Abstract

Jay M. Ver HoefI have enjoyed reading the paper by Gelfand et al. (2005). My congratulations goto the authors, as they have given us an important advance in the science of modelingspecies diversity. First, I would like to emphasize the importance of this topic. I fullyagree with the authors that species diversity has been a central concept in ecology formany years, yet the mechanisms that determine species diversity are still enigmatic.How then has this paper helped us?One of the rst problems in assessing species diversity is to know where a speciesoccurs. While this may seem simple, it is actually very dicult. The authors have avery ne data set that was systematically sampled in a very interesting, diverse part ofthe world, where high species diversity is compacted into a relatively small space. Oneof the questions that I want to ask is, \Can the methods of Gelfand et al. (2005)be usedmore generally? That is, can I use them in Alaska? Alaska is a rather large state, butif we consider plants, being far to the north, it is not really very diverse. We know ofonly about 1600 di erent plant species in Alaska. Rhode Island has more plant species(2600). The methods of Gelfand et al. (2005) are fairly complex, but in principle itseems that they could be adapted for hundreds (perhaps thousands) of di erent plantspecies as computational power increases. However, for a more general application,there are problems with species presence data that do not occur for Gelfand et al.(2005). Sampling has not occurred uniformly over my state, or any large geographicarea that I know of. For example, I’m pretty sure that if we added a covariate such asdistancetothenearestuniversity,therewouldbeahighlysigni cant,negativeregressioncoecient when modeling species presence or diversity. The reason is clear. For years,botany professors have been sending out legions of graduate students and classes tocollect plants, and they stay relatively close to home. Thus, not all zeros are createdequal. This is known as ascertainment bias in the epidemiology literature. Gelfand etal. (2005) have done an outstanding job in distinguishing other factors that do createzeros, such as transformed landscapes. This is an important step, but it is informationthat is relatively easy to gather as compared to e ort. Eventually, it will be importantto solve the e ect of e ort (ascertainment bias).Now, what about prior information? Gelfand et al. (2005) use a hierarchical modelwith vague priors. This makes sense, given the complexity of the model. Eliciting priorsfrom most plant collectors that I know would be very dicult. It would be hard forthem to make sense of priors on parameters in a model with the complications of thepotential and transformed surfaces, hidden random e ects, etc. Still, these same plantcollectors have a wealth of prior knowledge; they have spent years crawling through thebushes. Early in my career I collected plants as my job, and I lived by the maps drawnin Hulten’s (1968) Flora of Alaska. It was a big deal to extend any of the species rangesdrawn in his book. Plant collectors, such as Hulten, simply used their experience and

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call