Elevational Diversity Profiles of Aves and Lepidoptera (Sphingidae) – A Comparative Analysis in the Eastern Himalayas

Abstract

In this dissertation we present a comparative study of the diversity patterns of two disparate organismal groups – Lepidoptera: Sphingidae (hawkmoths) and Aves (birds) – along an elevational gradient spanning 200-2800 m in the eastern Himalayas of Arunachal Pradesh, India. The concurrent sampling of the two taxa along the same elevational transect should reduce the number of confounding factors due to the identical parameters associated with climate, vegetation and biogeography. We have investigated two important facets of diversity, viz. species and (functional) traits using multiple measures available for each (e.g. alpha diversity, beta diversity, evenness, etc.). It is hoped that this study would contribute to the growing field of identifying the taxon-specific and taxon-independent factors impacting diversity. To the best of our knowledge this is the first such simultaneous elevational diversity pattern study of two animal taxa in the eastern Himalayas of India. We recorded 4731 hawkmoth individuals from 13 elevations between 200-2800 m, spanning 80 morpho-species and 30 genera. We reliably measured body and wing sizes for 3297 individuals using field images, after correcting for image distortions, one of the largest and systematic compilation of insect trait data from a single locality, and achieved without having to collect specimens (and the consequent allocation of large amount of resources for their preparation and maintenance). For birds, line transects at 48 elevations yielded 15,867 individual records spanning 235 species, 150 genera and 48 families. We obtained species mean body mass and wing span data from literature. We also carried out a similar sampling of birds during winter but is not presented in this thesis. 1. Species abundance distributions (SAD): We have shown that evenness of species abundances within a community decreases with elevation for both hawkmoth and birds. The decline in evenness was consistent across multiple measures including parameters from models (e.g. standard deviation of a Log-normal fit), as well as model-independent metrics such as the width of the octave binned species abundance distributions, slope of rank-abundance curves and Pielou's evenness index. High evenness has been previously observed in stable, more productive ecosystems with high species richness and more resource/niche partitioning (like at low elevations and latitudes), whereas low evenness is linked to unstable, variable and less productive environments where few species dominate (like at high elevations and latitudes). We also found that the Log-series and Neutral models fit the individual elevational communities of hawkmoths the best, while bird communities are more mixed with Log-normal dominating the fits. However, for the regional pool as a whole the Log-normal was by far the best fit for both taxa. 2. Alpha and beta diversity profiles: Birds and hawkmoths, both exhibited a mid- elevation peak in alpha diversity, although, the peak was lower and broader in hawkmoths than in birds. However, interestingly, when each species was represented only by the mean elevation of its distribution, the bird profile turned out to be essentially independent of elevation, while hawkmoths continued to show a mid-elevation peak. Analysis of beta diversity revealed a similar reduction of dissimilarity with elevation for both birds and hawkmoths, with species turn-over dominating nestedness. Despite this similarity, there is some indication from NMDS analyses that elevational communities in moths are clustered into 5 groups while those in birds showed a smoother transition in community composition across the gradient. 3. Bergmann's Rule: Previous investigations, in the hundreds, present a very confusing picture of the validity of Bergmann's rule, raising issues of its nature (pattern or process), of its applicability (ectotherms or endotherms or both), and the scale of investigation (intraspecific, interspecific and assemblage). In this study both hawkmoths and birds showed a strong but contrasting trend at the assemblage level, with the mean body mass increasing for hawkmoths (an ectotherm!) and decreasing for birds (endotherm!). We also found that the same data shows a considerably reduced signal when plotted as species mean values (as has been done by most previous studies) and a very weak signal at the intraspecific level. One way of reconciling the contradiction is to ascribe the effect to species turnover. The converse-Bergmann pattern in birds has previously been attributed to reduced food availability during winters at high elevations. As facultative endotherms, thermal explanations for Bergmann's rule that are normally applied to endothermic vertebrates could potentially operate in hawkmoths, especially as they are most active during the hours of the (cold) night. Our work shows that the result changes with the taxonomic level of investigation. Given the strong contrast in patterns for hawkmoths and birds at the same location, and the fact that the relationship is contrary to the original expectations for ectotherms and endotherms, we support the notion that Bergmann's Rule should be treated as a pattern and the underlying process has to be investigated separately depending on the taxon and the context. 4. Wing allometry and Flight: We found that the allometry of wing area is remarkably similar across more than 4 orders of magnitude of body mass spanning hawkmoths and birds. This suggests that very similar, biophysical, factors are operating across these two taxa in deciding flight dynamics. Notwithstanding which, somewhat curiously, the allometric indices for all genera (but one) of hawkmoths lie below the assemblage average. This suggests that while basic biophysical principles of flight dominates across the range of body mass, the small-scale allometry is decided by factors like inter- and intra-specific competition, local climatic variables (air density, perhaps wind) and perhaps historical contingency. We also found that the assemblage-mean wing loading decreased with elevation for both hawkmoths and birds suggesting better flight capacities at higher elevations. In common with the result from Bergmann's Rule we found that the strength of this correlation with elevation changes with the taxonomic level at which it is investigated: strongest at the assemblage level and weak or undetected at the interspecific and intraspecific levels. 5. Functional trait space in hawkmoths: Given the large data set of (functional) trait values for 3297 individual hawkmoths we investigated several issues related to functional diversity, occupancy in trait space and their implication for community assembly. We observed a decrease in trait overlap across communities with increasing elevational distance using the primary traits of body length, thorax width, wing length and wing breadth, as well as derived traits of body volume, wing area and aspect ratio. We also showed that hawkmoth subfamilies, and even the most abundant genera neatly segregate with almost no overlap in trait space involving just the 4 primary traits listed above. We also showed the presence of strong internal filters at all elevations using T-statistics, with individuals within species significantly more similar than individuals across species in each elevational community. To sum up, using a large set of systematically collected diversity data on hawkmoths and birds along 2600 m of an elevational gradient we have been able to show several patterns with a high degree of statistical significance. We recognise that these are patterns and we need to progress towards processes. Yet, we believe that generating secure patterns is the first step towards understanding the underlying processes. Previous studies have highlighted the large “diversity” of patterns across different taxa and in different locations across the globe; and the quest for identifying the key processes has often floundered on the diversity of contexts of the different studies. We determined the patterns for two very different organismal groups, hawkmoths and birds, at the same place and during the same period. Our study fills an important geographical void in the multi-taxon investigations along elevational gradients. The comparison along multiple facets of diversity – alpha, beta, functional – is expected to complement the available meta-analyses of disparate data sets. We are in the process of adding more taxa and genetic diversity to this mix at our study site. We hope that it will be a small step towards a better understanding of the process by which communities are assembled.