Evaluation of species distribution models by resampling of sites surveyed a century ago by Joseph Grinnell
Evaluation of species distribution models by resampling of sites surveyed a century ago by Joseph Grinnell
复制标题
通过对一个世纪前约瑟夫·格林内尔调查的地点进行重新采样来评估物种分布模型
DOI：
10.1111/j.1600-0587.2013.00107.x
复制
发表时间：
2013
期刊：
影响因子：
5.9
通讯作者：
C. Moritz
中科院分区：
文献类型：
--
作者：
Adam B. Smith;Maria J. Santos;M. Koo;K. Rowe;K. Rowe;J. Patton;J. Perrine;S. Beissinger;C. Moritz
生态志36：1017-1031,2013DOI：10.1111/j.1600-0587.2013.00107.x©2013作者。生态学©2013北欧学会Oikos主题编辑：David Nogue-Bravo。接受2013年1月25日由约瑟夫·格林内尔·亚当·B·史密斯、玛丽亚·J·桑托斯、米歇尔·S·古永锵、凯伦·M·C·罗、凯文·C·罗、詹姆斯·L·巴顿、约翰·D·佩林、史蒂文·R·贝辛格和克雷格·莫里茨·A·B·史密斯(Adam@adamlilith.net)、古永锵、J·L·巴顿、S·R·贝辛格和C·莫里茨一个世纪前调查的地点的物种分布模型评估。加利福尼亚州伯克利，邮编：94720-3060.目前地址：密苏里州植物园保护与可持续发展中心，密苏里州63166，圣路易斯，邮政信箱299号。SRB还在：环境科学、政策和管理系，130马尔福德大厅，大学。加利福尼亚州伯克利，邮编：94720-3114.-M·J·桑托斯，美国西部空间历史项目和比尔·莱恩中心，斯坦福大学历史系，加利福尼亚州斯坦福，94305-2055年。-K·M·C·罗和K·C·罗，4科学部，维多利亚博物馆，GPO Box 666，墨尔本3001，澳大利亚VIC。-J.D.佩林，加州理工大学生物科学系，圣路易斯奥比斯波，加利福尼亚州93407-0401.物种分布模型(SDMS)通常用于预测物种对预期的全球变化的反应，但由于缺乏来自未来时间段的数据，因此无法评估其可靠性。相反，针对同一时代的测试数据的性能被认为与未来的准确性相关。此外，高概率的fi缺失数据是测试模型准确性所必需的，但通常是不可用的，因为一个物种可能在未被检测到时存在。在这里，我们评估了用历史(1900-1939年)或现代(1970-2009年)气候数据和18个哺乳动物物种的发生记录训练的8个SDMS的表现。模型被预测到相同或相反的时间段，并使用约瑟夫·格林内尔及其同事从1900年至1939年在加利福尼亚州内华达山脉进行的调查以及从2003年至2011年进行的现代再调查中获得的数据进行评估。占用模型被用来持续地分配fi在未检测到物种的试验点的缺勤。用物种的存在结合这个高fi缺失(Hca)集合、一个低fi缺失集合(其中未检测到表示缺失(LCA))和随机定位的“假缺失”(PSA)来评估SDMS。模型绩效随缺席质量的增加而显著提高(平均AUCSE：PSA为0.760.01，LCA为0.790.01，HCA为0.810.01)，SDM之间的表观fi差异随缺席质量的增加而减小。当使用与阈值无关的指标进行评估时，跨时间投影的模型的表现与在相同时间段内投影的模型一样好。然而，在跨时代的预测中，存在和缺席预测的准确性有时会下降。尽管性能上的大部分差异发生在物种之间，但自身性状与模型准确性的相关性很弱。我们的研究表明，a)ff评估数据的质量影响模型性能的评估；b)时代内的性能与跨时代的性能正相关，但不可靠；c)SDM可以可靠但谨慎地跨时间预测。人为的气候变化有望改写地球上S物种的生物地理，一些物种预计会受益，一些物种会失去，一些物种将改变它们目前的分布。因此，保护规划者需要可靠的方法来预测受关注物种的未来分布并优先保护ff或(Th OMAS等人。2004年，Carroll等人。2010年，Ogawa-Onishi等人。2010年，Saupe等人。2011年)。物种分布模型(SDMS)经常用于此类预测，该模型将物种出现数据与气候变量和其他指示生境质量的因素相关联，以生成环境适宜性地图。不幸的是，预测与保护相关的时间段的SDMS的可靠性在很大程度上仍然未知(Ara u Jo等人)。2005a，b，Dormann，2007，Elith和Leathwick，2009，Kharouba等人。2009年)。数十项研究使用时代内评估来评估不同ff的SDM算法的性能，并针对用于训练模型的相同地区和时间段的记录来测试模型(Elith等人。2006年，希曼斯和格雷厄姆2006年，西弗和富兰克林2009年)。然而，对SDMS的时代内评估可能会给出对跨时代表现的过于乐观的估计(Ara u Jo等人。2005a，希曼斯2012年)。虽然跨时代评估增加了培训和测试数据之间的独立性，但它需要来自两个感兴趣的时间段的数据，而这些数据很少用于与保护规划相关的时间跨度(即几十年或更长时间)。由于与生物学和建模有关的原因，在跨时间进行投影时，SDMS的可靠性应该低于同一时代(Ara u Jo等人)。2005a，b，Dobrowski等人。2011年)。从生物学的角度来看，如果物种分布不符合
Ecography 36: 1017–1031, 2013 doi: 10.1111/j.1600-0587.2013.00107.x © 2013 Th e Authors. Ecography © 2013 Nordic Society Oikos Subject Editor: David Nogues-Bravo. Accepted 25 January 2013 Evaluation of species distribution models by resampling of sites surveyed a century ago by Joseph Grinnell Adam B. Smith , Maria J. Santos , Michelle S. Koo , Karen M. C. Rowe , Kevin C. Rowe , James L. Patton , John D. Perrine , Steven R. Beissinger and Craig Moritz A. B. Smith (adam@adamlilith.net), M. S. Koo, J. L. Patton, S. R. Beissinger and C. Moritz, Museum of Vertebrate Zoology, 3101 Valley Life Sciences Building, Univ. of California, Berkeley, CA 94720-3060, USA. Present address of ABS: Center for Conservation and Sustainable Development, Missouri Botanical Garden, PO Box 299, Saint Louis, MO 63166, USA. SRB also at: Dept of Environmental Science, Policy and Management, 130 Mulford Hall, Univ. of California, Berkeley, CA 94720-3114, USA. – M. J. Santos, Spatial History Project and Bill Lane Center for the American West, History Dept, Stanford Univ., Stanford, CA 94305-2055, USA. – K. M. C. Rowe and K. C. Rowe, 4 Sciences Dept, Museum Victoria, GPO Box 666, Melbourne 3001, VIC, Australia. – J. D. Perrine, Biological Sciences Dept, California Polytech- nic State Univ., San Luis Obispo, CA 93407-0401, USA. Species distribution models (SDMs) are commonly applied to predict species ’ responses to anticipated global change, but lack of data from future time periods precludes assessment of their reliability. Instead, performance against test data in the same era is assumed to correlate with accuracy in the future. Moreover, high-confi dence absence data is required for testing model accuracy but is often unavailable since a species may be present when undetected. Here we evaluate the performance of eight SDMs trained with historic (1900 – 1939) or modern (1970 – 2009) climate data and occurrence records for 18 mammalian species. Models were projected to the same or the opposing time period and evaluated with data obtained from surveys conducted by Joseph Grinnell and his colleagues in the Sierra Nevada of California from 1900 to 1939 and modern resurveys from 2003 to 2011. Occupancy modeling was used to confi dently assign absences at test sites where species were undetected. SDMs were evaluated using species ’ presences combined with this high-confi dence absence (HCA) set, a low-confi dence set in which non-detections were assumed to indicate absence (LCA), and ran- domly located ‘ pseudoabsences ’ (PSA). Model performance increased signifi cantly with the quality of absences (mean AUC SE: 0.76 0.01 for PSA, 0.79 0.01 for LCA, and 0.81 0.01 for HCA), and apparent diff erences between SDMs declined as the quality of test absences increased. Models projecting across time performed as well as when pro- jecting within the same time period when assessed with threshold-independent metrics. However, accuracy of presence and absence predictions sometimes declined in cross-era projections. Although most variation in performance occurred among species, autecological traits were only weakly correlated with model accuracy. Our study indicates that a) the quality of evaluation data aff ects assessments of model performance; b) within-era performance correlates positively but unreliably with cross-era performance; and c) SDMs can be reliably but cautiously projected across time. Anthropogenic climate change promises to rewrite the bio- geography of Earth ’ s species, with some expected to gain, some to lose, and some to shift their current distributions. As a result, conservation planners require reliable methods to project future distributions of species of concern and to prioritize conservation eff ort (Th omas et al. 2004, Carroll et al. 2010, Ogawa-Onishi et al. 2010, Saupe et al. 2011). Species distribution models (SDMs), which correlate species occurrence data with climate variables and other factors indicative of habitat quality to produce maps of environ- mental suitability, are frequently used for such projections. Unfortunately, the reliability of projecting SDMs across time periods relevant to conservation remains largely unknown (Ara u jo et al. 2005a, b, Dormann 2007, Elith and Leathwick 2009, Kharouba et al. 2009). Scores of studies have assessed the performance of diff erent SDM algorithms using within-era evaluation, testing models against records from the same region and time period used to train the models (Elith et al. 2006, Hijmans and Graham 2006, Syphard and Franklin 2009). However, within-era assessments of SDMs may give overly optimistic estimates of cross-era performance (Ara u jo et al. 2005a, Hijmans 2012). While cross-era evaluation increases the independence between training and test data, it requires data from both time periods of interest, which are rarely available for time spans relevant to conservation planning (i.e. several decades or more). SDMs should be less reliable when projecting across time than within the same era for reasons related to both biology and modeling (Ara u jo et al. 2005a, b, Dobrowski et al. 2011). From a biological perspective model perfor- mance will be diminished if species distributions are not in