Faecal samples analysed per bat species for each location, season and habitat type in northern Madagascar

This table is part of the PhD thesis of Carme Tuneu-Corral, entitled 'Bats and rice: promoting Integrated Pest Management to enhance biodiversity conservation'. It is the Table A4.1 of the supplementary material of the Chapter 5 'Beyond borders: evaluating the role of protected areas in promoting bat-mediated pest suppression in rural areas of northern Madagascar', and compiles the information on the number of samples analysed per bat species for each location, season and habitat type. Buildings were all located in rural villages outside protected areas, and caves and forest were all located inside protected areas.

Methodology: 

Due to the subtropical climate of this country, we sampled bats in two periods representing the two main seasons for both Amber Mountain and Ankarana. Dry and wet season vary slightly in timing and extent, and we adjusted the field periods to cover the end of the dry season (October – November 2022), and the end of the wet season (May 2023, as in April one of the protected areas was inaccessible due to the poor state of the road caused by heavy rains). French Mountain was visited only in the dry season (October – November  2022) due to logistic limitations. We used mist-nets and harp traps to sample insectivorous bats in protected areas and rural villages. Bat captures were conducted for 7-10 consecutive nights in each protected area, changing sampling points daily to cover a wider area. We sampled bats in a total of six caves and four forest points in Ankarana; six caves and two forest points in French Mountain; and 10 forest points in Amber Mountain (there are no known caves in this protected area). Three school buildings were visited at 16 km from Ankarana borders, two public buildings at 3 km from Amber Mountain borders, and two school buildings at 1 km from French Mountain borders. When sampling forest habitats, mist-nets and harp traps were installed 30 minutes before sunset and remained open for at least four hours. When sampling bats at roost entrances (caves inside protected areas and buildings in rural villages), bats were captured before sunrise when returning to the colony after feeding. In these cases, we used mist-nets and harp traps set up four hours before dawn, which remained in place until 30 minutes after sunrise.

All insectivorous bats captured were measured, weighed and identified using different bibliographic references. Bat captures and handling were conducted following guidelines approved by the American Society of Mammalogists (Sikes and Gannon, 2011). Bats were kept in clean cloth bags until they defecated, generally for a maximum of one hour. In case the bat individuals had not defecated within that time interval, they were released after being measured, weighted and identified. Faecal pellets of each individual were immediately collected and stored in 2 ml tubes with 95% ethanol and labelled accordingly. All bats were released at the same site where they were captured. We gathered a total of 500 samples to assess bat diet: 250 samples collected in the dry season and 250 samples in the wet season. The number of samples collected per species varied according to their rate of capture at the different sampling points.

We were able to trap 15 different insectivorous bat species, 11 Madagascar endemics, and four regional endemics also occurring on nearby islands. Twelve bat species were captured exclusively inside the protected areas and those considered largely forest dependent are highlighted in bold (Chaerephon jobimena, Laephotis matroka, Macronycteris commersoni, Miniopterus aelleni, M. ambohitrensis, M. gleni, M. griveaudi, Myotis goudoti, Otomops madagascariensis, Paratriaenops auritus, Paremballonura tiavato, Triaenops menamena), and two were captured exclusively outside the protected areas (Chaerephon leucogaster, Mops leucostigma). Mormopterus jugularis was the only species found roosting both in caves and buildings, although it was primarily captured in natural roosts in this study. We collected faecal samples for all these species, obtaining operational genetic material (i.e. OTUs representing more than 1% of the total dietary reads of each sample) from 454 of the 500 faecal samples analysed (Table A4.1).