Arthropods as indicators for forest monitoring

Abstract

Arthropods comprise the largest proportion of terrestrial biodiversity and are in alarming decline globally caused by anthropogenic activities. Although arthropods are abundant, sensitive to environmental changes, and play various irreplaceable roles in ecosystem functions, their use in biodiversity assessment and conversation management are relatively uncommon due to low taxonomic intractability and the intensive demand of labour inputs. This thesis aims to evaluate the use of arthropods as indicators to assess forest ecosystem health by implementing standardized protocols in South China. First (Chapter 1), I collected arthropods using five standardized methodologies in two South China forest plots under the Forest Global Earth Observatory (ForestGEO) network with distinctive land-use histories: Tai Po Kau; (TPK), a 70 year-old secondary forest in Hong Kong, and Dinghushan (DHS), an old-growth forest in Guangdong province. I showed that arthropods at adequate taxonomic resolutions could be surrogate site indicators of TPK and DHS. Specifically, moth families/subfamilies of Geometridae and Arctiinae are the most useful and convenient indicator groups for DHS and TPK, respectively. I also identified three butterfly and six ant species as site indicators which could be potentially used for long-term monitoring of forest diversity changes. Second (Chapter 2), I further analysed the data of ants, butterflies, and moths to contrast taxonomic and functional structures of TPK and DHS. I found that full taxonomic and functional diversity of TPK had not entirely recovered to levels of an undisturbed state after 70 years of forest succession. Third (Chapter 3), I used five butterfly datasets from the ForestGEO network (including TPK & DHS) across three continents covering up to nine years of monitoring to determine how temporal butterfly species richness was structured at the levels of days, months, and years. I found that yearly sampling was more effective in capturing total species pools while lower-temporal level sampling (e.g. daily or weekly) might be more useful in examining common species demographic patterns. Fourth (Chapter 4), I used a dataset of a comparable butterfly monitoring programme based on a 10-year, monthly sampling in subtropical Hong Kong. I found that warmer temperatures were associated with advances in the spring appearance of butterflies, and temperature-sensitive species may provide useful predictions on how global warming may impact butterfly phenology. Overall, this thesis provides baseline diversity patterns and demonstrates that arthropods are robust indicators for biodiversity monitoring in South China. Long-term monitoring is particularly important for the inventory of species pools and examining temporal diversity patterns along forest succession and climate change. I encourage further research integrating the knowledge of species taxonomy and life history for surrogate indicators. Future monitoring of taxonomic and functional diversity should be implemented for the purpose of preserving species diversity and ecosystem functions.