The ‘River Lime’ Effect: detecting the signature of recreational use on a tropical river ecosystem
Our latest paper, published this week in the journal Ecology and Evolution, describes how we detected the ‘signature’ of disturbance on tropical freshwater fish communities, using the streams of Trinidad’s Northern Range as a model system.
“Liming” is a popular pastime on the Caribbean islands of Trinidad & Tobago. It describes the act of relaxing with friends and family – usually accompanied by food, drink and loud music. The island’s many picturesque streams are favourite locations for liming, which means that certain stretches are exposed to very high levels of recreational disturbance. In this study we sampled 8 pairs of sites in the Northern Range of Trinidad, multiple times over a 2 year period; each matched pair of sites consisted of one well-known liming spot and a nearby, less-disturbed, site.
Disturbance can impact natural communities in multiple ways, so we took a multi-level approach in our analysis. For each site, we looked at the total fish biomass, fish species richness and three different diversity indices that take ‘evenness’ into account (Shannon, Simpson and Berger-Parker).
We also decided to put the spotlight on one species within the community – the Trinidadian guppy (Poecilia reticulata) – to see if this well-studied, fast-breeding, widespread species would reveal any patterns at the population level. We looked at the proportion of guppy biomass (in relation to total biomass) and the sex ratio of guppies at our sites.
Overall we found that the more disturbed sites were associated with significantly greater biomass and species richness than the less disturbed sites. This could be due to greater nutrient input at the disturbed sites, which may allow it to support more individuals and consequently more species (known as the “more individuals hypothesis”).
We also found that higher levels of disturbance tended to correspond with more female-biased guppy populations. This may be due to a greater sensitivity to physiological stress in male guppies, perhaps as a result of males increased investment in bright colouration.
However, none of the indices that take into account the proportional abundance of species (evenness) showed any relationship with disturbance. The proportion of biomass attributed to guppies also showed no change in response to disturbance. This suggests a certain level of resilience to disturbance in the structure of these communities.
As human populations increase, recreational use of natural habitats is set to do the same – not least for tropical streams – making it increasingly urgent that we understand the effects we have on such ecosystems.
We hope that these results from Trinidad’s streams will add to our understanding of other tropical ecosystems, and help us learn more about how to conserve and manage those that are most vulnerable. Importantly, the fact that some of the most commonly employed indices of diversity did not detect any effect of disturbance in our system, emphasises that we should not rely on these ‘classical’ measures alone, and that further research into the functional implications of changes at all levels of community properties is urgently needed.
This work was carried out as part of the 5 year ERC-funded BioTIME project. The full paper can be accessed here: Deacon, A.E., Shimadzu, H., Dornelas, M., Ramnarine, I.W. & Magurran, A.E. (2015) From species to communities: the signature of recreational use on a tropical river ecosystem. Ecology and Evolution