本数据集包含构建北山羊（Siberian ibex）物种分布模型后得出的当前分布范围和未来2071-2100年三种情景下的分布范围。\nCurrentEM_BiasCorr_Simple ：当前适宜生境概率图。\nFuture126EM2070：北山羊未来SSP1-2.6情景下的适宜生境概率图。\nFuture370EM2070：北山羊未来SSP3-7.0情景下的适宜生境概率图。\nFuture585EM2070：北山羊未来SSP5-8.5情景下的适宜生境概率图。\n
", "ds_source": "野外调查和文献资料库检索。
", "ds_process_way": "为构建北山羊物种分布模型，首先系统收集了物种出现点位数据，这些数据代表物种在空间上的实际观测记录。同时，整合了与物种生态需求相关的环境变量图层（如气温、降水、海拔等栅格型气候与地形因子），作为驱动分布模型的关键预测因子。基于上述数据，通过以下流程生成物种当前分布范围：针对物种真实缺失数据难以获取的问题，采用最小凸多边形（MCP）外扩200公里缓冲区界定地理背景范围，并利用高斯核密度表面量化采样偏差，在该表面约束下生成10,000个伪缺失点，使其空间分布与实际调查努力相匹配，从而规避人为活动导致的虚假环境关联。将物种存在点与伪缺失点合并后，按75%:25%划分为训练集与测试集，运用七种算法（GLM、GBM、GAM、ANN、FDA、MARS、RF）各运行十次，产生70个单模型；筛选其中真实技巧统计值（TSS）高于整体均值的模型，按其TSS权重构建加权集成模型，并通过AUC值（>0.9为优）验证模型可靠性。\n北山羊未来分布范围预测是通过将训练好的集成模型投射至特定气候情景（如CMIP6下2071-2100年的SSP1-2.6，SSP3-7.0和SSP5-8.5），采用与当前建模相同的环境变量，将未来适生概率转换为可比的空间分布图。该方法直接量化物种分布的核心动态：栖息地收缩区（当前存在而未来消失）、扩张区（当前缺失而未来新增）以及避难所（持续适生区），同时揭示分布质心向高纬度/高海拔的迁移趋势，为识别濒危区域、设计生态廊道及制定保护策略提供空间决策依据。\n
", "ds_quality": "", "ds_acq_start_time": null, "ds_acq_end_time": null, "ds_acq_place": "", "ds_acq_lon_east": null, "ds_acq_lat_south": null, "ds_acq_lon_west": null, "ds_acq_lat_north": null, "ds_acq_alt_low": null, "ds_acq_alt_high": null, "ds_share_type": "apply-access", "ds_total_size": 29742115, "ds_files_count": 12, "ds_format": "", "ds_space_res": "1000米", "ds_time_res": "年", "ds_coordinate": "WGS84", "ds_projection": "GCS_WGS84", "ds_thumbnail": "f0962369-3454-4144-993a-da8a2edbde8b.jpg", "ds_thumb_from": 0, "ds_ref_way": "", "paper_ref_way": "", "ds_ref_instruction": "", "ds_from_station": null, "organization_id": "a5877b42-96ea-4f13-af7e-246f355413d6", "doi_value": "", "subject_codes": [ "180.57" ], "quality_level": 1, "publish_time": "2025-08-13 19:18:28", "first_publish_time": null, "last_updated": "2025-12-08 16:30:20", "protected": false, "protected_to": null, "lang": "zh", "cstr": "33110.11.ariddc.01212", "license": null, "extra": null, "files_shape": [ { "name": "当前和未来三种情景下北山羊分布数据", "size": null, "is_dir": true } ], "features": null, "data_level": 0, "i18n": { "en": { "title": "Current and future distribution data of northern goats", "ds_abstract": "This dataset includes the current distribution range obtained after constructing the Siberian ibex species distribution model and the distribution range under three scenarios in the future from 2071 to 2100.\nCurrentEM_BiasCorr_Simple: Probability graph of current suitable habitats.\nFuture126 EM2070: Probability map of suitable habitat for northern goats under future SSP1 -2.6 scenarios.\nFuture370 EM2070: Probability map of suitable habitat for northern goats under future SPSP3 -7.0 scenarios.\nFuture585 EM2070: Probability map of suitable habitat for northern goats under future SPS5 -8.5 scenarios.\n
", "ds_source": "Field survey and literature database search.
", "ds_process_way": "In order to build a species distribution model of the Northern Goat, data on the occurrence points of species were first systematically collected, and these data represented the actual observation records of the species in space. At the same time, layers of environmental variables related to the ecological needs of species (such as grid climate and terrain factors such as temperature, precipitation, and altitude) are integrated as key predictive factors driving the distribution model. Based on the above data, the current distribution range of the species is generated through the following process: In view of the difficulty in obtaining true missing data of species, the minimum convex polygon (MCP) is used to extend the 200-kilometer buffer zone to define the geographical background range, and the Gaussian kernel density surface is used to quantify sampling deviations. Under the constraint of this surface, 10,000 pseudo-missing points are generated, matching their spatial distribution with actual investigation efforts, thereby avoiding false environmental associations caused by human activities. After combining species presence points and pseudo-missing points, they were divided into training set and test set by 75%:25%. Seven algorithms (GLM, GBM, GAM, ANN, FDA, MARS, RF) were run ten times each to generate 70 single models; Models with true skill statistics (TSS) higher than the overall mean were screened, a weighted integrated model was constructed based on their TSS weights, and the reliability of the model was verified by AUC values (>0.9 is excellent).\nPrediction of the future distribution range of northern goats is carried out by projecting the trained integrated model to specific climate scenarios (such as SSP1-2.6, SSP3-7.0 and SSP5-8.5 for 2071-2100 under CMIP6), using the same environmental variables as the current modeling, and transforming the probability of future fitness into a comparable spatial distribution map. This method directly quantifies the core dynamics of species distribution: habitat contraction areas (currently existing but disappearing in the future), expansion areas (currently missing but newly added in the future), and shelters (continuously suitable areas), while revealing the distribution centroid to high latitudes/The migration trend of high altitude provides spatial decision-making basis for identifying endangered areas, designing ecological corridors and formulating protection strategies.\n
", "ds_projection": "GCS_WGS84", "ds_space_res": "1000 meters", "ds_time_res": "years" } }, "license_type": null, "doi_reg_from": "reg_local", "cstr_reg_from": "reg_local", "doi_not_reg_reason": null, "cstr_not_reg_reason": null, "is_paper_in_submitting": false, "ds_topic_tags": [ "分布预测", "模型", "物种" ], "ds_subject_tags": [ "动物学" ], "ds_class_tags": [], "ds_locus_tags": [ "中亚" ], "ds_time_tags": [ 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018, 2019, 2020, 2021, 2022, 2023 ], "ds_contributors": [ "徐靖雯" ], "ds_meta_authors": [ "徐靖雯" ], "ds_managers": [ "徐靖雯" ], "category": "生物多样性" }