We have moved!

The bigger, better, brand new DAPA blog is here (link)


Please note this Blog is not updated anymore.

We have moved! -- CLICK HERE --
Decision and Policy Analysis Research Area – DAPA
Homologue

Homologue

Homologue

Description

In the early 1980s, David Wood arrived one day in CIAT with a collection of Gliricidia germplasm and asked me where it would be best to plant it for conservation and future use. This led me to produce a series of algorithms for comparing climates from the CIAT climate database. Tony Bellotti used these to great effect in planning the transfer of predators of cassava mealybug (Phenacoccus manihoti Matile-Ferrero) from Latin America to Africa. About 10 years later, Nick Galwey came back to CIAT after some years’ absence, and together we worked up the central algorithm for what became FloraMapTM.

FloraMap is an algorithm for mapping the distribution of plants and other organisms in the wild. It works on the premise that we know nothing about the organism other than the geographic location of a calibration set of collection points. From these we fit a climate probability model. This approach has had considerable success, and is being used widely. However, it has some major drawbacks for many applications: it requires a calibration set, it only works on climate, and it has not been used successfully on cultivated crops where the farmer alters the environment.

So, what do we do for those who ask the simple question, “Where else in the world is like my plot of land?” We have no calibration set. We do not know what species we are considering. We do not have an algorithm for predicting the probabilities of relevant soil characteristics. The question may be simple, but the answer is not. Homologue has been developed to cope with the complexities of this simple question. Homologue uses the basic algorithm of FloraMap, generalized to fit a range of generic species designated by the user. It incorporates statistical probability calculations for the mapping of soil characteristics. If we know where else in the world is like my plot of land, we can infer, from the agricultural practices there, what may be applicable to my plot.

Homologue also can be used in the reverse sense. If we find an instance of the cultivation of an interesting species, we may look at where else it might be used. If we have a number of these instances, we might look for an environmental envelope, or “Cloud” as James Cock terms it, where the species might fit. Homologue therefore allows the combination of the probability maps for various
sites that circumvents the problem that FloraMap has with small sample sizes.

Download

Share this:

Leave a Reply

Your email address will not be published. Required fields are marked *


*

about CIAT

If you could answer these three short questions, that would be really appreciated http://dapa.ciat.cgiar.org/we-want-to-know-our-readers/

Our Latest Presentations