Modelling amphibian road crossing points in a dynamic environment

Roads are ubiquitous infrastructures that have detrimental effects on wildlife and contribute to increased mortality and fragmentation of animal populations. Although several mitigation measures are available to reduce road impacts, their planning rarely considers the dynamic nature of the environment, which may reflect into temporal variation in habitat suitability and connectivity for animal species. Consequently, the effectiveness of such measures may fall short of expectations. By combining high-resolution satellite imagery and connectivity modelling, we propose a generalizable approach to identify the most probable crossing points across a barrier at different time snapshots. This information may be pivotal in planning mitigation measures that can take into account dynamic components. We collected occurrence data of three farmland-adapted anurans and high spatiotemporal definition Sentinel-2 multispectral images to build habitat suitability models that capture the relationship betw..., , # Modelling amphibian road crossing points in a dynamic environment

Dataset DOI: [10.5061/dryad.cz8w9gjg2](10.5061/dryad.cz8w9gjg2)

## Description of the data and file structure

Reported data were the basis for Habitat Suitability Modelling and subsequent Connectivity Models.

Data includes the occurrence data of the three amphibian target species (*Hyla intermedia*, *Pelophylax* synkl. *esculentus*, *Bufotes viridis*), collected in 2021 and 2022. The values of the Sentinel-2 spectral bands we used as predictors are Copernicus Sentinel data [2025] and can be downloaded from the Copernicus Browser ([https://browser.dataspace.copernicus.eu/](https://browser.dataspace.copernicus.eu/)).

### Files and variables

#### File: occurencedata.csv

**Description:** 

Occurrence points of the three target species. Observations missing important information have already been removed from this dataset. Here you can find the data we used for all the modelling steps. They are the occurrence data of ...,

# 动态环境下两栖动物道路穿越点建模

道路作为无处不在的基础设施，会对野生动物造成不利影响，并加剧动物种群的死亡事件与栖息地破碎化问题。尽管现有多种可降低道路影响的减缓措施，但其规划极少考虑环境的动态特性，而这会体现在动物物种栖息地适宜性与连通性的时间维度变化上。因此，此类措施的实施效果往往不尽如人意。本研究结合高分辨率卫星影像与连通性建模方法，提出了一种可推广的研究路径，用于识别不同时间快照下障碍物区域最可能的动物穿越点位。该信息可为纳入动态要素的减缓措施规划提供关键支撑。本研究采集了三种适应农田生境的无尾目两栖动物的出现点数据，以及高时空分辨率的哨兵二号（Sentinel-2）多光谱影像，用于构建能够捕捉相关生态关联的栖息地适宜性模型……

数据集DOI：10.5061/dryad.cz8w9gjg2

## 数据与文件结构说明

本研究的栖息地适宜性建模及后续连通性建模均以本次报告的数据为基础。

本次数据集包含2021年与2022年采集的三种目标两栖动物出现点数据，分别为意大利树蛙（*Hyla intermedia*）、可繁殖型林蛙复合群（*Pelophylax* synkl. *esculentus*）以及绿蟾蜍（*Bufotes viridis*）。本研究用作预测变量的哨兵二号光谱波段数据源自2025年哥白尼哨兵（Copernicus Sentinel）数据集，可通过哥白尼浏览器（https://browser.dataspace.copernicus.eu/）下载获取。

### 文件与变量

#### 文件：occurencedata.csv

**文件说明：**
该文件收录了三种目标物种的出现点数据，所有缺失关键信息的观测样本均已从本数据集中剔除。此处包含的即为我们用于全部建模流程的相关数据，即……