When I started my doctorate working with John Stireman of Wright State University, he and I already had discussions about the chapters in my dissertation. We agreed that there should be a broad biogeographic or ecological component, a narrower phylogenetic component, and a taxonomic revision. I am happy to say that the taxonomic portion passed review in 2020 and is currently in final edits. The ecological portion was published last March in Insect Conservation and Diversity. I long ago promised that I would blog all my publications, but I've obviously dropped the ball on that here. So this is the story behind the paper.
In 2012, I started my doctoral program at Wright State in Dayton, Ohio. At the time, Stireman had been running malaise traps out in the woods adjacent to campus and in several other locations near the University. These flight intercept traps catch all sorts of things, but his main interest was to catch tachinid flies, of which he has a sizable personal collection. One of these locations was from Huffman Metropark; most of these data were published by his former student Diego Inclan. Stireman also had a large tachinid pan-trapping data-set from his doctoral research in Arizona. And, to top it all off, he had access to specimens from a large trapping project at Yanayacu Biological Station in Ecuador. These three regions, Ohio, Arizona, and Ecuador, are spread across 40 degrees of latitude from the temperate zone to the high tropics, which allowed a rough comparison of diversity between tropical and temperate tachinid communities.
The reason this was such an interesting idea, this latitudinal comparison, is that there is this historical idea that parasitoids are less species rich in tropical zones than in temperate regions, that they have a sort of "inverse diversity gradient", inverse to the normal strong gradient of lower to higher diversity as one moves towards the tropics. Dan Janzen wrote a article about this in the mid 1980s, where he used species diversity of parasitoid wasps across North American latitudes to suggest that these wasps have a mid latitudinal peak in species diversity, at around 30-25 degrees North Latitude, and that after that the diversity declines. And furthermore he, and other authors, suggested that tachinid flies would follow this pattern similarily or even more so.
Now, this particular hypothesis was a sort of creative impulse to write this chapter of my dissertation. But at the time of publication, and even ten years ago at the time I started writing this research, the "inverse latitudinal gradient" for tachinid flies was already a rejected hypothesis. The copious amount of new species coming forth from Janzen's own rearing project in Costa Rica made that abundantly clear. So, it was not my purpose to falsify that hypothesis with this paper, but it was an interesting place to start.
Over time I added several other datasets for comparison. Sonja Scheffer and Mathew Lewis at the University of Maryland allowed us to add their dataset for comparison. Karen Petersen, a former master's student under Stireman, had surveyed in another area at Yanayacu, so I added that small dataset. And Marc Pollet had a good sized survey of tachinid flies from Podocarpus-El Condor Reserve in Ecuador that he collected while trapping for Dolichopodidae. (Marc, Stireman, and Diego all joined me as authors on the final publication.) But by far the largest later addition was all the Tachinidae collected during the Zurqui All Diptera Biotic Inventory (ZADBI) project in Costa Rica, for which Stireman and I identified a large portion of the tachinid specimens. With the addition of that Costa Rica dot I also filled a latitudinal gap, so in the paper we had a general latitudinal spread of 40--30--10--0 degrees. The result was that, every time we thought we had the paper set for publication, it grew another dataset, so what should have taken a few years instead took most of a decade.
(As an aside, I'd like to reflect briefly on the ZADBI project. Since I wasn't a forefront author for those papers, I don't feel like it's right to do a full post about it. But I do want to comment on one point, that over half the tachinid species collected there were in the tribe Blondeliini, which is my taxonomic research focus. Blondeliine diversity is absolutely insane in the Neotropics, far beyond any other group, and the ZADBI project really bore that out.)
The unsurprising result was that we saw a general increase of tachinid species richness from the temperate to the tropical zone. But as I said this had already been established by taxonomists for decades. What is far more interesting to me, is the rarity of most species in all datasets. If you look at the accumulation curves in Figure 1, you can see that none, not ONE of those datasets reaches an asymptote. Even years into collecting, species new to the longer surveys were being recorded. And this is across the board, both large and small datasets, both datasets in the temperate region and in the tropics. In all cases, 55-68% of all species were collected only once (singlets or singletons) or twice (doublets or doubletons). This to me is absolutely astounding, that even in Ohio, more than half of all tachinid species can be counted as rare. And in Costa Rica, at Zurqui, more than two thirds of all species, 68 percent (!) were collected only once or twice. What this says to me, is that any accurate measure of tachind community richness requires a long period of collecting, perhaps decades of surveying. For tachinid flies, Preston's Veil seems wide.
Overall, I'd say I'm pleased with the paper. There's still that pesky gap at 20 degrees (Mexico) and it wasn't possible to make any strong conclusions due to the low number of datasets. There are more tachind survey datasets out there, but they're survey's of caterpillars that sometimes yielded parasitoids. Since caterpillar rearing is generally not a random selection process, it's not comparable with trap surveys. So, we could say that the metrics strongly suggest higher species richness in the tropics, but there is nothing in the paper that counts as a definitive test. My question is: If someone were to repeat the same survey at Zurqui, or Yanayacu, how much would those two datasets overlap? 90% species the same? 70%? We know nothing about the turnover of richness in these communities, how many species are transients or inhabitants, or even what "rare" means for Tachinidae. Does it mean "low number of individuals"? Does it mean "widely dispersed in the landscape", and thus rarely encountered? We know that these flies often hilltop, they gather at specific points in the landscape for mating, and we know they have the flight strength to travel long distances. After all, "tachinid" comes from the Greek word "táchos", meaning "speed". Maybe how we see landscapes, with our narrow surveys, means little to a speedy fly.