Building a conservation evidence base for the Critically Endangered Blue-crowned Laughingthrush Pterorhinus courtoisi

Understanding why species have restricted ranges is important to developing a conservation evidence base. The Blue-crowned Laughingthrush Pterorhinus courtoisi (BCLT) is a Critically Endangered passerine with an officially recorded population of c.320 birds found in northeast Jiangxi Province, China. It occurs in colonial, cooperative breeding groups close to villages and agricultural habitat, making it difficult to understand why it is not more widespread across other human-modified landscapes in China, and scant evidence exists explaining why the species is restricted to a small area or why the population remains small.

I combined, contrasted and analysed existing data, including two independently collected monitoring datasets, and citizen science data. Results showed the BCLT population may be 42.8% higher and occupy more breeding sites than previously believed. A species distribution model (SDM) showed BCLT occurrence is linked to annual temperature range, precipitation during the breeding season and land cover. I also conducted systematic interviews across the BCLT range at 39 villages. Results revealed two villages where BCLT may breed (but not officially recorded), the first evidence of recent trapping of the BCLT in southeast China, and regions which should be prioritised for conservation efforts. I also found an association between BCLT breeding sites and local landscape changes.

Previous research into BCLT habitat associations have compared relatively few breeding sites and not accounted for the wider mixed agricultural mosaic. I quantified habitat preferences with resource selection functions across 39 villages, comparing the home range and nest site scales. Nesting sites were significantly more likely to contain large broadleaved trees or stands of bamboo, fir or mixed forest, and be on flatter slopes. This supports past work linking BCLT to mature broadleaved trees, but also nuances past assumptions about BCLT nesting habitat.

To understand influences on range at the landscape scale, I compared seven SDM model algorithms and two pseudo-absence methods, using a land cover map with eight types, and k-fold cross validation. I showed that BCLT occurrence is linked to precipitation during the breeding season and in landscapes with higher tea plantation. This may be related to the presence of mature trees and non-crop vegetation in tea plantations across the study region, with implications for landscape policy.