Genereric pest spread models

Assessing spread of quarantine pest under new climatic conditions is not an easy task. This kind of model must combine climatic, spatial and pest climatic suitability data. Unfortunately that is not enough in most cases. To make model that resembles reality it should also include pest biology: it’s movement abilities, reproduction, maximum population density, host availability etc. Taking all above into account we may conclude that it is extremely hard to make generic spread model.

I am aware of only one suite of this kind of models (actually this suite include only one sensible generic spread model in my opinion). It combines data on climatic niche suitability from CLIMEX and basic biology of pest, to make an estimate of it spread in spatio-temporal manner. This suite is broadly described in Robinet et al. 2012 and also contain quite nice vignettes explaining how to use it.

Problems occurs when you try to execute their script on newer versions of R and raster package. It appears that function xyValues() is no longer supported by raster package giving you error:

Error in xyValues(host, cbind(xygrid[, 2], xygrid[, 1])) : could not find function "xyValues"

The solution is surprisingly easy, yet quite hard to find for people not really familiar with geo analysis. In newer versions of raster package, function xyValues() was exchanged with extract(). Simply find&repalace schema will make this code working.

Comment

Also this suite underscores problem of nowadays science. People engaged in research (I intentionally withhold from calling them Scientists) are in constant chase for money provided by bureaucrats. Of course, those money come from tax payers, whom do not have anything to say when it comes to how their money should have been spent. In this long lasting game, specific rules were coined and preserved, so actually no one involved will benefit from change. At least not immediately. In result we face situation presented above. Some years ago quite nice script was developed, but than left to rotten without proper maintenance. The reason is very simple. You won’t get points for maintaining and updating your work. You won’t be able to publish updated script or package. You won’t get a position in Academia when you spent to much time on maintaining or updating your work. The only thing that matters are new papers and new grants. Thus it is not in the interest of people involved in research to spent time on old things. They need to publish NEW THINGS ass many as possible, write as many grant proposals as possible. It is also beneficial for bureaucrats. They always can say to tax payers: Look how much projects we were able to fund with money we took from you. What nobody says is that results are biased, job is only half done, and half of this projects reinvent the wheel.

Robinet C, Kehlenbeck H, Kriticos DJ, Baker RHA, Battisti A, Brunel S, et al. (2012) A Suite of Models to Support the Quantitative Assessment of Spread in Pest Risk Analysis. PLoS ONE 7(10): e43366. https://doi.org/10.1371/journal.pone.0043366