Abstract Amphibians are one of the most vulnerable taxa at risk of rapid decline under climate change. Here, we evaluated the impact of different migration constraints on projected future distributions of four high elevation frogs, belonging to the genus Scutiger, in the Eastern Himalaya. We explored differences between the output of conventional models assuming no or unlimited migration versus models considering plausible migration rates to ascertain future species distributions under climate change. Distributions of the four Scutiger species, namely S. boulengeri, S. glandulatus, S. sikimmesis and S. tuberculatus, based on field data and other sources were modelled using MaxEnt and projected for three future time periods (2021–2040; 2041–2060; 2061–2080) under the relatively ambitious RCP4.5 and the more pessimistic RCP8.5 climate change scenarios using three global circulation models. Projected species distributions were compared at different spatial resolutions (1 km, 5 km and 10 km) and for five assumptions about species migration: (1) no migration; (2–4) low, medium and high migration abilities using the KISSMig model; and (5) unlimited migration. Without migration, the projected future distribution of all four species showed a significant decrease of −15% to −64% by 2080. In contrast, three out of the four study species were projected to expand their distribution under unlimited migration scenarios. Models with more realistic migration rates, however, demonstrated considerable deviance from both no migration and unlimited migration scenarios. These results were consistent across models with different spatial resolutions. Our study shows that ignoring realistic migration constraints can lead to ineffective conservation measures by overestimating the future distribution of Himalayan amphibians. The proposed framework can be used to project more realistic ranges of future species distributions by considering the accessibility of future suitable areas, a key factor for species persistence under climate change.

中文翻译：

---

有限迁徙的直接建模改善了气候变化下喜马拉雅两栖动物的预测分布

摘要 两栖动物是气候变化下面临快速衰退风险的最脆弱的类群之一。在这里，我们评估了不同迁徙限制对喜马拉雅东部四只属于 Scutiger 属的高海拔青蛙的预计未来分布的影响。我们探讨了假设没有或无限迁移的传统模型的输出与考虑合理迁移率以确定气候变化下未来物种分布的模型之间的差异。四种 Scutiger 物种的分布，即 S. boulengeri、S.glutenulatus、S. sikimmesis 和 S.tuberculatus，基于现场数据和其他来源，使用 MaxEnt 建模，并预测未来三个时间段（2021-2040 年；2041-2060 年） ; 2061-2080）在相对雄心勃勃的 RCP4.5 和更悲观的 RCP8 下。使用三种全球环流模型的 5 种气候变化情景。在不同的空间分辨率（1 公里、5 公里和 10 公里）和关于物种迁移的五个假设下比较了预测的物种分布：(1) 没有迁移；(2-4) 使用 KISSMig 模型的低、中和高迁移能力；(5) 无限迁移。如果没有迁徙，到 2080 年，所有四种物种的预计未来分布将显着减少 -15% 至 -64%。相比之下，在无限迁徙情景下，预计四分之三的研究物种将扩大其分布。然而，具有更现实迁移率的模型显示出与无迁移和无限迁移场景的相当大的偏差。这些结果在具有不同空间分辨率的模型之间是一致的。我们的研究表明，通过高估喜马拉雅两栖动物的未来分布，忽视现实的迁徙限制可能导致保护措施无效。通过考虑未来适宜地区的可达性，拟议的框架可用于预测未来物种分布的更现实范围，这是气候变化下物种持久性的关键因素。