Abstract:
The biological diversity of Samoa’s freshwater macroinvertebrates remains poorly understood. To date, limited research has been conducted on the diversity of damselflies and macrocrustacea, while much of Samoa’s remaining freshwater macroinvertebrate diversity is not characterised or catalogued. My study used DNA barcoding to identify commonly occurring invertebrates in the waterways of Upolu Island, Samoa, and tree building to characterise their phylogenetic relationships. By developing an understanding of the diversity of freshwater macroinvertebrates present in Upolu, and their distribution in relation to anthropogenic impacts, an important aim of this study was to assess if macroinvertebrate indices of stream health could be applied to monitor the ecological health of waterways in Samoa. Nine streams were selected for analysis, with stream sampling locations determined in partnership with the Ministry of Natural Resources and Environment (MNRE), Samoa, to match sample collection with sites where chemical sampling has previously been conducted. Three streams were sampled at three reaches (upper, mid and near coastal reaches) to collect samples across a gradient from most pristine (upstream) to most impacted (coastal) perceived anthropogenic intensity gradient. Five were sampled at one site each to collect additional data from more pristine sites, one stream was also sampled at two locations. The identity of taxa was determined morphologically, using a microscope where necessary, and also determined based on analyses of mitochondrial cytochrome c oxidase subunit (COI)genes, amplified from our invertebrate samples. Where polymerase chain reaction (PCR) amplification of COI genes failed in the first instance, the protocol was repeated, but at lower annealing temperatures to maximise the likelihood of amplification. PCR products were purified and sent to the Massey Genome Service for Sanger (ABI3730) DNA sequencing, where DNA was sequenced in both forward and reverse directions. Once the sequences were retrieved and quality checked, the DNA fragments were searched against the NCBI BLAST database for closest matches. The 100 best BLASTn pairwise matches were realigned with the invertebrate query sequence using Clustalx2.1 and weighted and unweighted neighbor joining trees were constructed to examine the relationships among the sequences in each of these datasets. A weighted Neighbor Joining tree was also constructed to visualize the genetic diversity among the determined macro-invertebrate sequences. The abundance of each taxon and the overall richness of taxa at each site could then be reported. Overall, 51 freshwater macroinvertebrates were morphologically identified. 27 taxa were given species level assignments, 9 taxa were identified to family and genera level using DNA methods. Because there was limited study of intraspecific variation and limited data available for interspecific comparison, these assignments are hypothetical taxonomic assignment that require further testing. Our study identified 29 taxa that had not previously been recorded in Samoa using DNA methods. This included taxa with very high COI sequence similarity to species reported elsewhere as well as taxa with low levels of COI similarity to existing GenBank sequences. The majority of flies found in our study have not been previously reported in the freshwaters of Samoa. This includes dragonfly, mayfly, Chironomids, vinegar fly, microcaddisfly and truefly. Newly reported insects including a “water strider beetle” (Gerrispingreensis), "elmid beetle" (Staphylinidae sp.) and an organism that we described as an"unknown bug" with COI sequence similarity to Microvelia americana. Shrimps not previously recorded in Samoa included Caridina jeani, C. nilotica, C. rubella, and a Stenoithoidae species. Worm species newly recorded for Samoa included types of “paddleworms” and segmented worms. We inferred if taxa were pollutant tolerant or sensitive by applying pollution sensitivity scores to each taxon using New Zealand’s macroinvertebrate community index (MCI) scores. There was no strong indication that pollutant sensitive taxa were more abundant, or that pollutant tolerant taxa declined in abundance in streams visibly less impacted by human activity. However, upstream sites located in populated areas that were visually polluted had lower counts of pollutant sensitive taxa. We found sites with high MCI scores, and low QMCI (quantitative macroinvertebrate community index) values have a high abundance of lowscoring taxa (pollutant tolerant are dominant), while sites with high QMCI values and lower MCI scores have a greater abundance of high scoring taxa (pollutant sensitive taxa). Remarkably, my study is among the very first to begin to catalogue the genetic diversity of freshwater macroinvertebrate communities in Samoa. Our study recorded the first communities of mayfly, dragonfly, Chironomids, vinegar fly, microcaddisfly and worms for Upolu. Some of our species are likely found in other Pacific countries such as some Chironomids, where BLASTn matches to French Polynesian species were found to be similar (91.8%). Although we were unable to identify clear gradients in the composition or richness of macroinvertebrate communities across sampling gradients that we perceived to reflect human impact, our study provides a first step in assessing the validity of a macroinvertebrate community index to monitor the health of Samoan freshwater. My study provides a stepping stone for future work into identifying impacted and clean streams and assigning tolerance scores to Samoan macroinvertebrate taxa and communities, aiding the development of an MCI for monitoring water quality.