning method, random forests (RFs). Five environmental and biologically relevant variables (bottom temperature, bottom salinity, current velocity, depth and distance to shore) were used to build the ecological niches between the presence/absence data and suitable habitat. In addition, the degree to which false absence data might impact model performance was evaluated. Our results indicated that RFs provided accurate predictions, with seasonal habitat suitability maps of L. tanakae differing substantially. Bottom temperature and salinity were identified as important factors influencing the distribution of L. tanakae. False absence data were found to have negative effects on model performance and the decrease in evaluation metrics was usually significant (P<0.05) after 30% or more errors were added to the absence data. Through identifying highly suitable areas within its geographic range, our study provides a baseline for L. tanakae that can be further applied in ecosystem modelling and fishery management in the BSYS. 相似文献
2.
Mireia Valle Ángel BorjaGuillem Chust Ibon GalparsoroJoxe Mikel Garmendia 《Estuarine, Coastal and Shelf Science》2011,94(2):144-154
Predicting species distribution and habitat suitability is of considerable use in supporting the implementation of environmental legislation, protection and conservation of marine waters and ecosystem-based management. As other seagrasses, Zostera noltii has declined worldwide, mainly due to human pressures, such as eutrophication and habitat loss. In the case of the Basque Country (northern Spain), the species is present only in 3 out of 12 estuaries. From the literature, it is known that at least 6 of these estuaries were formerly vegetated by this seagrass. Consequently, efforts to monitor and restore (potential) habitats have been enhanced. Therefore, we aim: (i) to determine the main environmental variables explaining Zostera noltii distribution, within the Basque estuaries based upon the Oka estuary; (ii) to model habitat suitability for this species, as a wider applicable management-decision tool for seagrass restoration; and (iii) to assess the applicability and predicted accuracy of the model by using internal and external validation methods. For this purpose, Ecological Niche Factor Analysis (ENFA) has been used to model habitat suitability, based upon topographical variables, obtained from bathymetric Light Detection And Ranging (LiDAR); sediment characteristics variables; and hydrodynamic variables. The results obtained from the ecological factors of the ENFA (Marginality: 1.00; Specialization: 2.59) indicate that the species habitat differs considerably from the mean environmental conditions over the study area; likewise, that the species is restrictive in the selection of the range of conditions within which it dwells. The main environmental variables relating to the species distribution, in order of importance, are: mean grain size; redox potential; intertidal height; sediment sorting; slope of intertidal flat; percentage of gravels; and percentage of organic matter content. The model has a high predicted accuracy (Boyce index: 0.92). Model validation using an independent dataset in the Bidasoa estuary has shown the applicability but also the limitations in extrapolating the habitat suitability model to select suitable transplantation areas in other estuaries with similar morphological and biogeographical characteristics. ENFA-technique, applied with an accurate selection of environmental predictors, could be a promising tool for predicting seagrass habitat suitability which could assist on seagrass conservation and restoration programs worldwide. 相似文献
3.
Recent habitat suitability models used to predict the occurrence of vulnerable marine species, particularly framework building cold-water corals, have identified terrain attributes such as slope and bathymetric position index as important predictive parameters. Due to their scale-dependent nature, a realistic representation of terrain attributes is crucial for the development of reliable habitat suitability models. In this paper, three known coral areas and a noncoral control area off the west coast of Ireland were chosen to assess quantitative and distributional differences between terrain attributes derived from bathymetry grids of varying resolution and information content. Correlation analysis identified consistent changes of terrain attributes as grain size was altered. Response characteristics and dimensions depended on terrain attribute types and the dominant morphological length-scales within the study areas. The subsequent effect on habitat suitability maps was demonstrated by preliminary models generated at different grain sizes. This study demonstrates that high resolution habitat suitability models based on terrain parameters derived from multibeam generated bathymetry are required to detect many of the topographical features found in Irish waters that are associated with coral. This has implications for marine spatial planning in the deep sea. Supplemental materials are available for this article. Go to the publisher's online edition of Marine Geodesy to view the free supplemental file. 相似文献
4.
Seismic surveys are frequently a matter of concern regarding their potentially negative impacts on marine mammals. In the Southern Ocean, which provides a critical habitat for several endangered cetacean species, seismic research activities are undertaken at a circumpolar scale. In order to minimize impacts of these surveys, pre-cruise planning requires detailed, spatio-temporally resolved knowledge on the likelihood of encountering these species in the survey area. In this publication we present predictive habitat modelling as a potential tool to support decisions for survey planning. We associated opportunistic sightings (2005–2011) of humpback (Megaptera novaeangliae, N=93) and Antarctic minke whales (Balaenoptera bonaerensis, N=139) with a range of static and dynamic environmental variables. A maximum entropy algorithm (Maxent) was used to develop habitat models and to calculate daily basinwide/circumpolar prediction maps to evaluate how species-specific habitat conditions evolved throughout the spring and summer months. For both species, prediction maps revealed considerable changes in habitat suitability throughout the season. Suitable humpback whale habitat occurred predominantly in ice-free areas, expanding southwards with the retreating sea ice edge, whereas suitable Antarctic minke whale habitat was consistently predicted within sea ice covered areas. Daily, large-scale prediction maps provide a valuable tool to design layout and timing of seismic surveys as they allow the identification and consideration of potential spatio-temporal hotspots to minimize potential impacts of seismic surveys on Antarctic cetacean species. 相似文献
5.
The aim of the study was the development of habitat models for Nephtys species (Polychaeta: Nephtyidae). The investigation area was the German Bight, the southeastern part of the North Sea. Models were developed based on field data collected between 2000 and 2006. In addition, data on environmental variables were retrieved from long-term monitoring data sets and from the sediment map by Figge [Figge, K., 1981. Nordsee. Sedimentverteilung in der Deutschen Bucht. Map No. 2900. Publisher: Deutsches Hydrographisches Institut, Hamburg]. The statistical modelling technique used was multivariate adaptive regression splines (MARS). Models were fitted individually for each species. Evaluation of predictive discrimination and predictive accuracy of the developed models was by calculation of the area under the receiver operating curve (AUC) or sensitivity and specificity, respectively. Habitat models with best predictive fit were selected for the presentation of habitat suitability maps.Six Nepthys species were found: Nephtys assimilis, N. caeca, N. cirrosa, N. hombergii, N. incisa and N. longosetosa. N. hombergii was most common whereas N. incisa and N. longosetosa were rare. Habitat preferences varied considerably among the species. For all investigated Nephtys species except N. longosetosa a habitat model could be developed based on four predictor variables. The habitat models with best predictive fit were those for N. cirrosa and N. hombergii. The N. caeca habitat model was of limited predictive accuracy and only accept predictive discrimination. The number of predictors as well as the relative importance of the respective predictors in the model varied among the different species. Direct comparison of most suitable habitats for the different species based on modelling revealed that in the mostly sandy regions parallel to the German coast in water depths up to 20 m an overlap between N. caeca, N. hombergii and N. cirrosa exists. In the deeper central German Bight with mostly fine sands with increased mud contents N. hombergii, N. assimilis and, at least partially and rare in numbers, N. incisa co-occur. It can be concluded that important sediment characteristics like grain size median and mud content as well as water depth and mean salinity are useful parameters to describe the habitat requirements of most Nepthys species in the German Bight. However, additional variables need to be incorporated into such analyses. 相似文献
6.
《Journal of Sea Research》2009,61(4):276-291
The aim of the study was the development of habitat models for Nephtys species (Polychaeta: Nephtyidae). The investigation area was the German Bight, the southeastern part of the North Sea. Models were developed based on field data collected between 2000 and 2006. In addition, data on environmental variables were retrieved from long-term monitoring data sets and from the sediment map by Figge [Figge, K., 1981. Nordsee. Sedimentverteilung in der Deutschen Bucht. Map No. 2900. Publisher: Deutsches Hydrographisches Institut, Hamburg]. The statistical modelling technique used was multivariate adaptive regression splines (MARS). Models were fitted individually for each species. Evaluation of predictive discrimination and predictive accuracy of the developed models was by calculation of the area under the receiver operating curve (AUC) or sensitivity and specificity, respectively. Habitat models with best predictive fit were selected for the presentation of habitat suitability maps.Six Nepthys species were found: Nephtys assimilis, N. caeca, N. cirrosa, N. hombergii, N. incisa and N. longosetosa. N. hombergii was most common whereas N. incisa and N. longosetosa were rare. Habitat preferences varied considerably among the species. For all investigated Nephtys species except N. longosetosa a habitat model could be developed based on four predictor variables. The habitat models with best predictive fit were those for N. cirrosa and N. hombergii. The N. caeca habitat model was of limited predictive accuracy and only accept predictive discrimination. The number of predictors as well as the relative importance of the respective predictors in the model varied among the different species. Direct comparison of most suitable habitats for the different species based on modelling revealed that in the mostly sandy regions parallel to the German coast in water depths up to 20 m an overlap between N. caeca, N. hombergii and N. cirrosa exists. In the deeper central German Bight with mostly fine sands with increased mud contents N. hombergii, N. assimilis and, at least partially and rare in numbers, N. incisa co-occur. It can be concluded that important sediment characteristics like grain size median and mud content as well as water depth and mean salinity are useful parameters to describe the habitat requirements of most Nepthys species in the German Bight. However, additional variables need to be incorporated into such analyses. 相似文献
7.
The quality of environmental data and its possible impact on the marine species habitat modelling are often overlooked while the sources for these data are increasing. This study selected sea surface temperature(SST) from two commonly used sources, the NOAA Ocean Watch and IRI/LDEO Climate Data Library, and then constructed habitat suitability index model to evaluate the influences of SST from the two sources on the outcomes of Ommastrephes bartramii habitat models for the months of July–October... 相似文献
8.
We developed habitat suitability curves (HSC) using generalised additive models (GAMs) for nine benthic macroinvertebrate taxa from a small New Zealand river for hydraulic-habitat modelling assessments of instream flow requirements. We included interaction terms between the primary variables (water depth, velocity, substrate) when significant, to address a longstanding criticism of univariate HSC. To date, only large-river univariate HSC have been available and these have been used in hydraulic-habitat applications on small rivers, despite doubt over the transferability of HSC between rivers of different size and type. We tested the outcome on the predicted abundance–flow relationship of applying the small-river habitat suitability GAMs versus large-river GAMs for two taxa on the same small river. We found the effects of flow allocation were overestimated by the large-river GAMs relative to the small-river GAMs. Further research to develop general HSC for categories of river size and type is needed to better inform hydraulic-habitat modelling applications. 相似文献
9.
Long‐term presence and habitat use of Cuvier's beaked whale (Ziphius cavirostris) in the Central Tyrrhenian Sea 
下载免费PDF全文
下载免费PDF全文 Cuvier's beaked whale (Ziphius cavirostris, G. Cuvier 1823) is a poorly known species and many international agreements have asked for a better understanding of its biology for conservation purposes. In the present study, systematic cetacean surveys were carried out from ferries along a trans‐regional fixed transect in the Central Tyrrhenian Sea (Civitavecchia, Latium – Golfo Aranci, Sardinia), just outside the southeastern border of the Pelagos Sanctuary. This research provided long‐term, consistent data on Cuvier's beaked whale during two research periods (1990–1992 and 2007–2011). The objective of the research was to compare the presence, distribution and habitat use of Cuvier's beaked whale between the two investigated periods. Summer data (June–September) from the two periods were compared in terms of frequency of sightings, group size and spatial distribution related to the main ecogeographical features. A presence‐absence model (generalized additive modelling) was performed to predict habitat suitability in the two study periods. The results highlight long‐term site fidelity of Cuvier's beaked whale in the Central Tyrrhenian Sea with encounter rates comparable to the ones reported for other key areas. Separate suitability models based on 1990s and 2000s data appeared to work for each individual time period but differences were evident between the two periods, indicating changes in habitat selection over time. Our findings of the study appear to expand the definition of suitable beaked whale habitat and underline how the temporal scale of the analysis can affect the results in habitat studies. Moreover, this research highlights the importance of the Central Tyrrhenian Sea marine region for Cuvier's beaked whale and the ability of continuous monitoring to identify changes in cetacean frequency and distribution, necessary for adaptive conservation management approaches. 相似文献
10.
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. 相似文献
11.
Andrew J. Davies Max Wisshak James C. Orr J. Murray Roberts 《Deep Sea Research Part I: Oceanographic Research Papers》2008,55(8):1048-1062
Ecological-niche factor analysis (ENFA) was applied to the reef framework-forming cold-water coral Lophelia pertusa. The environmental tolerances of this species were assessed using readily available oceanographic data, including physical, chemical, and biological variables. L. pertusa was found at mean depths of 468 and 480 m on the regional and global scales and occupied a niche that included higher than average current speed and productivity, supporting the theory that their limited food supply is locally enhanced by currents. Most records occurred in areas with a salinity of 35, mean temperatures of 6.2–6.7 °C and dissolved oxygen levels of 6.0–6.2 ml l−1. The majority of records were found in areas that were saturated with aragonite but had low concentration of nutrients (silicate, phosphate, and nitrate). Suitable habitat for L. pertusa was predicted using ENFA on a global and a regional scale that incorporated the north-east Atlantic Ocean. Regional prediction was reliable due to numerous presence points throughout the area, whereas global prediction was less reliable due to the paucity of presence data outside of the north-east Atlantic. However, the species niche was supported at each spatial scale. Predicted maps at the global scale reinforced the general consensus that the North Atlantic Ocean is a key region in the worldwide distribution of L. pertusa. Predictive modelling is an approach that can be applied to cold-water coral species to locate areas of suitable habitat for further study. It may also prove a useful tool to assist spatial planning of offshore marine protected areas. However, issues with eco-geographical datasets, including their coarse resolution and limited geographical coverage, currently restrict the scope of this approach. 相似文献
12.
13.
剑鱼（Xiphias gladius）是一种高度洄游性鱼类,其迁徙和栖息地利用受海洋环境影响明显,理解其空间分布格局形成的机制对于资源的养护和管理具有重要意义。本研究利用2017-2019年中国印度洋延绳钓渔业观察员数据中剑鱼的渔获物信息作为物种出现数据,结合西印度洋海域的海表温度、海面高度、叶绿素a浓度、混合层深度、海表盐度等环境数据,采用最大熵模型对剑鱼的栖息地适宜性分布进行了模拟。结果表明：（1）模型对印度洋西部剑鱼栖息地适宜性分布的模拟精度非常高,各个季节受试者工作特征曲线的曲线下面积都大于0.9,可用于模拟剑鱼潜在的栖息地适宜性分布;（2）研究区域内剑鱼适宜栖息地分布变化与实际作业位置变动基本一致,干季和湿季剑鱼栖息地高适宜性的区域分布都较为集中,但湿季分布范围要大于干季;（3）海表温度、海表盐度和混合层深度是影响西印度洋剑鱼栖息地适宜性分布的重要环境因子,在干季和湿季的最适范围分别为25.8～31.6℃、34.4～35.9、0.1～24.9 m和25.6～30.5℃、34.8～36.4、13.1～54.1 m。研究结果可为西印度洋海域剑鱼种群的可持续利用和科学管理提供必要参... 相似文献
14.
Different techniques of species distribution modeling were applied to evaluate the distribution of eight benthic marine species in Icelandic waters. The species examined were Symplectoscyphus tricuspidatus, Stegopoma plicatile (both Hydrozoa), Prionospio cirrifera, Amphicteis gunneri (both Polychaeta), Desmosoma strombergi, Eurycope producta (both Isopoda), Andaniella pectinata and Harpinia crenulata (both Amphipoda). Information on 13 environmental variables (temperature mean, temperature mean SD, temperature minimum, temperature maximum, salinity mean, salinity mean SD, oxygen content, particulate organic carbon, seasonal variation index, bottom roughness, sediment thickness, acidification) and records of occurrences of these eight species was collated in an ArcGIS project. Modeling methods applied were MARS, TreeNet, and MaxENT. According to area under the receiver operating curve (AUC) model assessment values, models with moderate to outstanding discriminatory power were found for all species. There was a good overlap in the overall pattern of prediction for most species independent on the modeling technique. Among the three applied techniques MARS seemed to generalize most whereas TreeNet predictions very precisely reflected information from the training data set. The distribution of the selected benthic invertebrate species in Icelandic waters could be linked to a variety of environmental factors related to oceanography, seabed topography and human impact. Their multivariate interactions acted as a structuring force of species distribution, instead of just their one by one individual influence. The selected predictors varied between the different models for the same species. They substituted each other in different models. The expected distribution of the examined species was mapped for a seascape of known environmental settings. Such maps will serve as excellent references in future impact studies and enable the detection of changes in the distribution of benthic marine invertebrates. 相似文献
15.
Explaining island‐wide geographical patterns of Caribbean fish diversity: A multi‐scale seascape ecology approach 下载免费PDF全文
下载免费PDF全文 Geographical patterning of fish diversity across coral reef seascapes is driven by many interacting environmental variables operating at multiple spatial scales. Identifying suites of variables that explain spatial patterns of fish diversity is central to ecology and informs prioritization in marine conservation, particularly where protection of the highest biodiversity coral reefs is a primary goal. However, the relative importance of conventional within‐patch variables versus the spatial patterning of the surrounding seascape is still unclear in the ecology of fishes on coral reefs. A multi‐scale seascape approach derived from landscape ecology was applied to quantify and examine the explanatory roles of a wide range of variables at different spatial scales including: (i) within‐patch structural attributes from field data (5 × 1 m2 sample unit area); (ii) geometry of the seascape from sea‐floor maps (10–50 m radius seascape units); and wave exposure from a hydrodynamic model (240 m resolution) for 251 coral reef survey sites in the US Virgin Islands. Non‐parametric statistical learning techniques using single classification and regression trees (CART) and ensembles of boosted regression trees (TreeNet) were used to: (i) model interactions; and (ii) identify the most influential environmental predictors from multiple data types (diver surveys, terrain models, habitat maps) across multiple spatial scales (1–196,350 m2). Classifying the continuous response variables into a binary category and instead predicting the presence and absence of fish species richness hotspots (top 10% richness) increased the predictive performance of the models. The best CART model predicted fish richness hotspots with 80% accuracy. The statistical interaction between abundance of living scleractinian corals measured by SCUBA divers within 1 m2 quadrats and the topographical complexity of the surrounding sea‐floor terrain (150 m radius seascape unit) measured from a high‐resolution terrain model best explained geographical patterns in fish richness hotspots. The comparatively poor performance of models predicting continuous variability in fish diversity across the seascape could be a result of a decoupling of the diversity‐environment relationship owing to structural degradation leading to a widespread homogenization of coral reef structure. 相似文献
16.
Species distribution maps are needed for ecosystem-based marine management including the development of marine spatial plans. If such maps are based on predictive models then modelling procedures should aim to maximise validation success, and any uncertainty in the predictions needs to be made explicit. We developed a predictive modelling approach to produce robust maps of the distributions of selected marine species at a regional scale. We used 14 years of survey data to map the distributions of plaice, sole and thornback ray in three hydrographic regions comprising parts of the Irish Sea, Celtic Sea and the English Channel with the help of the hybrid technique regression kriging, which combines regression models with geostatistical tools. For each species–region combination we constructed logistic Generalized Linear Models (GLMs) based on presence–absence data using the environmental variables: depth, bottom temperature, bed shear stress and sediment type, as predictors. We selected GLMs using the mean squared error of prediction (MSEP) estimated by cross-validation then conducted a geostatistical analysis of the residuals to incorporate spatial structure in the predictions. In general, we found that species occurrence was positively related to shallow areas, a bed shear stress of between 0 and 1.5 N/m2, and the presence of sandy sediment. Predicted species occurrence probabilities were in good agreement with survey observations. This modelling framework selects environmental models based on predictive ability and considers the effect of spatial autocorrelation on predictions, together with the simultaneous presentation of observations, associated uncertainties, and predictions. The potential benefit of these distribution maps to marine management and planning is discussed. 相似文献
17.
This article adopts Multivariate Adaptive Regression Spline (MARS) for prediction of Angle of Shearing Resistance(?) of soil. MARS is an adaptive, non-parametric regression approach. Percentages of fine-grained (FG), coarse-grained (CG), liquid limit (LL), and bulk density (BD) have been used as input variables of MARS. The developed MARS gives an equation for prediction of ? of soil. The results of MARS have been compared with Genetic Expression Programming (GEP), Artificial Neural Network (ANN), and Adaptive Neuro Fuzzy Inference System (ANFIS) models. These results demonstrate that the developed MARS can be used as a robust model for determination of ? of soil. 相似文献
18.
The pikeperch (Sander lucioperca (L.)) is an economically important fish species occurring in the fresh and brackish waters of Europe. To evaluate the distribution and extent of the reproduction areas in the northern Baltic Sea, a field survey was carried out in two separate coastal areas. Presence/absence data were used to develop a geographic information system (GIS)-based predictive spatial distribution model, where high resolution raster maps of the focal environmental variables and a logistic regression equation were used to predict the probability of larval occurrence. The results indicated that the pikeperch reproduction areas are located in the innermost archipelago zone where high water turbidity best explained their presence. Turbidity was related to several other variables such as fetch and depth. Contrary to our preliminary hypothesis, surface water temperatures measured during the survey had no significant effect in the model due to the low spatial variation in the measured values. Since turbidity is possible to determine by remote sensing methods, the probability maps can be cost-effectively extended to more extensive coastal areas with proper validation. 相似文献
19.
Giovanni Bearzi Arianna Azzellino Elena Politi Marina Costa Mauro Bastianini 《Ocean Science Journal》2008,43(4):175-182
Bottlenose dolphins are the only cetaceans regularly observed in the northern Adriatic Sea, but they survive at low densities
and are exposed to significant threats. This study investigates some of the factors that influence habitat use by the animals
in a largely homogeneous environment by combining dolphin data with hydrological and physiographical variables sampled from
oceanographic ships. Surveys were conducted year-round between 2003 and 2006, totalling 3,397 km of effort. Habitat modelling
based on a binary stepwise logistic regression analysis predicted between 81% and 93% of the cells where animals were present.
Seven environmental covariates were important predictors: oxygen saturation, water temperature, density anomaly, gradient
of density anomaly, turbidity, distance from the nearest coast and bottom depth. The model selected consistent predictors
in spring and summer. However, the relationship (inverse or direct) between each predictor and dolphin presence varied among
seasons, and different predictors were selected in fall. This suggests that dolphin distribution changed depending on seasonal
forcing. As the study area is relatively uniform in terms of bottom topography, habitat use by the animals seems to depend
on complex interactions among hydrological variables, caused primarily by seasonal change and likely to determine shifts in
prey distribution. 相似文献
20.
三疣梭子蟹(Portunus trituberculatus)是莱州湾最重要的经济蟹类,其资源丰度受栖息环境影响显著,为了解不同的环境因子对其栖息地分布的影响,根据2010~2020年夏季底拖网调查数据,研究莱州湾三疣梭子蟹栖息地适宜性及其影响因子。利用广义可加模型(generalized additive models,GAM)选取变量因子,通过提升回归树模型(boosting regression tree,BRT)对因子进行权重分析,构建了4种栖息地适宜性指数(habitat suitability index,HSI)模型,并通过实测值和预测值的Pearson检验,对模型比较和验证。结果表明,GAM和BRT优化的HSI模型好于其他3种模型(未优化模型,GAM优化HSI模型,BRT优化HSI模型),以生物量表征资源丰度的模型好于尾数表征资源丰度的模型,算术平均法构建的HSI模型的整体预测准确率和相关系数高于几何平均法,对莱州湾三疣梭子蟹栖息地有较强的预测能力。底层水温、底层盐度和水深对三疣梭子蟹栖息地影响较大,夏季栖息地适宜性较高的海域(HSI>0.7)主要分布在莱州湾南部、东南部和东北部海域,中部和西南部海域分布较少。研究结果为莱州湾三疣梭子蟹增殖放流工作提供了科学依据。 相似文献