A modified EDGE framework for the prioritization of arthropod conservation

[insectilluminatigetshrekt]

There is this framework called the EDGE of existence, which stands for Evolutionarily Distinct Globally Endangered. This has been applied extensively to vertebrates (Gumbs et al., 2024). Essentially its a score assigned to a species for conservation priority based on two factors, how evolutionarily distinct it is (measured in millions of years since its last common ancestor with its closest extant relatives on a phylogenetic tree) and how globally endangered it is (its red list status). For example, the tuatara or coelacanth would have very high EDGE scores while the brown rat or common pigeon would very low scores.

Part of the problem of applying this to arthropods is that the threat status for most arthropods is not assessed, and many of the ones that are are only done so under national/regional red lists which might not always use the same criteria or threat classification as the global red list (such as NatureServe). Another problem is that most arthropod species have not been included in comprehensive species level phylogenetic trees. But I think a set of modified EDGE criteria can be a good way to analyze them.

Instead of assigning scores to individual species, we could perhaps assign scores to families. I think this is justified because arthropods are often only identified to the family level during sampling studies, and family richness is usually considered a good proxy for species richness (Báldi, 2003; Hirst, 2008; Versluijs et al., 2023; Zou et al., 2020). Additionally, using families can be robust to species level taxonomic instability (lumping/splitting) even for better sampled groups like vertebrates (Gumbs et al., 2024), let alone arthropods.

My proposal for modified EDGE score for arthropods will be based on three factors. The first one, how evolutionarily distinct the family is on the phylogenetic tree, will be easy since I have already made a family level phylogenetic tree for Panarthropoda, which I have used in a previous analysis of inaturalist data. I am currently working on adding branch lengths and calibrations to the tree, which make ED quantifiable for each family.

Current tree with over 5,000 taxa. This includes fossil taxa, which will of course have to be pruned off the tree for it to be used in such a project.

Next is how many extant species each family, values I have also previously obtained for each family. These numbers can be obtained from group specific taxonomic databases or monographs/revisions. I will keep this to only known species since there is too much uncertainty with extrapolated hypothetical species counts. Number of species can indicate both distinctness (if a family has fewer species, each species within that family is necessarily more unique) and possibly endangeredness (a family with only 4 species is more likely to go totally extinct than a family with 4,000 species).

The third category is range size. Species level distributional data is extremely sparse for most arthropods, but aggregating to the family level may get a better picture. I would primarily base it off GBIF/OBIS occurrences. I would map the range of a family as all ecoregions with occurences of said family. I would use the World Wildlife Fund’s terrestrial ecoregions for mapping terrestrial families (Dinerstein et al., 2017), and their freshwater ecoregions for mapping freshwater families (Abell et al., 2008). For marine families, I would use a composite map with both the MEOW marine ecoregions (Spalding et al., 2007) and Longhurst’s biogeochemical provinces (Oostende et al., 2023), to include both continental and oceanic systems. MEOW ecoregions take priority over Longhurst’s provinces in places where they overlap. While ecoregions vary in size unlike standardized grid cells, they correspond tightly to gradients in elevation, temperature, and vegetation type, which are all things that heavily shape arthropod distribution. Additionally, ecoregions have been used to map the richness of beetle families in a previous study (Moodley et al., 2022). Families with no GBIF or OBIS occurrences may require their ranges to be determined by primary literature. The range size value would be given in square kilometers.

After each arthropod family is given a modified EDGE score, a global richness map of all arthropods would be made. The map would then be divided into standard sized grid cells (perhaps 1x1 km), and each grid cell would be assigned the mean or median of the EDGE scores of all the families who’s ranges occur inside of it. Places on Earth with particularly high concentrations of high scoring families would then be identified as conservation priorities.

Now, I want to pre-empt some potential questions.

1. How can species numbers per family be a good metric when we don’t know most arthropod species? This is a valid concern to have, but I think it wont be as much of an issue as you think, because unknown species richness seems to be concentrated in already very species rich lineages. Li et al. (2025) provides model predictions for future species richness of arthropod classes and insect orders we can analyze. The subsequent charts will be based on their data.

   
   
   Above, you can see the ratios of species richness of major arthropod classes in terms of known species.
   
   
   And here is total species richness predicted by the model. This seems extremely different, but the prediction for future insect species is quite conservative. If we take the lower bound of total insect species richness from Colwell et al. (2026), it gives us ratios between arthropod classes more similar to that of strictly known species as seen below.
   
   
   As seen below, the relative ratios of species richness across major insect orders is even more consistent than arthropod classes under the model predictions of Li et al. (2025), other than a modest decline in the relative share of lepidoptera.
   
   
   
   Additionally, its worth noting a few other things. Extremely specious families also tend to have cosmopolitan distributions. Its not likely they will find new weevils or braconids in large geographic areas where they previously were not known to exist, so the chances that ranges for highly diverse taxa will be under estimated are low. Additionally, errors in EDGE scores from incomplete species inventories will mostly be affecting extremely large families which would not be high priorities under this framework to begin with. In order words, there is not a huge practical difference here weather Curclionidae or Cecidomyiidae has more species, as both are very low priority families under this framework. On the flip side, the chances that there are thousands or millions of undiscovered Grylloblattids, Anaspidaceans or horseshoe crabs is exceedingly implausible.

   With that being said, it can be hard to tell if an apparently rare family is actually rare or just very under sampled. While its not likely that the relative ratios of family species in arthropods will change dramatically such that there is a million species of horseshoe crabs, it’s quite probably that many families may either have considerably more species and/or a wider geographic range than what is currently known. I think, for the sake of caution, it would be beneficial to treat these like we would genuinely rare/species poor/small ranged families simply because we don’t know that they aren’t until more data is uncovered.

2. What about families that have species across multiple realms? Good question, but the answer is simple. Families with both marine and freshwater species will be mapped based on both marine and freshwater ecoregions. Families with both terrestrial and marine species will be mapped based on terrestrial and marine ecoregions, though I can’t think of any examples of this other than Coenobitidae. Families with both terrestrial and freshwater regions will just be mapped in terms of freshwater ecoregions, since freshwater ecoregions cover terrestrial land area and the terrestrial habitation of said clades is likely to be tied to proximity to freshwater.

3. What about rare or endangered species within otherwise common, specious wide ranging families? This EDGE framework would not capture said species, that is a limitation. All I can say is that such species are beyond the scope of this project and are better served in other frameworks.

4. Wouldn’t invasive species artificially inflate the ranges of families? That can be an issue, but as far as I know most invasive species come from already diverse cosmopolitan families, so there would not be huge areas where they establish where other members of their families are not already living, so it wouldn’t result in a substantial range expansions for the family as a whole. An exception to this is islands. A famous example is that Hawaii has no native ants, but there are introduced ants there now. Perhaps some islands/archipelagos and other isolated systems like Antarctica with well catalogued arthropod fauna can be subjected to a special measure where all records from families known to be non native are filtered out.

5. Isn’t geographic sampling uneven? Yes it is, and this is a big reason why number of occurrences is not a factor to determine the EDGE score. Under sampling of certain countries is partially alleviated by the use of ecoregions which cross national borders. If a family occurs in South Korea, it almost certainly occurs in adjacent similar habitats in North Korea even if we have no direct evidence of it.

6. What’s even the point of this? Well, one, this would be the first major step in the application of the EDGE framework to arthropods. Second, I just don’t find most of the existing studies on arthropod conservation to be all that satisfactory. They either focus narrowly on specific taxa or on specific habitats (often agroecosystems). And even broad scale studies are not really about Arthropoda as a total group but rather just terrestrial arthropods, or sometimes even just whatever flying insects are collected at malaise traps. Marine arthropods, to the extent they are analyzed at all in this respect, are often lumped into broader categories like zooplankton or marine invertebrates. A lot of studies also have a large focus on “ecosystem services” (I dislike this term) like the status of pollinators. This study would de-emphasize a lot of hyper diverse insect taxa and put into the spot light more phylogenetically isolated taxa, parasites, meiofauna, deep sea fauna, soil microfauna, troglo/stygofauna, monotypic families, and rare micro endemics, as well as more recently described families who’s distribution is mostly unknown. Think of this as less a work of “Entomology” and more like “Arthropod Zoology”

What do you think of this? Any questions, concerns, or criticisms I may not have addressed?

References

Abell, R., Thieme, M. L., Revenga, C., Bryer, M., Kottelat, M., Bogutskaya, N., Coad, B., Mandrak, N., Balderas, S. C., Bussing, W., Stiassny, M. L. J., Skelton, P., Allen, G. R., Unmack, P., Naseka, A., Ng, R., Sindorf, N., Robertson, J., Armijo, E., … Petry, P. (2008). Freshwater ecoregions of the world: A new map of biogeographic units for freshwater biodiversity conservation. BioScience, 58(5), 403–414. https://doi.org/10.1641/b580507

Báldi, A. (2003). Using higher taxa as surrogates of species richness: a study based on 3700 Coleoptera, Diptera, and Acari species in Central-Hungarian reserves. Basic and Applied Ecology, 4(6), 589–593. https://doi.org/10.1078/1439-1791-00193

Colwell, R. K., Guzman, L. M., Steinke, D., Chao, A., Janzen, D. H., Hallwachs, W., Baker, A., Fernández-Triana, J. L., Hebert, P. D. N., Joyce, F., Puschendorf, R., Quicke, D. L. J., Rougerie, R., Smith, M. A., Zamora, N., & Sharkey, M. J. (2026). Constructing a lower-bound estimate of the global number of insect species on a hyperdiverse empirical foundation. Proceedings of the National Academy of Sciences, 123(27). https://doi.org/10.1073/pnas.2524283123

Dinerstein, E., Olson, D., Joshi, A., Vynne, C., Burgess, N. D., Wikramanayake, E., Hahn, N., Palminteri, S., Hedao, P., Noss, R., Hansen, M., Locke, H., Ellis, E. C., Jones, B., Barber, C. V., Hayes, R., Kormos, C., Martin, V., Crist, E., … Saleem, M. (2017). An ecoregion-based approach to protecting half the terrestrial realm. BioScience, 67(6), 534–545. https://doi.org/10.1093/biosci/bix014

Gumbs, R., Scott, O., Bates, R., Böhm, M., Forest, F., Gray, C. L., Hoffmann, M., Kane, D., Low, C., Pearse, W. D, Pipins, S., Tapley, B., Turvey, S. T., Jetz, W., Owen, N. R., & Rosindell, J. (2024). Global conservation status of the jawed vertebrate Tree of Life. Nature Communications, 15(1). https://doi.org/10.1038/s41467-024-45119-z

Hirst, A. J. (2008). Surrogate measures for assessing cryptic faunal biodiversity on macroalgal-dominated subtidal reefs. Biological Conservation, 141(1), 211–220. https://doi.org/10.1016/j.biocon.2007.09.025

Li, X., Yang, D., Wang, L., & Wiens, J. J. (2025). The past and future of known biodiversity: Rates, patterns, and projections of new species over time. Science Advances, 11(49). https://doi.org/10.1126/sciadv.adz3071

Moodley, S., Procheş, Ş., Perera, S. J., Lubbe, E., & Leschen, R. A. B. (2022). Analysis of the diversity and distributional patterns of coleopteran families on a global scale. Zootaxa, 5138(5), 575–583. https://doi.org/10.11646/zootaxa.5138.5.5

Oostende, M. van, Hieronymi, M., Krasemann, H., & Baschek, B. (2023). Global ocean colour trends in biogeochemical provinces. Frontiers in Marine Science, 10. https://doi.org/10.3389/fmars.2023.1052166

Spalding, M. D., Fox, H. E., Allen, G. R., Davidson, N., Ferdaña, Z. A., Finlayson, M., Halpern, B. S., Jorge, M. A., Lombana, A., Lourie, S. A., Martin, K. D., McManus, E., Molnar, J., Recchia, C. A., & Robertson, J. (2007). Marine ecoregions of the world: A bioregionalization of coastal and shelf areas. BioScience, 57(7), 573–583.

Versluijs, T. S. L., Zhemchuzhnikov, M. K., Kutcherov, D., Roslin, T., Schmidt, N. M., van Glis, J. A., & Reneerkens, J. (2023). Site-specific length-biomass relationships of arctic arthropod families are critical for accurate ecological inferences. PeerJ, 11, e15943–e15943. https://doi.org/10.7717/peerj.15943

Zou, Y., Werf, W., Liu, Y., & Axmacher, J. C. (2020). Predictability of species diversity by family diversity across global terrestrial animal taxa. Global Ecology and Biogeography, 29(4), 629–644. https://doi.org/10.1111/geb.13043

[insectilluminatigetshrekt]

I want to make an additional post highlighting some real stand outs I think would have very high EDGE scores.

Mormotomyiidae: It’s not an especially ancient or evolutionarily distinct family being part of the superfamily Ephydroidea which underwent its main radiation in the early Cenzooic. But it makes up for this by being a monotypic family with only one species, Mormotomyia hirsuta, and being a microendemic that occurs on only one mountain cave in Kenya.

Craterostigmatidae: The only family in the order Craterostigmomorpha. It has only two species, one in Tasmania and one in New Zealand. It also likely diverged from other centipedes in the Devonian.

Punciidae: A family of Ostracods with only 3 species and which are only known to occur in shallow seas above New Zealand, they also diverged from their closest relatives the Podocopid ostracods since at least the Carboniferous

Proteonematalycidae: A strong candidate for perhaps the highest EDGE score of any arthropod family. It has only one species, Proteonematalycus wagneri, is only known to occur on the southern shores of Lake Michigan, and is a very early diverging branch of Acariformes, with its ancestors pre dating the split between Trombidiiformes and Sarcoptiformes all the back to at least the Devonian and perhaps even further back.

[jhbratton]

The comment is often made in the forum that the concept of species is much more artificial than many people realise. Isn’t that artificiality much greater for families? The EDGE score could vary immensely depending on which taxonomy you used and would be susceptible to sudden change when a taxonomic revision happens.

[insectilluminatigetshrekt]

Well, there is always a compromise one has to make between precision and coverage. Its true that families are not immune to taxonomic issues of lumping vs splitting.

But species level data is is far to sparse for most arthropods. Family level data allows the aggregation of many sparse individual species records (as well as records not identified to species which would be unusable in a species level analysis) into a larger more usable dataset. Also, family is a very commonly used taxonomic unit in studies of insects/arthropods/invertebrates, I gave some sources justifying this in the original post.

Its true taxonomic revisions can change a score. But keep in mind there is over 3,000 extant families of arthropods. I dont think all of them are undergoing constant enormous taxonomic revisions such that the big picture results would change substantially.

[insectilluminatigetshrekt]

Something else I want to add is that this goes in an entirely different direction than the typical studies and initiatives of arthropod conservation. I already mentioned this in the original post, but would like to elaborate on it a little bit. There is this huge emphasis on “ecosystem services” like pollination, being food for other animals, break down of organic matter, pest control (though the pests in question being controlled are usually arthropods themselves too), etc.

My idea here would ignore this framework entirely and just look at arthropods from a perspective we usually reserve for vertebrates. As unique evolutionary history. Due to the fact that species which provide “ecosystem services” usually come from large, cosmopolitan families, there might be no correlation, or in fact even a negative correlation, between regions of the world with high average EDGE scores among its resident arthropod families and how much “ecosystem services” are being provided by said arthropods.

Who or what is being “served” by members of Sigilliidae, Gelyellidae, Valeseguyidae, Mimopidae, Holothyridae, Heterojapygidae, Proterorhagiidae, Trogloraptoridae, Hadruridae, Mictocarididae, or Glypheidae? Not a lot I imagine, at least from what we know. But a lot of obscure arthropods are not only not assessed or not protected, they are essentially never even considered in a conservation context at all. When was the last time anyone asked “How are the Arguloids doing” or “What’s going on with the Schizomids”?

And an additional thing I want to mention is that places in the world with high concentrations of high scoring arthropod families are also likely to just be very unique ecosystems outright. So I think this method would be a good way to identify such systems for conservation priority, even if the threat status of any particular species cannot be ascertained. There is so many arthropod species that system level protection is often more meaningful than species level anyway.

[insectilluminatigetshrekt]

Does anyone have any thoughts on they would like to share?