Wes Jackson on why perennials and why not annuals

Jim Wood wrote last week in defense of traditional agriculture.  He was reacting to something he heard Wes Jackson, director of The Land Institute, say in an interview with Mark Bittman on the New York Times site, and that was that "agriculture is the worst mistake in the history of the human race."  Jared Diamond may be responsible for this idea gaining traction as it was the title of an article he wrote for Discover Magazine in 1987.

Fertilizer being applied to a corn field; Wikimedia Commons

Jim noticed that the idea was perpetuated on the Land Institute's website, and it got his hackles up because, as an anthropologist, he believes that most traditional agriculture isn't all bad.  Of course, yesterday's post by archeology graduate student Reed Goodman points out that depending on what is 'traditional' agriculture has had serious consequences before today's industrialized, fossil-fuel-based version that Wes is referring to.

The Land Institute is a forward-thinking, some might say radical hot-shot on the sustainable ag horizon.  They have been working for 37 years now to perennialize grain, for many important ecological and economic reasons. We blogged about The Land recently, but of course Wes Jackson describes their work much better, and more fully than we can. He is passionate because he wants to motivate people to undertake fundamental change, and that is difficult to achieve.

Indeed, Wes emailed us a reply to Jim's blog post, explaining what he thinks is wrong with modern agriculture, and why perennializing grains will eliminate some of the worst ecological effects of today's reliance on annual crops, monocrops at that.  He has kindly agreed that we post his comments here.

Wes Jackson

First off, I am not an anthropologist. I have spent little time in undeveloped countries. I have visited a few farms in Mexico, rural Chile and Argentina and also Tunisia. I am not doubting the observations of Professor Wood with his vast “on the ground” experience with subsistence farmers. There is little to disagree with on any of the “bullet points” and wondering if perhaps he is not aware that we are mostly plant breeders (perennializing the major crops and domesticating some promising wild species) and ecologists at work to put them in mixes that mimic something close to the vegetative structure of a prairie.

I am attaching a.) my preface and prologue of New Roots for Agriculture, 1980 and b.) Chapter 6 of my 2012 book Consulting the Genius of the Place. [Please email me, avbuchanan at gmail, if you would like copies of either of these.]

1.     The Land Institute is not suggesting that we go back to gathering and hunting. Too much water has gone through the turbines for that to happen.

2.     We do not criticize farmers of any variety. We are all locked into a system in which there is little opportunity for life outside it.

3.     We are saying that annual grain monocultures represent a kind of “hardware” around which we have historically built several versions of software. Some have been sustainable and have stood the test of time, but most acreage with annual hardware erodes.

4.     Most of nature’s land based ecosystems tend to feature perennials grown in mixtures. There soil capital accumulates. Grain agriculture features annuals which tend to be grown in monocultures.

5.     Here and there and now and then (sometimes for long periods) are examples where the ecological capital has been sustained, for whatever reason. It depends.

6.     We can now come up with herbaceous perennial seed producing polycultures, something humanity could not do until the present. Here is the reason. Essentially all annuals tend to self, i.e. accept their own pollen and therefore the genetic load does not build up. Also, a desirable mutation, such as shatter resistance, can be quickly fixed in the population. This makes the first step to domestication easy. Perennials are likely to have a huge genetic load. Early agriculturists would have been quickly discouraged at getting high yield as they did with the annuals through crossing. There would have been lots of aborted embryos. Also they would have had to wait a long time for a desirable trait, such as shatter resistance, to become fixed. In our time, given our computational power and knowledge not available then, we can overcome the problem of our agricultural ancestors.

7.     I am not an anthropologist and have spent little time in undeveloped countries. I have visited farms in Mexico, rural Chile, Argentina and Tunisia. I am not doubting the observations of Professor Wood with his vast experience “on the ground” visiting subsistence farmers and liking them. Seems I have been influenced by others who have studied the history of earth abuse through agriculture. To name a few: Dr. Walter C. Lowdermilk, Plato and Aristotle, Curtis N. Runnels, Paul B. Seers and many others.

8.     The United Nations says we are losing some 30 million agriculture acres per year due to agricultural land degradation. Also, one study shows that for 1700-2000 the globe lost slightly less than 1 billion acres of ag land due to degradation of some form.

Here are a few points we believe we can defend as to why we should be optimistic about the potential of perennial grain mixtures.

1.     Production: “In most parts of the world, human activities, and agriculture in particular, have resulted in decreases in net primary productivity from the levels that likely existed prior to human management. (Chris Field, 2001, Science)

2.     More Efficient:  Perennials have greater access to resources over a longer growing season.

3.     Time-tested: Diverse, perennial plant communities have been successful micro-managers of landscapes for millions of years.

4.     Reallocation Potential: Perennials have “excess capacity” that can be reallocated to grain production.

5.     Sustained yield: Perennial grains have the advantage over annuals in terms of sustained yield on marginal landscapes.

6.     The revolutionary transformation of wild species into crops has been done before (with annuals).

7.     The Millennium Ecosystem Assessment (MEA) report said: Production (of food) at the expense of conservation (of the wild biodiversity) OR conservation at the expense of production. The either-or assumption can be replaced with “conservation as a consequence of production” with perennial polycultures.

In addition to the greenhouse problem being addressed with perennial grains, there are two other realities which Natural Systems Agriculture addresses:

1.     Agriculture is the “largest threat to biodiversity and ecosystem function of any single human activity.” (MEA)

2.     Humans are poor micro-managers of complex, dynamic processes taking place across large landscapes. Diverse perennial systems are elegant micro-managers of nutrients and water.

And more succinctly: 

I do not criticize grain farmers who have been forced to come up with software to do the best with the “hardware” with which they have been stuck for 10 millennia. Several small scale enterprises have come up with software (agronomic methods) to make do, depending on where they are and the history of their culture. With the new hardware (perennials on the horizon), now ecology can enter the picture in a robust way for the first time. What has to run ahead of everything else on the global scale is an historical reality. On a large scale annual agriculture destroys the ecological capital behind our food supply. It has done so long before the industrial era because annual systems are poor micromanagers of nutrients and water.

 Summing up?
A lively discussion about how things were, are, and ought to be is a salubrious thing for the social media.  In this case, the differences of views expressed reflect different perspectives and time scales and the like, not different assessments of the issues that Wes has devoted so much attention and creative work to.  And Jim Wood, like many anthropologists, is interested in other aspects of agriculture.

The more these discussions can be aired, and the more widely, the more likely it could be that these serious issues are addressed.

The lessons of the Land: part IV

Our 3-part series last week on lessons learned in modern genomically based plant breeding was intended to address its conceptual relation to evolution and genetic causation generally, and to problems in human genetics specifically.  We hadn't intended a 4th part, but thanks very much  to conversations with Ed Buckler, the Cornell plant geneticist whose ideas were featured in the Land Report that motivated this series, we wanted to add some further comments.

The idea of genome-based selection (GS) is that you take a sample from a population, of maize or goats, say, and measure phenotypes of interest in each individual, then genotype each individual for a large number of genome-spanning variable sites (markers), just as in genomewide association studies (GWAS).  You use these data to evaluate the contribution that every marker site makes to your trait, thus optimizing a phenotype-predicting score from the genotype.  Then, you use this score to select individuals for breeding.  After a number of generations, you expect an improved stock.

This is very similar in nature to what is done with human populations in some recently advocated methods of using genomewide data to make individualized predictions.  Peter Visscher is probably the author most prominently recognized as developing these methods, though many others are now also involved.

In both human genetics and agriculture, we use a current data set of achieved traits--kernel yield, muscle mass, human stature, blood pressure or disease.  But this is retrospective assessment of genetic associations, and it may only partially reflect genetic causation.  For example, environmental factors may be unmeasured.  Also much of our variation in natural population will be captured in one but not a next sample or not exist in other populations.  These facts place some limits on the predictive power of genomewide data.

Nonetheless, like using parents to predict traits in offspring, if the genetic component is substantial (for example, by measures like heritability, or trait correlations among relatives) there must be regions of the genome that are responsible and that is what this approach finds.  How advantageous it is over measurements of phenotypes in relatives can be debated, as can the amount of contribution to the trait, like disease risk, of large numbers of very rare, never to be seen again, variable sites.  And the prediction is of a net result, which need not be (and often will not be) due to a tractable set of genome sites.  So the biomedical dream of 'personalized genomic medicine' may or may not answer the dreams of its advocates.

The idea should work much better in agricultural breeding, because the population is closed, and genetic variation is systematically, and strongly favored.  Thus, over a few generations the genetic variation can be highly enhanced in the desired direction--at least under the controlled environmental conditions of the selective breeding.

The discrepancy between breeding experience, and the observational setting of human biomedicine--and of evolutionary biology--may, if carefully considered, provide ways in which the former can inform the latter.  There are reasons to think that important changes in view may result.

Learning the lessons of the Land: part III

This is the third and final post on a recent article in The Land Institute's Land Report, describing advances and methods to identify and isolate desirable genetic variation in plant species, with the goal of sustainable agriculture by scientific, but efficient, methods.

This is pure modern genetics, combined with traditional Mendelian-based empirical breeding as has been practiced  empirically over many thousands of years and formally since the mid-19th century.

The discussion is  relevant to the nature and effects of natural selection, which, unlike breeding choice, is not molecularly specific and is generally weak.   That's why it's difficult to find desirable individual plants in the sea of natural variation, and why intentional breeding is so relatively effective: with a trait in mind, we can pick the few individual plants that we happen to like, and then isolate them for many generations, under controlled circumstances, from members of their species without that trait.

By contrast, natural selection seems usually to act very slowly.  Among other things, if selection were too harsh, then perhaps a few lucky genotypes would do well, but the population would be so reduced as to be vulnerable to extinction.  Strong selection can also reduce variation unrelated to the selected trait, and make the organisms less responsive to other challenges of life.  If environments usually change slowly, selection can act weakly and achieve adaptations (though some argue that selection has its main, more dramatic effects very locally).

With slow selection, even if consistent over many generations, variation arises at many different genes that can affect a trait in the favored direction.  Over time, much of the genome may come to have variants that are helpful. But they may do this silently: even if variation at each of them still exists, there can be so many different 'good' alleles that most individuals inherit enough of them to do well enough to survive. But the individual alleles' effects may be too small to detect in any practical way.

These facts explain, without any magic or mystical arguments about causation, why there is so much variation affecting many traits of interest and why their genetic basis is elusive to mapping approaches such as GWAS (genomewide association studies).  

Of course, even highly sophisticated breeding doesn't automatically address variable climate, diet, etc. conditions which can be relevant--indeed, critical, to a strain's qualities.  Molecular breeding is much faster than traditional breeding, but still takes many generations.  Think about this:  even if only 10 generations, in humans that would mean it would take250 years (the age of the USA as a country) to achieve a result for a given set of conditions.  So how could this kind of knowledge be used in humans....other than by molecular based eugenics (selective abortion or genotype-based marriage bans)--days we surely don't want to  return?

Breeders might eventually fix hundreds of alleles with modern, rapid molecularly informed methods.  But we can't do that in humans, nor as a rule identify the individual alleles, because our replicate samples come not from winnowing down over generations in a closed, or controlled, breeding population, but from new sampling of extant variation each generation, in a natural population.  

The data and molecular approaches seem similar in human biomedical and evolutionary genetics, but the problem is different.  As currently advocated, both pharma and 'personalized genomic medicine' essentially aim at  predictions in individuals, based on genotype, or treatment that targets a specific gene (pharma will wise up about this where it doesn't work, of course, but lifetime predictions in humans could take decades to be shown to be unreliable).

It's hard enough to evaluate 'fitness' in the present, much  less the past, or to predict biomedical risk from phenotype data alone, though such data are the net result of the whole genome's contributions and should be of predictive value.  So how to achieve such prediction based on specific genotypes in uncontrolled, non-experimental conditions, if that is a reasonable goal, is not an easy question.

In ag species, if a set of even weak signals can be detected reliably in Strain B, they can be introduced into a stock A strain by selective breeding.  It need not matter that the signals that only explain a fraction of the desired effect in strain B aren't detected by the mapping effort because repeated iteration of the process can achieve desired ends.  With humans, risk can be predicted to some extent, from GWAS and similar approaches. But so far most of the genetic contribution detected has been elusive, weakening the power of prediction.

In humans, the equivalent question is perhaps how and when molecular-assisted prediction will work well enough in the  biomedical context, or in the context of attempting to project phenogenetic correlations back into the deep evolutionary past accurately enough to be believable.  Perhaps we need to think of other approaches.  Aggregate approaches under experimental conditions is great for wheat. But humans are not wheat.

Learning the lessons of the Land: part II

This series of commentaries (beginning yesterday) was inspired by the latest issue of the Land Institute's Land Report, that describes efforts to use modern science to develop sustainable crops that can conserve resources yet feed large numbers of people.  We were motivated by the thought that not only is this important work, but it should inform our ideas about human--and evolutionary--genetics.

Van Gogh, Farmhouse in a Wheat Field, public domain

In our previous post we introduced the idea of molecular breeding, a genomewide association study (or GWAS)-like approach that experimental breeders in agriculture are taking to speed up and focus their efforts to breed desired traits into agricultural plants.  Here, we want to continue that discussion, to relate the findings made in agricultural genetics to what is being promised for GWAS-like based personalized genomic medicine.

Essential personalized medicine means predicting your eventual disease-related phenotypes from your inherited genotype (and here, we'll extend that beyond just DNA sequence, to epigenetic aspects of DNA modification, assuming that will eventually be identifiable from appropriate cells).

If breeders had been finding that once seed with desired traits had been identified, genome-spanning genetic markers (polymorphic sites along the genome) pointed to a small number of locations with big effects, then we would quickly be able to find, and perhaps use the actual genes diagnostically.  This seems to be true for some plants with small genomes, for traits that seem to be due to the action of variants in one or only a few genes.  This is just what we find for the 'simple' human diseases, or the subset of complex diseases that segregates in families in a way that follows Mendel's principles of inheritance.  There are many examples.

But for many traits, including most complex, delayed onset, life-style related, common disorders that are the main target of the GWAS-ification of medicine in the Collins era of NIH funding, what is being found is quite different.  Mapping is finding hundreds of genes, almost all of which have either very small individual effects, or if larger effects, that are so rare that they are of minimal public health importance (even if very important to those who carry the dangerous allele).  For these, the question is what to do with the countless, variable regions of the genome that make up the bulk of the inherited risk.

This is the same situation as faced in the agricultural breeding arena, for many of the traits, like water- or drought- or pest-tolerance, nutrient yield, or other characteristics desired for large scale farming.  The traits are genomically complex. Even with large samples and controlled and uniform conditions--very unlike the human biomedical situation--it is not practicable or practical to try to improve the trait by individual gene identification.  Nor is it likely that introducing single exotic transgenes will do the trick (as many agribusinesses are acknowledging).

Instead, molecular breeding takes advantage of the plant's own natural variation to select those variants that do what is desired simply by choosing the plants that transmit those variants, and without attempting to engineer or even to identify what they are.  We do not know how reliable the prediction of phenotype from genotype in these circumstances typically is, but the idea is if that you keep selecting plants with the desired regions of the genome that mapping identifies, and breeding them for example with strains that you like for other reasons, you can reduce reliance on individual prediction because eventually every individual will be alike, for the traits you were interested in.

Once that is the case, regardless of the genes or regulatory regions that are involved, you have your desired plants, at least under the conditions of nutrients, climate, and so on, in which the strain was developed.

Clearly this experience is relevant for human genetics, and for evolutionary genetics of the same traits, a topic to which we will turn in our final post in this series.....

Learning the lessons of the Land: part I

This post is inspired by the latest issue of The Land Report, the thrice-yearly report by The Land Institute, of Salina, Kansas.  This organization is dedicated to research into developing sustainable agriculture that can conserve water and topsoil, reduce industrial energy dependence, and yet produce the kind of large-yield crops that are needed by the huge human population. 

Minnesota cornfield, Wikimedia Commons

An article in the spring 2011 issue concerns an approach called molecular breeding.  Here the idea is to speed up traditional empirical breeding to improve crops, as a different (and better, they argue) means than traditional GM transgenic approaches, that insert a gene--often from an exotic species such as a bacterium--into the plant genome.

The important point for Mermaid's Tale is that crop breeders have been facing causal complexity for millennia, and from a molecular point of view for decades.  Their experience should be instructive for the attitudes and expectations we have for genomewide association studies (GWAS) and other 'personalized genomic medicine.'   To develop useful crop traits means to select individual plants that have a desired trait that is genetic--that is, that is known to be transmitted from parent to seed, and to replicate the trait (at least, under the highly controlled, standardized kinds of conditions in which agricultural crops are grown).  For this to work, one needs to be able to breed, cross, or inter-breed seeds conferring desired traits to proliferate those into a constrained strain-specific gene pool.  Traditionally, this requires generations of breeding, and selection of seed from desired plants, repeated for many generations.

The idea of molecular breeding is to use genome-spanning sets of genetic markers--the same kinds of data that human geneticists use in GWAS--to identify regions of the genome that differ between plants with desirable, and those with less desirable, versions of a desired trait.  If the regions of the genome that are responsible can be identified, it is easier to pick plants with the desired genotype and remove some of the 'noise' introduced by the kind of purely empirical choice during breeding that farmers have done for millennia.

Relating phenotype to genotype in this way, to identify contributing regions of the genome and select for them specifically is in a sense like personalized genome-based prediction.  As discussed in the article, 'Biotech without foreign genes', by Paul Voosen in The Land Report (which, unfortunately, doesn't seem to be online) molecular breeding is a way to greatly speed up the process of empirical crop improvement.  What we mean by empirical is that the result uses whatever genome regions are identified, without worrying about finding the specific gene(s) in the regions that are actually responsible (this means, in technical terms, using linkage disequilibrium between observed 'marker' genotype, and the actual causal gene).

For crops with small genomes, like rice, breeders have been more readily able to identify specific genes responsible for desired traits.  But for others, the large size of the genome has yielded much more subtle and complex control that is not dominated by a few clearly identifiable genes. Sound familiar?  If so, then we should be able to learn from what breeders have experienced, as it may apply to the problem of human genomic medicine and public health.

We'll discuss that in our next post.

Something to root for

Yesterday, we commented on the typical kind of societal, scientific, and biological complexity associated with genetically engineered food crops. Will evolution have its way, or will we keep one step ahead of famine? The issues have to do with world population demanding industrial-scale agriculture (or, at least, that's how modern societies are constructed).

Assuming that climate change doesn't change the entire name of the game, Mega-dust-bowl fashion, we are still feeding the world on the back of 'portable' fossil fuel (for fertilizers, insecticides, tractor and shipment fuels, and so on). We lay out huge fields of a single crop, as far as the eye (or a soaring eagle's eye) can see. These are largely grasses, the staple foods of most of the world. But the way we do this leads to disaster.

We are losing topsoil faster than a stripper loses clothes, and pretty soon farmland will be as bare as a stripper, too. We have water pollution problems due to runoff not just of soil but of fertilizer and pesticide. One reason is that we rely so heavily on annual grass crops. These need to be plowed under each year, baring and loosening the soil, which makes it vulnerable to runoff.

And mega-agribusiness removes family farms and their communities. Small farms traditionally had a mix and rotation of crops, grazing and other livestock (and their manure), and so on. And though the work was very hard, they had community. This lifestyle is not for everyone, but many pine for it, as it disappears.

However, a different kind of genetically engineered crop is being worked on by a few researchers in the US and elsewhere. Most notable is The Land Institute, in Salina, Kansas. There, under the direction and inspiration of Wes Jackson, new ideas are taking root. Many wild grasses related to commercial food grasses are perennials rather than annuals. They last for years, keeping roots to anchor the soil, reducing the need for fertilizers, keep the topsoil in place, and so on.

Jackson and his group of dedicated experimenters are using traditional agricultural breeding, and where possible, scientific methods, to get commercially useful annual crops to take on the relevant aspects of perennial stock, so that we can have perennial wheat, corn, and other crops. With even modest investment, they will likely be able to succeed. If they do, it will meld real science, both high-tech and traditional, with sound long-term future implications for our physical bellies and our social hearts.

Nobody claims that this will be easy, but it is easy to claim that we have little choice. The Land Institute has a web page where you can learn about these efforts: landinstitute.org