Hi! I'm Felipe Maresca, student from Uy studying at @UdelaR and this is my #BESaquaticECRTweets @BES_AquaEco presentation. I’m gonna tell you about my research on discontinuities in body size distributions (1/18)

Body size is a key trait of organism ecology. Body size distributions allow us to infer ecological/evolutionary mechanisms that favor or limit organism viability. Several studies have shown the existence of discontinuities in body size distribution along scales and taxa. (2/18)

This pattern of gaps and clumps in body size distribution (BSD from now on) has been reported in scales ranging from continental scales to local communities, showing sizes with better or worse relative fitness. In order to explain this pattern, Holling came up with the TDH.(3/18)

Texture Discontinuity Hypothesis (TDH) proposes that discontinuities in BSD are a reflection of discontinuities in resources and opportunities at different times and scales. This is one of the several, non exclusive, proposed hypotheses or mechanisms. (4/18)

Community ecology suffers from a deficit between theoretical predictions and empirical backup, and the BSD theoretical frame is not the exception. Few studies have explored the identification of determinants in the distribution of gaps and clumps in BSD. (5/18)

#Metacommunity theory has proven central to understanding the role of regional landscape structure as a determinant of local community structure. Landscape structure and size dependent dispersal has been identified as a fundamental determinant of biodiversity patterns. (6/18)

Using null models we assessed the existence of discontinuities in the body size distribution in 48 local assembles of carnivorous fishes in the Rio Negro basin, Uruguay, and looked into the relation between discontinuities, and landscape structure + local conditions (7/18)

I worked with a database that contained information on 3761 carnivores, as well as environmental data from each of the 48 sites. Information about sampling methods and the construction of the graph representing the metacommunity can be seen in Borthagaray et al 2020 (8/18)

A PCA was done with some of the variables that were highly correlated, the first axis of the PCA captures the Headwater-Outlet gradient, which was used to represent landscape structure. Out-closeness was taken from the graph shown above. (9/18)

If isolation and local conditions didn’t affect BSD, then it would be a random sample of the whole metacommunity. Using this logic a null model was generated which estimated BSD doing 2000 random samples for each local community. Results were compared to the observed BSD (10/18)

Values of standardized Z distribution, which represents the magnitude of the difference between observed BSD and the BSD expected from the null model, were obtained from the comparison at each local distribution. (11/18)

Max and Min values of Z indicate clumps and gaps, and the mass value at which they occurred were registered. Values of Z and mass were then used to perform GLMs. Results show that this values are associated with the headwater-outlet gradient and local nutrient load (12/18)

Larger fish, having more dispersal capacity, have an advantage in intermediate parts of the headwater-outlet gradient because they are able to forage in other communities, be subsidized populationally and recolonize in the event of local extinction. 13/18

As we move to more central, outlet communities this advantage that larger fish had is now available to smaller fish because the communities are better connected. Here, smaller carnivores are favoured because they have lower energetic demands. (14/18)

In peripheric communities smaller fish would suffer the worst part of perceiving the landscape as fragmented. Low population subsidies and low chance of recolonization in case of local extinction determine the presence of a gap at smaller sizes in these communities. (15/18)

The relation between local nutrient load and BSD could be caused by energetic constraints: More nutrient load is related to productive systems. However, the cause could have to do with the different geological substrates of the basin. (16/18)

In conclusion, this study shows that patterns in discontinuous BSD are determined by processes happening at two distinctive scales, local and metacommunitarian, and that this fenomena should be analyzed looking at both dimensions (17/18)

I want to thank my advisors @matiasarim and Ana Borthagaray, and my cool collegues at @Cureoficial. I would also like to thank my friend Vec for helping me with some of the images. You will soon be able to check out info regarding sampling and graph construction here:(18/18)
Bye!

PS: I should also make clear that the first image i used is taken from Allen et al 2006. doi: 10.1111/j.1461-0248.2006.00902.x