Do bats use guano and urine stains to find new roosts? Tests with three group-living bats

These data were collected for a study looking at whether three species of bats (Desmodus rotundus, Eptesicus fuscus, and Molossus molossus) use the scent from guano and urine to locate new roosts. The project included 8 different experiments (2 field experiments and 6 captive experiments). The two field experiments were done in Ohio testingEptesicus fuscus and in Panama testing Molossus molossus. The experiment involved placing paired artificial roosts in the field where one was treated with guano and urine and the other served as a control. Each artificial roost was outfitted with a Wildlife Acoustics SM4+ bat device and microphone that recorded any bat calls near the entrance of the roost. We could determine approximately how close the bats were to the roost based on how loud the calls were. The txt files labeled 20190714_Call_Analysis_Compilation_Molossus and 20190921_Call_Analysis_Compilation_Molossus are measures of the .wav files collected from the microphones in Panama. These measures were generated through Avisoft. Details on the parameters used in Avisoft can be found in the methods of the manuscript. The R file labeled 20200501_Molossus_FlyBy_Analysisis the R script used to analyze the data in the aforementioned txt files. The txt file labeled are measures of the .wav files collected from the microphones in Ohio. These measures were generated through Avisoft. Specifications for the parameters used in Avisoft to automatically processthe .wav files can be found in the associated paper. The original files that include details about all of the .wav files recorded on the microphones can be found in the following .txt files (audubon.txt, blacklick.txt, highbanksone.txt, highbankstwo.txt, highbanksthree.txt, highbanksfour.txt, highbanksfive.txt, pickerington.txt,sciotogrove.txt, slaterun.txt, and walnutwoods.txt). Each of these files is all of the data collected from one of the seven sites in Ohio.The R script labeled 20200501_Eptesicus_analysis was also used to analyze the data in the txt files for the Ohio portion of the project. This R script also created a .csv file labeled epfu2. The epfu2.csv file is a portion of the txt file that only includes .wav files that were likely to include Eptesicus fuscus calls, as opposed to all bat calls recorded on the microphones. The six captive experiments were done with both Desmodus rotundus (5 studies) and Molossus molossus (1 study). Experiment1data.csv was collected from a project looking at whether 5 Desmodus rotundus from captivity would chooseto roost in a tube with or without guano and urine. These data include information on which bat was being tested, how many tubes there were to choose from (either 4 tubes with 3 unscented control tubes and 1 scented tube or 2 tubes with 1 unscented control tube and 1 scented tube), and which roost tube the bat chose. The Experiment2data.csv includes data collected from a project looking to see if 20 Desmodusrotundus from a zoo would choose to hang on a surface in a small cage with guano more than on one without.These data include information on which bat was being tested and how many times it was found on both the control and the scented side. The Experiment3data.csv includes data collected from a project similar to Experiment 2, but it was done with wild-caught Desmodus rotundus and it was filmed. The Experiment4adata.xlsx and Experiment4bdata.xlsx files were from another study similar to Experiments 2 and 3, but this data set includes which side the bat was on at certain intervals, which bat was being used, whether it was part of a short (1 hour) or long (4 hour) trial, and which repetition of the trial that particular bat was on. The Experiment6data.xlsx file is from a study with the same set-up as Experiment 4, but this one was done with Molossus molossus instead of Desmodus rotundus. Finally, there are two files associated with Experiment 5. The experiment5data.csv file includes data on vampire bats placed into a y-maze with one of three options. First, they might be choosing between a scented arm of the maze or an unscented arm of the maze. These are labeled scent under trial type. Second, they might be choosing between an arm of the maze with playback callsbeing emitted from it or one without those calls. These are labeled sound under trial type. Finally, they might have been choosing between an arm with both cues or an armwithout either cue. These are labeled both under trial type. Each bat was placed in all three trial types, and so the order in which they experienced these trial types is also recorded. We measured the amount of time they spent in each arm (duration) and the amount of time it took them to choose that arm (latency). This can also be used to calculate speed, or how fast they moved into the arm of their choice. The ymaze_results.csv is a printout of a permutation done with experiment 5 data. This work was approved by the University of Maryland Institutional Animal Care and Use Committee (Protocol R-10-63), the Smithsonian Tropical Research Institute Animal Care and Use Committee (#2015-0915-2018-A9, #2019-0501-2022), the Panamanian Ministry of the Environment (#SE/A-76-16, #SE/A-36-19), and the Columbus and Franklin County Metro Parks through an access permit.

本数据集采集自一项探究三种蝙蝠——普通吸血蝠（Desmodus rotundus）、大棕蝠（Eptesicus fuscus）与墨西哥犬吻蝠（Molossus molossus）——是否借助蝙蝠粪与尿液的气味定位新栖息场所的研究。本项目共包含8组实验：2项野外实验与6项圈养实验。
两项野外实验分别于俄亥俄州针对大棕蝠开展，以及在巴拿马针对墨西哥犬吻蝠开展。实验布设方案为在野外设置成对的人工栖息箱，其中一组施加蝙蝠粪与尿液气味处理，另一组作为空白对照。每处人工栖息箱均配备Wildlife Acoustics SM4+蝙蝠声学记录仪与麦克风，用于录制栖息箱入口附近的蝙蝠鸣叫声。研究人员可通过鸣叫声的响度，大致判断蝙蝠与栖息箱的距离。
名为20190714_Call_Analysis_Compilation_Molossus与20190921_Call_Analysis_Compilation_Molossus的文本文件，为巴拿马地区麦克风录制的.wav音频文件的量化分析结果，上述指标通过Avisoft软件生成，其参数设置细节可参见论文的方法部分。名为20200501_Molossus_FlyBy_Analysis的R脚本，用于对上述文本文件中的数据进行分析。
另有一批文本文件为俄亥俄州地区麦克风录制的.wav音频文件的量化分析结果，同样通过Avisoft软件生成，该软件用于自动处理.wav文件的参数规范可参见相关论文。包含俄亥俄州7个采样点全部录制数据的原始文件详情，可从以下文本文件获取：audubon.txt、blacklick.txt、highbanksone.txt、highbankstwo.txt、highbanksthree.txt、highbanksfour.txt、highbanksfive.txt、pickerington.txt、sciotogrove.txt、slaterun.txt及walnutwoods.txt。针对俄亥俄州项目的文本文件数据，可使用名为20200501_Eptesicus_analysis的R脚本开展分析，该脚本同时生成了epfu2.csv文件，该文件仅包含大概率含有大棕蝠鸣叫声的.wav文件数据，而非麦克风录制的所有蝙蝠鸣叫声数据。
6项圈养实验分别针对普通吸血蝠（5项研究）与墨西哥犬吻蝠（1项研究）开展：
1. Experiment1data.csv采集自一项探究5只圈养普通吸血蝠是否会选择带有蝙蝠粪与尿液气味的栖息管的实验，数据包含受试蝙蝠编号、可供选择的栖息管数量（分为两种设置：4管组含3支无气味对照管与1支气味处理管；2管组含1支无气味对照管与1支气味处理管）以及受试蝙蝠选择的栖息管信息。
2. Experiment2data.csv包含一项实验的数据：探究20只动物园来源的普通吸血蝠是否更倾向于停留在涂有蝙蝠粪气味的小型笼舍表面，数据包含受试蝙蝠编号、其在对照侧与气味侧停留的次数。
3. Experiment3data.csv采集自一项与实验2类似的研究，但受试对象为野生捕获的普通吸血蝠，且实验过程全程录像。
4. Experiment4adata.xlsx与Experiment4bdata.xlsx来自另一项与实验2、3类似的研究，该数据集包含以下信息：受试蝙蝠在特定时间间隔所处的一侧、受试蝙蝠编号、实验时长（短期1小时或长期4小时）以及该受试蝙蝠参与的实验重复次数。
5. Experiment6data.xlsx来自一项与实验4设置一致的研究，但受试对象更换为墨西哥犬吻蝠而非普通吸血蝠。
6. 实验5包含两个相关文件：experiment5data.csv包含关于吸血蝠Y型迷宫实验的数据，受试吸血蝠有三种选择方案：其一为选择气味侧或无气味侧，对应试验类型为scent；其二为选择播放蝙蝠鸣叫声的一侧或无播放的一侧，对应试验类型为sound；其三为选择同时带有气味与鸣叫声的一侧或均无的一侧，对应试验类型为both。每只受试蝙蝠均需完成全部三种试验类型，因此其参与试验的顺序也被记录。研究人员测量了受试蝙蝠在各臂停留的时长（duration）以及作出选择所需的时间（latency），该数据也可用于计算移动速度，即受试蝙蝠进入选择臂的快慢。ymaze_results.csv为基于实验5数据开展置换检验的结果输出文件。
本研究的相关工作已通过马里兰大学机构动物护理与使用委员会（协议编号R-10-63）、史密森尼热带研究所动物护理与使用委员会（编号#2015-0915-2018-A9、#2019-0501-2022）、巴拿马环境部（编号#SE/A-76-16、#SE/A-36-19）以及哥伦布与富兰克林县都市公园的访问许可审批。