Meta-analyzing diversity

If you’ve just arrived from Tangled Bank, welcome. And be aware that there’s a follow-up post.

A couple of meta-analyses on the menu today.

ResearchBlogging.org Devra Jarvis and Bioversity International colleagues, together with numerous co-authors from national programmes around the world, have a paper in PNAS summarizing the results of a 10-year effort to establish the scientific bases of on-farm conservation of agrobiodiversity. ((Jarvis, D.I., Brown, A.H., Cuong, P.H., Collado-Panduro, L., Latournerie-Moreno, L., Gyawali, S., Tanto, T., Sawadogo, M., Mar, I., Sadiki, M., Hue, N.T., Arias-Reyes, L., Balma, D., Bajracharya, J., Castillo, F., Rijal, D., Belqadi, L., Rana, R., Saidi, S., Ouedraogo, J., Zangre, R., Rhrib, K., Chavez, J.L., Schoen, D., Sthapit, B., Santis, P.D., Fadda, C., Hodgkin, T. (2008). A global perspective of the richness and evenness of traditional crop-variety diversity maintained by farming communities. Proceedings of the National Academy of Sciences DOI: 10.1073/pnas.0800607105))

Varietal diversity ((The unit of analysis was the farmer-recognized and named variety.)) data on 27 crops grown on 64,000 ha by 2,041 households in 26 communities in 8 countries on 5 continents were pulled together in a stunning feat of synthesis. Are any generalizations possible from such a massive dataset? Well, perhaps surprisingly, yes. Let me pick out the highlights:

  1. Households growing traditional varieties generally grow more than one (1.38-4.25).
  2. Households within a community tend to grow somewhat different sets of traditional varieties.
  3. Larger fields generally have more traditional varieties, but smaller fields tend to be more different in varietal composition.

There’s much more to this rich analysis than that, but the take-home message can be pretty easily stated: crop genetic diversity can still be found on-farm because even neighbouring families choose to grow different traditional varieties, and generally more than one. Especially families tending smaller fields, who will presumably be poorer and living in more marginal conditions. The conoscenti will recognize a familiar meta-narrative, but it is good to have solid data from a wide range of crops and from all over the world.

The next paper I want to talk about looked at genetic diversity in wild clonal species as it relates to their breeding system. ((Honnay, O., Jacquemyn, H. (2008). A meta-analysis of the relation between mating system, growth form and genotypic diversity in clonal plant species. Evolutionary Ecology, 22(3), 299-312. DOI: 10.1007/s10682-007-9202-8))

Summarizing 72 genetic diversity studies, including of a couple of crop relatives, the authors found that populations of self-incompatible clonal species tended to have fewer genotypes, more unequally distributed (i.e., with a few dominant clones), than populations of self-compatible clonal species. It would be interesting to see if this relationship is also present in vegetatively propagated crops. I don’t think the previous dataset would help with that, however. Only two clonal crops were included in the on-farm analysis, cassava and taro. Interestingly, they had the highest average levels of community-level varietal richness (33) compared to seed-propagated species.

9 Replies to “Meta-analyzing diversity”

  1. Aside from an obscure reference to “crop complexes” in a table caption, I couldn’t find anything in the paper on intercropping. Also, they say that “community richness (omitting cassava) is 8-fold that of farm richness.” So there may be high diversity at the landscape scale, moderate diversity within farms, but not necessarily much diversity at the scale of meters. One question I will be exploring in Darwinian Agriculture, and which I would welcome input on, is how the benefits from a given level of diversity depend on the spatial and temporal patterns in which it is deployed.

  2. Yes, this paper did not look at intercropping, but I’m sure the project generated plenty of data on inter-specific diversity, and no doubt in time there will be a meta-analysis of that too. Regarding your request for input, if you’d like to write a paragraph or two of background, we could publish it as a regular post for more visibility.

  3. Thanks, Luigi, for your thoughtful comments. Perhaps another point that could be made is that the Jarvis et al. data on crop diversity from these 8 countries and 27 crops shows that there is a close relationship between their richness measures (varietal diversity) and eveness – particularly for farm level analyses. This may be worth following up as we look for indicators of diversity on farm that could be used nationally or globally – we may be able to explore how measures of eveness might be used to provide such indicators.

  4. @Toby

    I’m not sure how you intend to use evenness as a measure of diversity, when the paper suggests that there are at least two different kinds of relationship.

    For staples, evenness is low for a given level of richness, suggesting that farmers are growing a little of the “extra” varieties as insurance for the future.

    For non-staples, evenness is higher for a given level of richness, suggesting that farmers are growing all the varieties in the quantities they need for some immediate use or purpose.

    So what is the nature of the “close relationship between … richness measures (varietal diversity) and eveness” that will be so illuminating across crops, countries and regions? Is it the same for each crop species?

  5. I don’t understand this long-term insurance explanation of the difference between the evenness/richness relations of staples and non-staples.

    I would expect that farmers grow varieties for the market or bulk consumption in large quantities while planting speciality varieties of these same crops in small quantities, just for direct needs or petty trade (like non-staples).

    Wouldn’t that be a more obvious explanation for the observed pattern?

    Speciality varieties of staple crops defy the distinction between staples and non-staples.

Leave a Reply

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