Impacts of abiotic and biotic factors on terrestrial leeches in

Haemadipsid leeches are ubiquitous inhabitants of tropical and sub-tropical forests in the Indo-Pacific region. They are increasingly used as indicator taxa for biomonitoring, yet very little is known about their basic ecology. For example, to date no study has assessed the occurrence and distribution of haemadipsid leeches across naturally occurring gradients within intact habitats. We analysed a long-term data set (2012-2020) on the closely related tiger (Haemadipsa picta) and brown (Haemadipsa spp.) leech species to investigate if and how abiotic and biotic factors influence their occurrence across a gradient of forest types at an undisturbed tropical rainforest site in Indonesian Borneo. We compared a series of negative binomial mixed models and found that, of the abiotic factors, soil moisture had the largest positive effect on encounter rates of both leech species. Among biotic factors, forest type had differential effects on counts of the two species: while tiger leech counts were greater in low elevation forest types, brown leech counts were greater in high elevation forest types. Additionally, we found that the presence of one species had a positive effect on the presence of the other species. Finally, our results show that the tiger leech has a narrower distribution, being restricted to lower elevation forest types with higher water retention, suggesting that the tiger leech could be more sensitive to lower soil moisture levels.Microsoft Excel & Program RFunding provided by: University of MichiganCrossref Funder Registry ID: http://dx.doi.org/10.13039/100007270Award Number: Funding provided by: Victoria University of WellingtonCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001538Award Number: Funding provided by: Leakey FoundationCrossref Funder Registry ID: http://dx.doi.org/10.13039/100005966Award Number: Funding provided by: Idea WildCrossref Funder Registry ID: http://dx.doi.org/10.13039/100007142Award Number: Funding provided by: Mohamed bin Zayed Species Conservation FundCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100011672Award Number: Funding provided by: Disney Conservation FundCrossref Funder Registry ID: http://dx.doi.org/10.13039/100012168Award Number: Funding provided by: AZA Ape TAG Initiative*Crossref Funder Registry ID: Award Number: Funding provided by: Anne Nacey Maggioncalda and Jeffery Maggioncalda*Crossref Funder Registry ID: Award Number:In 2000, AJM established a series of 14 vertebrate survey transects across the seven primary forest types at the Cabang Panti Research Site in West Kalimantan, Borneo. The mean length of survey transects was 3.4 ± SD 0.24 km. Some transects were contained entirely in a single forest type, and others spanned multiple forest types. For analysis, transects were divided into segments that were restricted to a single forest type and varied in length from 50 to 550 m. We measured the altitude of each forest type at its approximate midpoint using a Suunto Vector Altimeter. From November 2012 to September 2020, we systematically counted leeches along these vertebrate transects. We walked each transect at a constant, slow speed (~1 km per hour) twice per month (starting at opposite ends), beginning at 0530 h. Surveys were normally carried out by a single observer, although occasionally two observers would walk the same transect (e.g., when new staff were trained). Every 500 m along transects or at the beginning of each segment, observers paused to examine themselves thoroughly for a period of two minutes. Any leeches attached to their bodies were counted, identified, and removed. When two surveyors were present, both individuals searched for and counted leeches, and the recorded number of leeches was combined for the two observers. In addition to leech species and counts, within each segment, observers made a note of soil moisture and rainfall history. For soil moisture, observers examined the dampness of soil and categorized it as "wet", "moist" or "dry". They categorized rainfall history for the whole survey day as "Raining now", "Rained last night", "Rained yesterday", "Rained day before yesterday" or "Rained over 48 hours ago" based on conditions at the research station. Rainfall can sometimes be highly localized at our site, so it is possible that occasionally a segment's rainfall history assessment based on conditions at the research station may have been inaccurate. We have no reason to believe such rare mischaracterizations were biased in any particular direction, however. Data tidying and analysis were carried out using program R (R Core Team, 2020; RStudio Team, 2021)