Biotropica 46(5) Editor’s Choice: Genetic Diversity & Phylogeography of trees in Tanzania’s Eastern Arc Mountains
The September 2014 Editor’s Choice article is: Alistair S. Jump, Martin Carr, Antje Ahrends and Rob Marchant. 2014 Genetic Divergence During Long-term Isolation in Highly Diverse Populations of Tropical Trees Across the Eastern Arc Mountains of Tanzania. Biotropica 46(5): 565-574.
Biotropica is well known for it’s contributions to tropical ecology, with an early emphasis on the Neotropics. However, it has always been a biological journal – Pantropical in scope and advancing our understanding of the ecology and evolution of tropical communities. This month’s selection is an excellent example of that – a contemporary take on a long-standing evolutionary question, conducted in an understudied system, and with important implications for conservation. Below the authors relate the motivation behind their work and summarize its importance (hence the British rather than American spellings!). They also provide some spectacular photos of their field sites. EB
The Eastern Arc Mountains (EAM) are a chain of isolated mountain blocks rising suddenly from the savannahs of eastern Tanzania and southern Kenya. Cooler temperatures and consistent orographic rainfall and mist have given the EAM potential to support large areas of tropical moist forest. However, fertile soils and often abundant water has resulted in these mountains being highly attractive for human settlement and agriculture, while many of the forest trees and other species have high commercial value. Consequently, only fragments of the once extensive forests remain.
The EAM forests are strikingly diverse and some authors have proposed that they are amongst the oldest on Earth, having been maintained through successive periods of climate change, buffered from regional and global climate changes by moisture laden air from the Indian Ocean. Biological uniqueness, antiquity and threat, combined with their importance for ecosystem services for surrounding towns and villages has led to the region becoming the focus of significant research and conservation effort over recent decades. Yet our biological knowledge and thus conservation priorities across the area remain spatially biased towards sites of early exploration. Furthermore, one of the central tenets underpinning understanding of the EAM region, the great antiquity of the forests themselves, is based on very little direct evidence.
As part of a wider research programme into the structure, function and history of the EAM forests we set out to test hypotheses concerning the distribution and persistence of diversity across the EAM. Contemporary patterns of diversity can be observed from modern survey work, while historical diversity and changes over time can be glimpsed through palaeoecological techniques. However, palaeoecology can only take us back in time so far, and so we turned to phylogeography – the exploration of spatial patterns of genetic diversity and population divergence within a species, to look for persistent signals of colonisation and dispersal of tree species within the EAM. Looking at one species can tell us a lot about that individual species, but comparative work across a number of species can give us information about the forest ecosystem as a whole, and so we sought to compare phylogeographic information across a range of common tree species of the EAM forests. Following a stay with botanist Iddi Rajabu at the Amani Nature Reserve to hone ID skills, and then armed with an experienced tropical botanist and Swahili speaker (Antje Ahrends), palaeoecologist (Jemma Finch) and environmental modeller (Philip Platts) and with the support of local colleagues and villagers, we set off to sample and survey across the EAM forests.
High diversity often results in low density of and consequent difficulty finding individual species – a simple relationship between the number of things you can fit into a finite space, and the number of individuals of each particular type that can be represented. Consequently, our initially larger panel of candidate species was reduced to the five used in the final work. Furthermore, not all species occurred across all mountains – ecological reality tempering the desire for a balanced sample design. Many of the EAM forests are remote and difficult to enter. Logistics thwarted us on a few occasions – a bridge lost in a flood left us with no access to sample in one forest, while Dr Finch’s nerves of steel (and considerable experience of four wheel drive) rescued our vehicle from toppling into a ravine when the track collapsed beneath us on another occasion. Our accommodation varied from a creaking catholic mission (and bat refuge) through basic village huts to a honeymoon suite at a medical research institute/occasional wedding venue – narrowly avoiding having to bunk in a brothel along the way, while in the forest my notoriously leaky tent did not do well in the near constant rain. Yet target trees were located, plant presses and the tubes of silica gel were filled with samples, and some two months later the lab work began in earnest back in the UK.
Many months later, and following the frustrations of Martin Carr and I dealing with analysis of the inevitably gappy dataset, the DNA from the leaves collected in the forests yielded surprising results. Comparing individuals across the sections of chloroplast DNA that we analysed showed that geographical structuring of genetic diversity was weak-to-absent across the region. However, populations on different mountain blocks were also highly divergent and often genetically distinct, with species showing little similarity to one another in the patterns identified.
When looked at as a whole, and in the light of other work, we believe that the data support a model of gradual accumulation of random changes in DNA sequence rather than effective recent exchange of genes between populations – or divergence in isolation rather than ongoing seed transfer between mountains. Additionally, while diversity is high across the region, no single forest stands out as being consistently more genetically diverse than others when all species are combined. This work adds another layer to our understanding of the diversity of EAM forests and lends further support to their hypothesised long-term persistence. Furthermore, our findings suggest that biodiversity research and conservation resources should be more evenly distributed across the region, since we find no evidence that currently prioritised mountains are any more genetically diverse or divergent than their neighbours.
University of Stirling