Genetic structure and life history are key factors in species distribution models of spiny lobsters

AimWe incorporated genetic structure and life history phase in species distribution models (SDMs) constructed for a widespread spiny lobster, to reveal local adaptations specific to individual subspecies and predict future range shifts under the RCP 8.5 climate change scenario.LocationIndo‐West Pacific.MethodsMaxEnt was used to construct present‐day SDMs for the spiny lobster Panulirus homarus and individually for the three genetically distinct subspecies of which it comprises. SDMs incorporated both sea surface and benthic (seafloor) climate layers to recreate discrete influences of these habitats during the drifting larval and benthic juvenile and adult life history phases. Principle component analysis (PCA) was used to infer environmental variables to which individual subspecies were adapted. SDM projections of present‐day habitat suitability were compared with predictions for the year 2,100, under the RCP 8.5 climate change scenario.ResultsIn the PCA, salinity best explained P. h. megasculptus habitat suitability, compared with current velocity in P. h. rubellus and sea surface temperature in P. h. homarus. Drifting and benthic life history phases were adapted to different combinations of sea surface and benthic environmental variables considered. Highly suitable habitats for benthic phases were spatially enveloped within more extensive sea surface habitats suitable for drifting larvae. SDMs predicted that present‐day highly suitable habitats for P. homarus will decrease by the year 2,100.Main conclusionsIncorporating genetic structure in SDMs showed that individual spiny lobster subspecies had unique adaptations, which could not be resolved in species‐level models. The use of sea surface and benthic climate layers revealed the relative importance of environmental variables during drifting and benthic life history phases. SDMs that included genetic structure and life history were more informative in predictive models of climate change effects.     

不同营林面积情景下鹿科动物的潜在生境分布

在有人为干扰的森林景观中开展鹿科动物适宜生境分布研究,对于解决大尺度生境保护与小面积森林经营的矛盾问题有着重要的参考意义,也符合我国林区的现实需求.2013—2015年冬季进行的多次野外调查收集196处鹿科动物出现点信息,将这些点作为分布点数据,选取地形、景观类型、植被特征和人类干扰4类17种因子作为环境变量,利用最大熵模型方法,分析4种林下经营面积情景下小兴安岭铁力林业局马鹿和狍的潜在适宜生境分布特征及其对环境因子的响应.结果表明:模型预测精度达到优秀水平,稳定性好,鹿科动物的适宜生境主要集中在东部区域;不同情景下,两种鹿科动物的主要环境影响因子相似,均为距农田距离、距居民点距离、距河流距离、距营林区距离和海拔因子,其中,距营林区距离因子的贡献率稳定在4%～6%;两种鹿科动物躲避人类经营活动干扰的距离(1200～1300 m)较为接近.在无林下经营情景中,鹿科动物的适宜生境分布较广、面积较大;随着经营面积的增大,适宜生境面积减少;当经营面积扩大到现实情况的2～3倍时,鹿科动物栖息地面积缩减较为严重.     

Relative predation risk in two types of habitat for juvenile Australasian spiny lobsters,Jasus edwardsii

The decline in kelp habitat on coastal reefs resulting from changes in ocean climate and the distribution and abundance of herbivorous species is common in many temperate regions of the world. Kelp habitat is highly productive, biodiverse and provides a complex habitat into which many organisms recruit, including spiny lobsters, such as the Australasian red spiny lobster, Jasus edwardsii. The displacement of kelp habitat by less-complex barren reef habitat has the potential to influence the risk of predation for early juvenile lobsters. Therefore, relative predation risk on the juvenile spiny lobster, J. edwardsii, was compared for kelp and barren habitats on the northeast coast of New Zealand using juvenile lobsters held in transparent containers and recording predators with a video recorder. In total, 188 predation attempts were observed within 420?h of video recordings gathered over 3 weeks of sampling. There was an overall higher predation risk in barren habitats. Daytime predation attempts were higher in barren compared to kelp habitat; however, there was no difference between the habitats for night time, dawn or dusk observations, when juvenile lobsters are emergent from shelters and vulnerable to predation. Similar numbers of predatory species were identified in kelp (13) and barren habitat (12). Other factors, such as food availability and time spent away from shelter, especially during night and crepuscular periods, need consideration in future studies when investigating the cause of differences in juvenile lobster mortality among habitats.     

不同气候情景下木梨潜在地理分布格局变化的预测

木梨的抗旱、抗寒、抗盐碱和抗梨锈病的能力较强,是我国西北地区梨的主要砧木类型之一,具有较高的生产价值。但目前其生境破坏较为严重。预测不同气候情景下木梨的地理分布可为木梨资源的合理开发利用及多样性保护提供重要的科学依据。本研究利用木梨全面且精确的分布信息和高分辨率环境数据,基于MaxEnt模型和ArcGIS空间分析,构建其当代及未来(2050和2070年)的潜在空间分布格局,评价环境因子对分布模型的重要性。结果表明: 目前木梨适宜生境面积为33.2万km²,主要位于我国青海东部、甘肃南部、宁夏南部、陕西中部、山西南部和河南西部地区,紫外线辐射量最少月份的平均紫外线量和海拔是限制其分布的主要环境因子。随着全球气候变暖,在不同CO₂浓度情景下,2050和2070年木梨的潜在适生境将逐渐减少,应加强对木梨群体的实时监测。     

小兴安岭南麓马鹿冬季适宜生境评价

通过2014-2015年两次冬季野外调查, 将收集的79处马鹿(Cervus elaphus)出现信息作为分布点数据, 选取地形、景观类型、植被特征和人类干扰4类19种因子作为环境变量, 利用最大熵(maximum entropy, MaxEnt)模型, 分析了小兴安岭铁力林业局辖区马鹿种群冬季潜在适宜生境分布特征和主要环境因子对马鹿分布的影响。结果显示: 模型预测精度较高, 训练集与验证集的平均AUC(area under the curve, 受试工作者曲线下面积)值分别为0.949和0.958; Jackknife检验结果表明: 景观类型因子对马鹿生境选择的影响较大; 坡向、距大路距离、距混交林距离、距灌草地距离和距农田距离是影响马鹿生境分布的主要环境因子, 其综合贡献值依次为27.8%、23.9%、19.5%、15.3%和10.4%; 距小路距离对马鹿分布影响较小。我们依据MaxEnt模型最大约登指数, 找到最佳中断点0.22作为阈值将马鹿冬季栖息地划分为适宜和不适宜两个等级, 其面积分别为663.49 km²和1,378.85 km², 分别占研究区总面积的32%和68%。马鹿的适宜生境主要分布在铁力林业局辖区的东部山地和中部林地等区域; 南部地区接近铁力市区, 人类活动频繁, 不适宜马鹿栖息。对马鹿种群的保护管理措施提出3点建议: 控制人为干扰; 构建多样性景观; 优先保护马鹿的潜在适宜生境分布区。     

气候变化下考氏白盾蚧的潜在分布区预测

【目的】评估园林植物害虫考氏白盾蚧Pseudaulacaspis cockerelli当前和未来在全世界的潜在分布区,揭示未来气候变化下考氏白盾蚧的分布动态,明确气候环境因素对其潜在分布的影响。【方法】以考氏白盾蚧为研究对象,基于考氏白盾蚧在全球的118条有效地理分布记录和19个环境变量,运用优化的MaxEnt模型和ArcGIS软件,推测气候变化下当前、2050年和2070年考氏白盾蚧的潜在分布格局,采用响应曲线确定环境变量的适宜区间,定量确定考氏白盾蚧未来气候条件下潜在地理分布动态。【结果】MaxEnt模型运算的平均曲线下面积(area under the curve, AUC)值为0.7182,表明该预测模型的预测精度比较高。当前考氏白盾蚧潜在地理分布的总适生区面积约为2.73×10⁷ km²,其中高适生区面积大约为4.37×10⁶ km²,占潜在可入侵总面积的16％,该区域主要位于美国与巴西西南沿海地区,印度西部地区及西部沿海区域,孟加拉国,越南北部大部,中国西南大部及华东华中大部,以及日本南部地区;在未来气候条件下,伴随着CO₂浓度的升高,考氏白盾蚧的高适生面积将显著增加。影响考氏白盾蚧的潜在地理分布的主要环境变量为平均月温差、昼夜温差与年温差比、最湿季平均温度和降水季节性,其中昼夜温差与年温差比的贡献率最高,达到38.8%。【结论】本研究结果表明考氏白盾蚧适宜生境主要受平均月温差和昼夜温差与年温差比的影响。本研究为考氏白盾蚧的综合防治提供重要依据和数据支撑。     

Habitat distribution modeling of endangered medicinal plant Picrorhiza kurroa (Royle ex Benth) under climate change scenarios in Uttarakhand Himalaya,India

Climate change has been the key factor in changing the alpine vegetation's habitat and causing it to migrate to higher latitudes. The present study aims to model the current and future potential habitat distribution of endangered medicinal plant Picrorhiza kurroa Royle ex Benth in Uttarakhand Himalaya using the maximum entropy (MaxEnt) modeling. We initially select twenty-two environmental variables (bioclimatic + topographic) got from the Fifty-four (54) species occurrence points, which were further reduced to nine variables to prevent multicollinearity. Shared Socioeconomic Pathways (SSP1–2.6 and SSP2–4.5) from the CMIP6 (BCC-CSM2-MR) climate model for the periods 2041–60 and 2061–80 were used to predict the current and future habitat distribution of P. kurroa. Results showed that the precipitation of the driest month (Bio 14; 33.8%), isothermality (Bio 3; 20.2%), mean temperature of warmest quarter (Bio 10; 12.7%), and temperature annual range (Bio 7; 12.2%) were the important bioclimatic variables influencing the habitat of P. kurroa. Overall, there is a decrease in the habitat of P. kurroa under climate change scenarios. The present results may prove insightful for the decision-makers to identify suitable sites in the wild for the further propagation of P. kurroa.     

基于MaxEnt模型西南地区高山植被对气候变化的响应评估

采用1∶100万的中国植被类型图以及19个气候环境变量数据,基于最大熵(MaxEnt)算法和ArcGIS空间分析模块构建西南地区高山植被地理分布的气候适宜性预测模型,模拟其在基准期(1960—2000年)和不同气候情景下(A2、A1B和B1)的气候适宜性分布格局,并评价其对气候变化的适应性。结果表明:MaxEnt模型分析研究区高山植被地理分布气候适宜性的适用性非常高(AUC=0.93);最暖月均温、最湿季均温、最冷月均温等温度变量是限制其地理分布的主要气候因子;研究区高山植被地理分布的气候适宜区主要集中在西藏自治区、青海省、四川省西部及云南省西北部的部分地区;完全适宜、中度适宜、轻度适宜、不适宜的面积所占总面积比例约为1∶1∶2∶5;1960—2050年研究区高山植被潜在地理分布的气候适宜性面积有不同定程度的减少;未来3种气候变化情景下高山植被地理分布对气候变化的适应性分布格局基本一致,均为不适应区所占总面积比例较大;伴随气候变化,研究区高山植被的适应性减弱,体现在其潜在地理分布对气候变化的适应区分布范围减少;海拔5000—5500m适应性较强,适应区所占面积比例最大(53%左右);3500—4500m适应性最弱,适应区所占面积比例最小(5%左右)。     

基于MaxEnt生态位模型预测对齿藓属(Didymodon)植物在新疆的潜在地理分布

该文基于MaxEnt模型,利用获得的132个对齿藓属(Didymodon)植物在新疆分布的信息,结合RCP45 CO2排放情景下2050年和2070年的19个生物气候数据预测该属在当代、2050年和2070年的潜在分布区域。结果显示,最湿季平均温度、年平均气温、最干季降水量和年降水量是影响该属分布最主要的气候因子,其贡献率分别为33.6%、22.2%、16.4%和14.6%;模型模拟准确度高(AUC值达0.84);在当代气候条件下,对齿藓属植物的适宜生境面积占新疆总面积的38.51%;最适分布区域是中部的天山山脉、南部昆仑山脉的东部和西部的帕米尔高原;与当代的分布预测结果相比,未来(2050年和2070年)该属适宜栖息地分布范围总体上呈现退缩趋势;退缩后的适宜生境面积分别占新疆总面积的36.56%和37.87%。温度和降水量可能是引起对齿藓属地理分布退缩的重要气候因子。研究结果可为探讨气候变化对干旱、半干旱区苔藓植物物种分布的影响提供参考资料。     

Predicting the distribution of green turtle nesting sites over the Mediterranean with outcoming climate driven changes

Nesting beaches have a critical role in the life cycle of sea turtles and their survival. Many different factors affect nest site selection, ranging from the composition of the sand to the vegetation of the beach. These factors are subject to change due to the onset of climate change. We aimed to determine the possible changes in nesting beaches according to the future climate scenarios of Chelonia mydas nesting sites in the Mediterranean by ecological niche modeling. Nineteen bioclimatic variables and Representative Concentration Pathway scenarios (RCP2.6 and RCP8.5) were used to generate past, current, and future nesting site projections. The datasets were prepared with ArcGIS v10. and bioclimatic variables were analyzed using the Pearson Correlation Analysis. The ecological niche modeling was made with the MaxEnt v4.1.0. Model outputs, mean temperature of warmest quarter (22.01 %), precipitation of coldest quarter (15.32 %), mean temperature of the driest quarter (13.60 %), isothermality (12.30 %), mean diurnal range (9.22 %), the max temperature of the warmest month (6.60 %), precipitation seasonality (5.87 %) and annual mean temperature (4.73 %) are the parameters that most affect the estimated distribution of the species and the other parameters have the least effect on the estimated distribution (each < 2.60 %). The prediction accuracy of the model is measured by the Area Under the Curve (AUC) values, which is between 0 and 1, where values closer to 1 have a greater prediction accuracy. In our model results, the AUC values vary between 0.961 and 0.990. The majority of current green turtle nesting sites will continue to be suitable for nesting into the 2100′s. But the habitat suitability of the current nesting beaches in Syria and Lebanon will decrease. Conservational efforts should be developed to protect not only the current nesting beaches but also other possible nesting beaches that might become viable in the future.     

Climate Change May Alter Breeding Ground Distributions of Eastern Migratory Monarchs (Danaus plexippus) via Range Expansion of Asclepias Host Plants

Climate change can profoundly alter species’ distributions due to changes in temperature, precipitation, or seasonality. Migratory monarch butterflies (Danaus plexippus) may be particularly susceptible to climate-driven changes in host plant abundance or reduced overwintering habitat. For example, climate change may significantly reduce the availability of overwintering habitat by restricting the amount of area with suitable microclimate conditions. However, potential effects of climate change on monarch northward migrations remain largely unknown, particularly with respect to their milkweed (Asclepias spp.) host plants. Given that monarchs largely depend on the genus Asclepias as larval host plants, the effects of climate change on monarch northward migrations will most likely be mediated by climate change effects on Asclepias. Here, I used MaxEnt species distribution modeling to assess potential changes in Asclepias and monarch distributions under moderate and severe climate change scenarios. First, Asclepias distributions were projected to extend northward throughout much of Canada despite considerable variability in the environmental drivers of each individual species. Second, Asclepias distributions were an important predictor of current monarch distributions, indicating that monarchs may be constrained as much by the availability of Asclepias host plants as environmental variables per se. Accordingly, modeling future distributions of monarchs, and indeed any tightly coupled plant-insect system, should incorporate the effects of climate change on host plant distributions. Finally, MaxEnt predictions of Asclepias and monarch distributions were remarkably consistent among general circulation models. Nearly all models predicted that the current monarch summer breeding range will become slightly less suitable for Asclepias and monarchs in the future. Asclepias, and consequently monarchs, should therefore undergo expanded northern range limits in summer months while encountering reduced habitat suitability throughout the northern migration.     

Environmental change,shifting distributions,and habitat conservation plans: A case study of the California gnatcatcher

Many species have already experienced distributional shifts due to changing environmental conditions, and analyzing past shifts can help us to understand the influence of environmental stressors on a species as well as to analyze the effectiveness of conservation strategies. We aimed to (1) quantify regional habitat associations of the California gnatcatcher (Polioptila californica ); (2) describe changes in environmental variables and gnatcatcher distributions through time; (3) identify environmental drivers associated with habitat suitability changes; and (4) relate habitat suitability changes through time to habitat conservation plans. Southern California's Western Riverside County (WRC ), an approximately 4,675 km² conservation planning area. We assessed environmental correlates of distributional shifts of the federally threatened California gnatcatcher (hereafter, gnatcatcher) using partitioned Mahalanobis D ² niche modeling for three time periods: 1980–1997, 1998–2003, and 2004–2012, corresponding to distinct periods in habitat conservation planning. Highly suitable gnatcatcher habitat was consistently warmer and drier and occurred at a lower elevation than less suitable habitat and consistently had more CSS , less agriculture, and less chaparral. However, its relationship to development changed among periods, mainly due to the rapid change in this variable. Likewise, other aspects of highly suitable habitat changed among time periods, which became cooler and higher in elevation. The gnatcatcher lost 11.7% and 40.6% of highly suitable habitat within WRC between 1980–1997 to 1998–2003, and 1998–2003 to 2004–2012, respectively. Unprotected landscapes lost relatively more suitable habitat (?64.3%) than protected landscapes (30.5%). Over the past four decades, suitable habitat loss within WRC , especially between the second and third time periods, was associated with temperature‐related factors coupled with landscape development across coastal sage scrub habitat; however, development appears to be driving change more rapidly than climate change. Our study demonstrates the importance of providing protected lands for potential suitable habitat in future scenarios.     

气候变化下寒露林蛙在中国的潜在地理分布

近几十年来,全球变暖对全球生物多样性及其地理分布产生了重要影响,特别是对气候变化敏感的两栖动物。寒露林蛙(Rana hanluica)是中国特有种,但在濒危物种红色名录中处于无危状态。为了评估寒露林蛙种群的生存现状,掌握该物种在中国的潜在分布区,以及在未来气候变化条件下适宜生境区的变化,本研究利用最大熵(MaxEnt)生态位模型和地理信息系统,对中国未来气候变化情景下(2050和2070年)寒露林蛙的适宜生境区进行识别。基于47个寒露林蛙分布位点和20个典型环境因子,建立了寒露林蛙在当前和未来气候条件下的适宜生境模型,并分析了相关的环境影响因子。结果表明: MaxEnt模型的预测准确度较高,受试者工作曲线面积值达0.993;寒露林蛙在当前气候条件下的潜在适宜生境面积为36.36万km²,潜在地理分布区域主要位于湖南省和贵州省;影响潜在地理分布的主要环境因子为最干月降水量和海拔。在未来2种典型浓度路径的气候情景下(SSP1-2.5和SSP5-8.5),寒露林蛙适宜生境区均出现不同程度的缩减,导致总适宜生境面积呈减少趋势;其高适宜生境向高纬度地区转移,其核心分布区仍以湖南省为主。     

基于最大熵模型的不同尺度物种分布概率优化热点分析: 以红色木莲为例

单一空间尺度构建的最大熵(maximum entropy, MaxEnt)模型是否具有代表性, 是MaxEnt模型应用与发展中面临的重要问题。本研究基于有效的地理分布位点数据, 利用最小凸多边形法(the minimum convex polygon method)在三江并流、云南省及全国3个空间尺度下分别识别了红色木莲(Manglietia insignis)的建模区域, 并进一步建立MaxEnt模型: 使用ROC曲线分析法与遗漏率(omission rate, OR)检验评估MaxEnt模型预测精度; 基于ArcGIS分析分布概率及其热点区域的分布趋势, 并通过分区统计工具Zonal识别潜在适宜分布区域的质心位置; 采用刀切法检验环境因子贡献率。结果表明: (1)不同尺度下红色木莲的MaxEnt模型都有良好的预测效果, 三江并流、云南省及全国尺度下的AUC值分别为0.936、0.887和0.930, OR值分别为0.18、0.15和0.20; (2)各尺度红色木莲的适生区格局呈现一致性分布趋势, 集中在独龙江、怒江和澜沧江3个流域; (3) 3个空间尺度下红色木莲的地理分布受不同环境因子影响, 存在着尺度依赖效应。由此可见, 红色木莲在不同空间尺度下的预测模型有着稳定的性能表现与良好的预测效果。此外, 我们建议在野外实地调查与野生生物资源保护中加强对普通物种的关注, 在预测物种地理分布的研究中将MaxEnt模型与热点分析结合使用。     

基于MaxEnt模拟人类活动与环境因子对青藏高原特有植物祁连獐牙菜潜在分布的影响

以青藏高原特有植物祁连獐牙菜(Swertia przewalskii Pissjauk.)为材料,基于该物种18个种群分布点及8个生物气候变量、海拔变量以及人类活动强度变量,运用最大熵模型(Max Ent)和ArcGIS技术分别构建当前气候情景下及人类活动影响下祁连獐牙菜的适宜生境预测模型,研究人类活动及自然环境变量对祁连獐牙菜空间分布的影响。结果显示,人类活动影响下的训练集和测试集的AUC值均小于无人类活动干扰的AUC值,人类活动与祁连獐牙菜分布呈负相关。限制祁连獐牙菜分布的主要变量为海拔、等温性、人类活动足迹指数及平均温度日较差。当前气候情景下祁连獐牙菜的最适宜生境占祁连山国家公园青海片区总面积的36.6%,有利于该物种的保护和恢复,而位于门源县和祁连县保护区内一般控制区的潜在生境受到人为干扰的可能性较大,应加强关注和保护。     

A practical guide to MaxEnt for modeling species’ distributions: what it does,and why inputs and settings matter

The MaxEnt software package is one of the most popular tools for species distribution and environmental niche modeling, with over 1000 published applications since 2006. Its popularity is likely for two reasons: 1) MaxEnt typically outperforms other methods based on predictive accuracy and 2) the software is particularly easy to use. MaxEnt users must make a number of decisions about how they should select their input data and choose from a wide variety of settings in the software package to build models from these data. The underlying basis for making these decisions is unclear in many studies, and default settings are apparently chosen, even though alternative settings are often more appropriate. In this paper, we provide a detailed explanation of how MaxEnt works and a prospectus on modeling options to enable users to make informed decisions when preparing data, choosing settings and interpreting output. We explain how the choice of background samples reflects prior assumptions, how nonlinear functions of environmental variables (features) are created and selected, how to account for environmentally biased sampling, the interpretation of the various types of model output and the challenges for model evaluation. We demonstrate MaxEnt’s calculations using both simplified simulated data and occurrence data from South Africa on species of the flowering plant family Proteaceae. Throughout, we show how MaxEnt’s outputs vary in response to different settings to highlight the need for making biologically motivated modeling decisions.     

基于MaxEnt模型的甘肃安西极旱荒漠国家级自然保护区北山羊生境评估

明晰甘肃安西极旱荒漠国家级自然保护区珍稀濒危物种北山羊的分布格局,并阐释气候变化和人类活动对北山羊的影响,对今后北山羊生境管理和物种保护具有重要意义。基于北山羊实测分布点记录和环境变量数据,结合MaxEnt模型和ArcGIS空间分析功能,利用CMIP6的8个气候模式均值预测中度发展路径（SSP2-4.5）下,基准期（1970-2000年）和未来气候（2041-2060年、2081-2100年）变化情景下,甘肃安西极旱荒漠国家级自然保护区北山羊的潜在适生区分布范围及变化,并综合贡献率和置换重要性值对北山羊生境选择关键环境因子进行了分析。研究结果表明：（1） MaxEnt模型的预测精度较高,三种气候条件下ROC曲线下面积（AUC,Area Under Curve）>0.97,且真实技巧统计（TSS,True Skill Statistic）>0.90,模拟结果可靠。（2）影响北山羊生境选择的主要环境因子为气候条件（降水量季节性变异系数和等温性）、海拔和人为干扰（距泉和居名点距离）。水是保护区北山羊生存的最基本要素,气候条件共同控制北山羊生境条件。此外,北山羊习性决定其生境宜选择高海拔和远离人类活动影响地区。（3）基准期保护区北山羊主要分布在北片和南片高海拔山区,面积365.77 km²（占整个保护区的4.31%）,北山羊适生区面积北片>南片、中高等适生区主要位于保护区北片。（4） CMIP6未来气候变化情景下,随着保护区生态环境改善和濒危物种保护措施的实施,北山羊潜在适生区面积呈增加趋势,但是受北山羊近亲繁殖的影响,整体上北山羊数量和适生区面积增加并不显著且有向南部及高海拔地区转移趋势。     

Machine learning of large‐scale spatial distributions of wild turkeys with high‐dimensional environmental data

Species distribution modeling often involves high‐dimensional environmental data. Large amounts of data and multicollinearity among covariates impose challenges to statistical models in variable selection for reliable inferences of the effects of environmental factors on the spatial distribution of species. Few studies have evaluated and compared the performance of multiple machine learning (ML) models in handling multicollinearity. Here, we assessed the effectiveness of removal of correlated covariates and regularization to cope with multicollinearity in ML models for habitat suitability. Three machine learning algorithms maximum entropy (MaxEnt), random forests (RFs), and support vector machines (SVMs) were applied to the original data (OD) of 27 landscape variables, reduced data (RD) with 14 highly correlated covariates being removed, and 15 principal components (PC) of the OD accounting for 90% of the original variability. The performance of the three ML models was measured with the area under the curve and continuous Boyce index. We collected 663 nonduplicated presence locations of Eastern wild turkeys (Meleagris gallopavo silvestris) across the state of Mississippi, United States. Of the total locations, 453 locations separated by a distance of ≥2 km were used to train the three ML algorithms on the OD, RD, and PC data, respectively. The remaining 210 locations were used to validate the trained ML models to measure ML performance. Three ML models had excellent performance on the RD and PC data. MaxEnt and SVMs had good performance on the OD data, indicating the adequacy of regularization of the default setting for multicollinearity. Weak learning of RFs through bagging appeared to alleviate multicollinearity and resulted in excellent performance on the OD data. Regularization of ML algorithms may help exploratory studies of the effects of environmental factors on the spatial distribution and habitat suitability of wildlife.     

Prediction of potential distribution of soybean in the frigid region in China with MaxEnt modeling

Soybean (Glycine max (L.) Merr.) is one of the most important grains and oil-producing plants grown in China. Understanding the potential suitable characteristics of areas where soybean is grown and predicting its potential habitat under different climate scenarios are a significant part of ensuring food security. This study compiled 65 occurrence locations of soybean and 32 environmental variables obtained from the WorldClim database. Nine environmental variables were selected for model training. We identified potential suitable distribution areas for soybean in the frigid region and predicted changes in its geographical distribution under four shared socioeconomic pathways, SSP1–2.6, SSP2–4.5, SSP3–7.0, and SSP5–8.5, for the periods from 2021 to 2040, 2041 to 2060, 2061 to 2080, and 2081 to 2100 using the MaxEnt model. The results showed that annual mean temperature, elevation, and April solar radiation were the dominant factors affecting the distribution of soybean, contributing 48.8%, 17.9%, and 15.7% of the variability in the data, respectively. Highly suitable habitats (defined as having a suitability variable P of 0.66–1.0) for the current conditions included the Songnen and Sanjiang plains, covering about 2.36 × 10⁵ km². The total areas of highly (as defined above) and moderately suitable (0.33–0.66) habitats would be reduced under the four climate scenarios. However, the centroids of the highly suitable habitat had a small mobile range under different scenarios. These results along with previous research on the potential distribution of soybean offer useful information; ecological modeling approaches need to be considered in future crop planting management and land use.     

The fate of European breeding birds under climate,land‐use and dispersal scenarios

Many species have already shifted their distributions in response to recent climate change. Here, we aimed at predicting the future breeding distributions of European birds under climate, land‐use, and dispersal scenarios. We predicted current and future distributions of 409 species within an ensemble forecast framework using seven species distribution models (SDMs), five climate scenarios and three emission and land‐use scenarios. We then compared results from SDMs using climate‐only variables, habitat‐only variables or both climate and habitat variables. In order to account for a species’ dispersal abilities, we used natal dispersal estimates and developed a probabilistic method that produced a dispersal scenario intermediate between the null and full dispersal scenarios generally considered in such studies. We then compared results from all scenarios in terms of future predicted range changes, range shifts, and variations in species richness. Modeling accuracy was better with climate‐only variables than with habitat‐only variables, and better with both climate and habitat variables. Habitat models predicted smaller range shifts and smaller variations in range size and species richness than climate models. Using both climate and habitat variables, it was predicted that the range of 71% of the species would decrease by 2050, with a 335 km median shift. Predicted variations in species richness showed large decreases in the southern regions of Europe, as well as increases, mainly in Scandinavia and northern Russia. The partial dispersal scenario was significantly different from the full dispersal scenario for 25% of the species, resulting in the local reduction of the future predicted species richness of up to 10%. We concluded that the breeding range of most European birds will decrease in spite of dispersal abilities close to a full dispersal hypothesis, and that given the contrasted predictions obtained when modeling climate change only and land‐use change only, both scenarios must be taken into consideration.