Quantifying species diversity with a DNA barcoding-based method: Tibetan moth species (Noctuidae) on the Qinghai-Tibetan Plateau

Abstract

With the ongoing loss of biodiversity, there is a great need for fast and effective ways to assess species richness and diversity: DNA barcoding provides a powerful new tool for this. We investigated this approach by focusing on the Tibetan plateau, which is one of the world's top biodiversity hotspots. There have been few studies of its invertebrates, although they constitute the vast majority of the region's diversity. Here we investigated species diversity of the lepidopteran family Noctuidae, across different environmental gradients, using measurements based on traditional morphology as well as on DNA barcoding. The COI barcode showed an average interspecific K2P distance of 9.45±2.08%, which is about four times larger than the mean intraspecific distance (1.85±3.20%). Using six diversity indices, we did not detect any significant differences in estimated species diversity between measurements based on traditional morphology and on DNA barcoding. Furthermore, we found strong positive correlations between them, indicating that barcode-based measures of species diversity can serve as a good surrogate for morphology-based measures in most situations tested. Eastern communities were found to have significantly higher diversity than Western ones. Among 22 environmental factors tested, we found that three (precipitation of driest month, precipitation of driest quarter, and precipitation of coldest quarter) were significantly correlated with species diversity. Our results indicate that these factors could be the key ecological factors influencing the species diversity of the lepidopteran family Noctuidae on the Tibetan plateau.

Figure 1. (a) Phylogenetic tree of Tibetan moth species from the family Noctuidae, inferred using neighbor-joining analysis of 320 COI DNA sequences.

Different colours represent different genera. (b) Plot of genetic distances showing the barcoding gap. Intraspecies in red, interspecies in blue. Fitted normal distribution curves are shown in red.

Figure 2. Species diversities of Tibetan moth species (Noctuidae) for each community, calculated using different diversity indices.

(a) Shannon index; (b) Simpson index; (c) Brillouin index; (d) index; (e) Exponential Shannon index; (f) Transformed Simpson index.

Figure 3. Rank-abundance curves of each community based on DNA barcoding (DB) with different reference database (a–c) or traditional morphology (TM) (d).

(a) DB-based rank-abundance curves for each community based on reference size of 10% (01 ref). (b) DB-based rank-abundance curves for each community based on a reference size of 30% (03 ref). (c) DB-based rank-abundance curves for each community based on a reference size of 50% (05 ref). (d) TM-based rank-abundance curves for each community.

Figure 4. Correlations between species diversity and environmental factors.

(a) Significant correlation between species diversity and three environmental factors, including precipitation of driest month, precipitation of driest quarter, and precipitation of coldest quarter. (b) Distribution of sampling sites/communities along different environmental gradients, including radiation and rainfall. Clustering of seven communities based on PCA of 22 environmental factors. (c) Spatial distribution of species diversities between West and East regions. (d) Rank-abundance curves of West and East communities.

Figure 5. Sampling sites in the Qinghai-Tibetan Plateau in China.

Samples were collected from seven locations from West to East in Tibet, representing different ecological conditions. RKZ - Shigatse, Gyantse - JZ, Lhasa - LS, Shannan - SN, Pagsum Co - BSC, Mainling - ML, and Bome - BM.

Figure 6. Species identity inferred via DNA barcoding for a community with a Bayesian method.

(a) An ecological community with species. (b) Bayesian formula used to infer species membership for unknown samples. (c) DV-Curve approach to construct DV-Matrix for Bayesian inference. is the th species in the community, is an unknown sample with DNA sequenced, is the probability of a sample belonging to . is the conditional probability of the unknown sample belonging to given that its DNA is sequenced. is the probability of having DNA sequence of the unknown sample given that DNA sequences of are known.