Feeding Wessex without fossil fuels

The last time we were in Wessex, I showed that its denizens circa 2039 could probably feed themselves quite comfortably using organic farming methods with 20% of the population concentrating largely on neo-peasant subsistence farming using 40% of existing lowland farmland, and the remaining 80% of the population fed by larger-scale, more cereal staple oriented farming from the remaining 60% of the farmland, plus a bit of upland grazing.

However, as it stands that scenario does depend on a fossil energy-intensive ‘business as usual’ approach on the large-scale farms. It seems worth pondering an alternative, zero fossil energy scenario. Here we begin to exceed even my own generous comfort zone for idle speculation about the future – if there’s no fossil fuel use in Wessex farming in 2039 (or beyond), what might be the social and economic correlates? Probably not one with 80% of the population still happily residing in towns and working as video game programmers, conservatory salesmen or whatever.

Still, I don’t propose to worry about that too much in this post. For now, let’s just consider the farming side of it, and see if we can find another way to power the food production for 80% of Wessex’s population.

That immediately plunges us into a speculative debate about the shape of the future energy mix which could go on until…well, 2039. So here I’m going to curtail it brutally by making the following doubtless highly debatable assumptions. I’m going to assume that there won’t be enough renewably generated electricity to power electric, fuel cell or electro-synthesised hydrocarbon tractors. I’m going to assume that none of the magic, much-touted next-generation or generation-after sources of limitless clean power such as thorium or nuclear fusion have come through. And I’m going to assume that wood methanol isn’t a viable source of agricultural energy, as a couple of people have suggested to me that it might be. The way I read the runes on that one is as follows:

You get about 27 litres of methanol from a tonne of wood, and you get about 3 tonnes of wood from a hectare of managed woodland, so you get about 80l of methanol from a hectare of woodland. Methanol has about half the energy density of diesel. You need about 100l of diesel (so 200l of methanol) to farm a hectare of arable land each year. I’ll assume you need about a quarter of that to farm a hectare of permanent grass, minimally, about as much again to manage the rest of the production and transport economy around food. That works out at about 1.2 million hectares of managed woodland to service 1.8 million hectares of farmland, which would exceed the land area of Wessex by nearly a third (while also neglecting the energy needs of the woodland management). Methanol can be made from other carbon-rich waste, but it seems to me a stretch to think it could be a major agricultural energy source unless anyone can provide some radically more promising figures.

Another suggestion I received was to put aside my West Country obsession with cows and make methane instead of milk from the grass via anaerobic digestion. Now, I’ve always regarded these straight-to-methane schemes as a dastardly vegan plot to deny me the froth I so badly need on my morning cappuccino, but after crunching a few numbers I’ve got to admit that the plan has something to commend it. In fact, the numbers seem to stack up so spectacularly well that I feel I must have made a terrible error somewhere, so let me run through my arithmetic in some detail with the hope that someone can either corroborate it or else point out the error of my ways.

Let’s start by calculating how much energy we need to run our Wessex food system. I’m going to assume that we need 100 litres of diesel per hectare on the farm for arable operations, and 25 litres for grassland management. Then to fuel the entire food economy from farm to fork, I’m going to assume we need another 200 litres of diesel equivalent per hectare (for both arable and grassland) – an assumption loosely based on the emissions scenarios in Tara Garnett’s Cooking Up A Storm. Diesel has an energy content of 38.6 MJl-1. So if we take our 166,000 ha of cropped arable at 300 l/ha diesel and our 795,000 ha of permanent grassland and arable ley at 225 l/ha and multiply that sum by 38.6 MJ we get a total energy requirement of about 8.8 billion MJ (or 8.8 PJ if you prefer).

On the supply side I’m assuming 20 tonnes of fresh silage per hectare1 (or 5.5 tonnes dry matter), grown organically (average conventional yields are more than double that), and 160m3 of biogas per tonne of silage2, with an energy content of about 22 MJ/m3 – so that works out at about 68,000 MJ/ha. If we take a quarter of our permanent pasture – some 223,000 ha – and set it aside for silage as biogas feedstock, that’ll give us 15.1 PJ of energy, which is nearly double our energy requirement. As I understand it, methane-powered tractors are already a reality at engine efficiencies similar or above those of conventional diesel, and though the biogas coming out of the digester needs a bit of refining, the process efficiency is quite high. Embodied energy of plant construction seems to turn out at around 10% of total energy output3, so the overall energy costs seem manageable.

Obviously we need to re-run our food productivity figures in the light of taking out a quarter of the permanent pasture (hopefully rotating cows over it and returning some or all of the digestate to it will keep the silage production sustainable). But since this part of the farm system otherwise produces relatively low-output grass-fed cows, the overall loss of productivity may not be too severe. And so it proves – removing 25% of the permanent pasture for biogas drops the supply/demand ratio for food energy from 1.07 to 0.99, with all the other nutritional ratios remaining >1. An energy ratio of 0.99 is doubtless a bit too close for comfort, but it shouldn’t be too difficult to find an extra bit of productivity. The lazy way would be to plough some more permanent pasture for wheat – about 22,000 ha or 3% of the total permanent pasture diverted to wheat would restore food energy productivity to the 7% surfeit we were experiencing with fossil diesel (call it 6% to make provision for a ley). But there would be other more elegant, if more labour intensive, ways of doing it. And remember that I’m making a lot of conservative assumptions about yields.

Originally I’d been thinking in terms of biodiesel from oilseed rape as the way we’d have to go in a fossil-fuel free Wessex. That method produces almost, but not quite, as much fuel energy per hectare as biogas from organic silage, but only by devoting a big chunk of precious cropland to the oil crop. And the rape would have to be grown conventionally, using synthetic fertiliser and pesticides, with additional energetic and environmental implications. An advantage of rape is that the meal or press cake from the oil extraction process yields a high energy livestock feed, which partially compensates for the loss of cropland. But rape just doesn’t seem to me to stack up as well as biogas – particularly since it looks like I can keep enough cows to get my cappuccino in the morning and still have fuel to start up the tractor. Another advantage of anaerobic digestion and biodiesel over the photovoltaics we were discussing in my last post is that the basic engineering technologies in both cases seem simpler, which perhaps gives them a better chance of making it through the climacteric as per the previous discussion.

Well, there you have it. As I’ve said many times before, I’m not trying to suggest in this exercise that it would a simple or even a likely thing for a future Wessex to feed itself, especially if it were as energy-constrained as the one I’ve been discussing here. I don’t want to come over all ecomodernist (not that ecomodernists have much time for such down home energy technologies as anaerobic digestion). But my proposition for discussion is that it may be a possible thing.


  1. See, for example, the Organic Farm Management Handbook, or this.
  1. http://www.biogas-info.co.uk/about/feedstocks/
  1. http://opus.bath.ac.uk/22984/1/UnivBath_PhD_2010_W_Mezzullo.pdf

40 thoughts on “Feeding Wessex without fossil fuels

  1. I know very little about the technologies involved and was surprised by the amount of energy that could be harvested growing grass as feed-stock for bio-gas. Back in the summer I was watching silage being harvested (my 2yr old son loves it) – big self-propelled forager and several large tractors pulling trailers – I’d been reading about BECCS and seeing that machinery at work my first thought was that BECCS must be pipe dream, that there simply wouldn’t be sufficient net energy to make it a realistic proposition. Your figures suggest that it may indeed be possible (assuming carbon capture technologies that don’t really exist yet).

  2. Would there be any feed value to whatever was left over after making methane in the way you talk about here?

    And I’m curious why you don’t mention draft animals in this post. I’m far from convinced that they’d be your answer, but I’d be interested in hearing your thoughts about draft animals and their cost-benefit analysis for future Wessex. The fact that they were the main power source on Wessex farms before fossil fuels suggests to me that they deserve very strong consideration for a post fossil fuel scenario.

    • As someone familiar with draft animals and I believe that they could be the key to our future but for some strange reason they are rarely considered.

  3. Congrats, you’ve done it. All our troubles and difficulties dashed to the sidelines. What a relief.

    Ok, too snarky. And I really shouldn’t go there. I actually like the biogas (or at least the biomass) approach. With all the grass production you have ready alternative use (feed for herbivores) which allows you some flexibility to manage longs and shorts. In a very favorable year you should have excess grass (which can be dried or ensiled (at appropriate moistures) and thus saved for lean times. Dried grass can be rehydrated and then digested for fuel OR fed to herbivores such as your froth producing cows. Silage also makes an excellent feed.

    It would be great if there were some herbivore that could serve as a converter of excess grass to meat on a fairly short time scale. You know, something that could scale up in months rather than years to take advantage of a little surplus feed when available. Let me see, goats and sheep… not too bad, some twinning, so flocks could be ramped up faster than cows. But isn’t there some critter we could turn to that is even more fecund than sheep, and suitable to small farm production? My hunch – build a hutch… quick like a bunny. 🙂

  4. I think you are overthinking this wayyy to much.
    How would society look like with a non-fossil fuelbased agriculture? Look in your historybooks, chapters prior to the industrial revolution and you get a pretty good idea.
    A lot of the folks simply will not be there anymore. The herd will be culled drastically and those remaining will most likely live a rural life, growing a lot of their food themselves.
    No society in their right mind would divert farmingland from the direct production of food to directly feed the people. There will be no need to go the farmland-crops-transport-productionplant-transport-machines-farmland-food-people route.
    The route will be a lot simpler, a lot shorter and will feed a lot less mouths.

  5. Chris:
    May I draw attention to the SRUC link you list in the first note… Grassland Production & Utilisation?? In it, on page 2 where they consider re-seeding… first paragraph… “New varieties on average have been increasing yield by approximately 1% per annum.” This sounds strangely like the rate of progress due to efforts at plant breeding. If a grass sward is re-seeded every 6 years one might hope for an increase in productivity (due to improved varieties) of just over 5% (compound interest on a per annum basis). Earth shattering? Well, no… but hardly the sort of result to be ignored either.

    • Periodic re-seeding is easier said then done, especially in an organic neo-peasant context. How would you re-seed if not by killing the whole stand of grass with herbicide and then no-till drilling to new varieties? Over-seeding is another possibility, but that will yield a much more marginal benefit (if any at all) without any comparable reduction in the seed cost.

  6. Yes Eric, most anything is easier said than done. And to really take advantage of newer genetics one would want to replace rather than over-seed. As a piece of a rotation where the pasture is used as a ley and a year or so of other crops are part of the rotation you might avail yourself of a clean tilled seedbed. No-till seeding into wheat or rye stubble might work as well. The rotation described in the same publication (page 4 – Michael Sannon’s rotational grazing piece) starts with silage followed by swift or kale; in year 2 cereals are “undersown” (which I hope someone in England understands… it gets past me)… anyway, subsequent years are grazed and it leaves me to imagine the seeding of forage species follows the cereals.

    Mr Shannon’s grazing cycle goes out to as many as 7 years, one presumes the quality of the stand is a consideration in a decision to restart the rotation. I picked 6 years above.

    To be completely transparent here Mr Shannon is using fertilizers in his system and a fair level of infrastructure to manage the system described in the piece. But these latter specifics of the example system don’t automatically negate the possibility that one can seed a new pasture using the latest genetics in an organic or any other appropriate system. Availability of organic seed stock may be an issue, but with the organic systems working here in the States there are ways to accommodate a first use of a seedstock that is not yet available in the market in an organic form.

    So, not as easily accomplished as said… but very little seems to be from where I stand.

    • I think we ought to recognize more than just a generic “easier said than done” rule to all of farming, true as that may be. There’s a huge difference between the *relative* applicability of re-seeding practices to large, heavily mechanized, conventional farms and to small organic farms, and that relative difference is what I meant to point out. Re-seeding has its applications in just about any style of grass farming — I didn’t mean to deny that in any way — but I think the potential annual gains attainable from genetic improvements to pasture/hay species in a small-scale organic context would be a whole lot closer to 0% than 5%. I also strongly suspect that the gains you’re attributing to genetic improvements in a conventional context should really be contributed more to synthetic and nonrenewable fertilizers (and I’m highly skeptical of compounding annual 5% gains even then): it may be necessary to breed new strains of grasses to realize the gains achievable with such high input systems, but in that case I don’t think it’s fair to say that breeders are breeding higher yielding grasses (and other forages) so much as they’re breeding grasses that are more compatible with unsustainably high input levels. I would be inclined to characterize a lot of modern breeding “successes” (including much of the “green revolution”) similarly.

      • I have no quarrel with looking beyond “easier said than done”… and I will also stipulate that the relative difference(s) between large scale mechanical farming and smaller scale organic is real (though I’m less convinced the difference is “huge” as far as varietal choice is concerned).

        If you will have a look back at the first comment where I indicated the rate of increase in yield attributed to new varieties you see that the annual rate is estimated at 1%. This is the rate published in the cited SRUC document and not something I “think” is real. I took the liberty of calculating the expected rate of increase after an intervening 6 years in the rotation (1% per year, and then a 1% on top of that for the following year… and so forth. Compound interest accrued annually). For instance, Farmer Brown seeds a pasture in 2007 with the most recent varieties available to him. In 2013 he cuts the pasture for silage and seeds kale for a winter forage. This is followed by a cereal and then the pasture is reseeded in 2014 to grass for grazing. Regardless of the size of his operation and the relative difficulty of his effort to reseed this pasture – it is a rotation worthy of his time and effort independent of the grass variety he chooses to plant.

        To your point of the grasses in question being bred for “synthetic and nonrenewable fertilizers”… I will merely observe that all sources nitrogen fertilizer other than atmospheric N2 are synthesized… some by us, some by soil bacteria, some by blue-green algae, and some by lightning. Whether or not the nitrogen we humans fix is sustainable actually calls for a rather specific LCA and can take us far off the current path. And without much real benefit to the current discussion – because even if we callously dismiss human synthesized nitrogen as unworthy we would miss the point that regardless of source of N these grasses are improved over previously available strains.

        There is no citation in the SRUC piece for the assertion that new grass varieties have realized a 1% per year increase in productivity. There are measures of how much nitrogen Michael Shannon uses in his farming – and his levels may well be *relatively* more difficult for a purely organic producer to match. But your strong suspicion isn’t supported by anything yet in evidence here.

        Changes in plant responses due to breeding can be realized at all levels of resource input. While not a grass breeder myself, I do know a little about breeding with a certain legume grown on more than a couple hundred million acres all around the planet every year. The conditions under which soy is planted range from very high input to very low. At both ends of this input range there are easily demonstrable gains in grain yield due to simple and sustainable plant breeding efforts.

        The green revolution trope you raise actually makes my point. There are wheat and rice genetics today that work better under limited resource conditions than did the prevailing wheat and rice genetics available decades ago. The Green Revolution “package” of stiff-strawed, lodging resistant lines used in conjunction with high input husbandry is more than breeding per se. The “package” and the following widespread application of the combined technology has given us plenty to debate concerning the merits of the Green Revolution… but the breeding alone is not guilty of any wrongdoing.

        • First of all, I have to apologize for mis-reading your original comment. I somehow thought you had claimed 5% annual gains. Even 1% per year is substantial, though, as you say.

          I’m not sure what you mean when you say you’re “less convinced the difference is ‘huge’ as far as varietal choice is concerned.” What do you mean by “as far as varietal choice”? Do you mean to say that there are improved varieties more suitable to small-scale organic farms and other varieties more suitable to large-scale conventional farms and that that both categories are seeing significant improvements through breeding? That would depend on your argument about organic versus synthetic sources of N.

          What you say about N may be technically true, but the point I intended to make had nothing to do with any differences between sources of N per se. The issue with organic sources of N is simply that they are limited in a way that synthetic sources aren’t (in the short term.) If we’re considering complete systems like Chris is with Wessex, and we’re talking about organic fertilizer, then there’s only so much N to go around. Of course, one can grow N-fixing cover crops, but everything comes with costs, and it certainly wouldn’t be the case that the average farm in Wessex could double rates of N application over any historic (pre-Haber-Bosch) level for double the cost (because of increasing marginal costs that would probably increase exponentially). With synthetic fertilizers over the short term, on the other hand, one could apply 10x the N for approximately 10x the cost and maybe even less by realizing some economies of scale. So that’s the difference I meant to point out.

          If the gains in breeding for varieties suitable to small-scale organic farms depends on the assumption that organic N comes at more or less a fixed cost per lb no matter how much is used, then there would be a lot more truth to your claim. But if the amount of N available to any given acre is very much limited (by increasing marginal costs), then I’ll confidently wager your claim is much less true. What kind of plant breeders would even consider a farming framework of N with increasing marginal costs? And if they’re not thinking in those terms, they’re not breeding for those kind of farms. And moreover, it would obviously be a lot more difficult to breed for increased yield without increased inputs. And that kind of breeding work doesn’t yield dramatic enough results for governments or seed companies to want to invest in, and small-scale organic farmers are difficult/reluctant customers anyways (and to the extent they’re not difficult customers, they’re gullible enough to buy seeds that aren’t really bred for or suitable to their circumstances anyways, such that in neither case does it pay to invest in breeding varieties of grass for them.)

          I know much less about soybeans than I do about wheat, for example, but I assume my personal observations with wheat and corn (maize) and other field crops isn’t so different from soybeans, and my experience with those crops is that the yield difference between old heirloom grains and modern varieties when grown with limited fertilizer (which is pretty comparable to how heirloom grains would commonly have been grown before synthetic fertilizers and hybrids) and otherwise in ways pretty comparable to old ways… 200 years of breeding hasn’t led to noticeable differences in yield. Certainly modern wheats are a lot shorter, but I can grow 5′ tall wheat with very little lodging so long as I don’t apply too much N (and adjust my seeding rate accordingly.) None of that is to say that “the breeding is guilty of any wrongdoing,” only that improvements are due chiefly to nonrenewable inputs (or at wildly nonrenewable/unsustainable rates) and the breeding improvements consist chiefly of breeding varieties more compatible with modern input rates rather than “pure” breeding improvements or improvements with much value to other systems. That’s also not to deny that “there are wheat and rice genetics today that work better under limited resource conditions than did the prevailing wheat and rice genetics available decades ago,” but that all depends on your definition of “limited resources,” and I haven’t seen any evidence (and especially not in my personal growing experience) that those gains aren’t dependent on at least limited increases in inputs.

          • We’re getting a little closer to a consensus here… I do want to continue the case for breeding, and with your permission will start a new comment thread below so that I have a bit wider window to type in…

  7. That Mezzullo thesis from your note 3 is quite fascinating and detailed. Good stuff.
    I’m still digesting it : ).

    Two areas specific to your Wessex scenario I think need some thought are the balance of plant size and logistics for feedstock transport, and the fact that the study ( as best I can tell so far) is really about input blends of animal manure and co-substrates such as your grass silage. I know that the health and vigor of the microbial workers can be rather touchy, and wonder if a straight grass feedstock would get the production assumed?

    Anyway, I am going back in to more thoroughly read the energy LCA and net energy sections when used for transport fuel, to make sure I follow them enough to compare to your assumptions.

    Props for the work you are putting in to this.

  8. Thanks for that interesting set of comments. To respond briefly, working down the page:

    Bruce: Yes, I was surprised too by how well the figures turn out. Regarding BECCS, that would obviously be great but I was thinking of this approach more just as a way to fuel agriculture in the event of an energy squeeze, regardless of the emissions implications (which presumably would be much less anyway), so I’m not sure that the proposals would depend on BECCS as such.

    Eric/Clem: I didn’t really think about draft animals – I have done occasionally in the past, but my fairly casual enquiries haven’t turned up many good figures on inputs/outputs for them. I’ll take your comment as a cue for further enquiry. I doubt an animal-powered agriculture could sustain a large population of 80% non-producers…but some might argue that would be no bad thing. There’d no doubt be some potential for grazing the silage fields, but if it’s organic you’d probably only get one good cut a year here so it would have to be judicious. Rabbits? Well, maybe. The Romans did introduce them here as a food animal, after all. I’m not sure later generations have thanked them for it much.

    Ron: I’m not sure I really understand your logic. There’s no particular reason to suppose that post-fossil fuel history has to look the same as pre-fossil fuel history. Doubtless there are good reasons to think that humanity may be in for a future Malthusian catastrophe, after which the population-land-technology equation would certainly be very different from the one I’m construing here. But I don’t see the logic of assuming such a catastrophe a priori and then invoking it to critique a programme for a more sustainable agriculture based on present circumstances. To my mind, such a catastrophe is worth trying to avoid however we can. ‘Farmland’ isn’t some fixed entity outwith human history that has to be treated in a particular way. People have always provided for their food and energy needs from the land around them – what’s so different with these proposals? I see much crazier uses of rural land around me today than using some of the less productive grassland to fuel agriculture.

    Eric/Clem: thanks for that very interesting and informative debate. I always like to see the case for and against plant breeding solutions! Maybe it’s worth mentioning that the land uses I’m employing in this analysis come from government definitions, and ‘permanent grass’ (a large proportion of Wessex farmland) implies that it’s been in situ for at least 5 years. So the opportunities for reseeding and rotating on that basis would be limited – doubtless yields would be higher with temporary grass, but the cost/benefit equation of it would be debatable, especially in an energy-constrained Wessex. I’m interested in the issues you’re debating around N responsiveness – my feeling is that Eric’s points about crop improvement in organic, constrained N situations have weight, but I’d be interested in further instalments… I’m not necessarily assuming that the farming for the 80% has to be of the small-scale, neo-peasant sort, but given the financial constraints Wessex will likely be under, perhaps we shouldn’t assume anything too heavily capitalised.

    Steve: thanks for that – sounds like you’ve read the Mezzullo thesis more thoroughly than I have. Remember to keep breathing while you digest it… Plant size and logistics is certainly an issue. I’m a bit reluctant to get into a full Wessex energy audit, not only because of the work involved but also because I fear over-specified utopias, but perhaps I’ll do a bit more work on this. On the feedstocks, I drew the figure from the reference in my second footnote, but I’m sure you’re right that there are optimisation issues with a blend of feedstocks. It ought to be possible to have blended feedstocks in Wessex, but as I understand it the basic energy potential of silages is superior.

    • What I meant to say is that, in case of no fossil fuels, I think people will revert to the known options that worked before; horsedrawn mechanics and manual labour. In that case there would be no need for biofuels and thus no land would be wasted in growing crops for that.
      Maybe mankind would be tempted to come up with hitech gadgets to be able to keep going, but considering the entire logistical system needed to keep those running in a profitable way…. I do think people are smarter than that.
      It isn’t just fuel; it’s lubricants, spare parts like tires, hoses, electronics etc. What would they be made of and at what agricultural cost?
      That’s what I meant with being right in the mind. It would be far to costly and labourintensive to do, let alone waste good farmland on it, where you could grow food that can be harvested and eaten.
      That you know worse ways of how farmland is used, supports my remark about the oit if its mind society. Just look around.

      • I believe that you are right Ron. Civilisation will revert to the use of draft animals by default. It worked very well in the past. As James Kunstler points out, “if we are lucky we might stop for a while at 1900 on our way back to the middle ages.

        • I think we’d struggle to find space for draft animals at current population levels. But I need to look into it some more – if anyone has some figures I’d be interested. Essentially we need to know what the net energy output of horses or oxen would be per hectare of land they take in order to compare it with the mechanical figures.

          I’ll write some more about the social side of this next year – whatever happens in the future, I’d be astonished if it turned out to be ‘going back to the middle ages’, except in a very loosely metaphorical sense.

          • Oxen would be preferable over horses, since they are lower maintenance when it comes to feeding, so areas unsuitable for crops can be used as pastures. Plus oxen yield more bi-products like meat and skin. Even other bodyparts.
            Reverting to the middle ages? Hell no. We have moved on, but in the scenario you drew a good number of things from the past work much better in agriculture, are much easier to get and much easier to operate. Cheaper too.

          • Well, I’ll try to look at some figures for oxen – I’m not convinced that it’ll turn out better than biogas. Certainly less engineering complexity involved though, other than the social engineering of returning virtually everyone to the land.

            Sounds like we’re agreed then that Kunstler’s view that we’re going ‘back to the middle ages’ is misplaced.

    • It’s hard for me to get enthusiastic about using grass instead of diesel during an “energy squeeze”, since it’s likely that the folks growing and transporting food will be the very last users of diesel ever. Once the diesel is gone, assuming that tractor parts will be available is getting into “assume a can opener” territory.

      The main reason for my reply is that I object to your comment thatthere’s no particular reason to suppose that post-fossil fuel history has to look the same as pre-fossil fuel history.

      The very good reason that the histories will look similar is that pre-fossil-fuel people were not any less intelligent than we are now. If there was a much more easy or efficient way to grow food in the pre-fossil-fuel era, someone, somewhere in the world figured it out. Any dissimilarities will be to the detriment of the future history.

      Future farmers will be lucky to do as well as people did farming before fossil fuels. A great deal of local agricultural knowledge has been lost as the tsunami of cheap energy and other agricultural inputs has washed away subsistence agriculture, particularly in the northern hemisphere. We are going to have to discover many basic operating procedures all over again. I hope I can look to Small Farm Future for help with that.

    • Ron, Joe:

      First, thanks for the clarification, Ron – I think I understand better where you’re coming from.

      I’m not convinced that the issue of grease, tyres, spare parts etc. is all that salient. The energetic cost of these things is quite low relative to energy-in-use, and would be lower in an energy-constrained society than it presently is. So what I take from what you’re saying is that the kind of society I’m construing wouldn’t be feasible unless it retains complex levels of social organisation that would permit an engineering industry and infrastructure. That, I think, is true. And it may be true that such a society won’t be possible in the future. But as a matter of historical philosophy, I don’t think that you can assume it will be true a priori as the basis for critiquing the feasibility of a society in which it isn’t true. To me, that’s another kind of can opener.

      In this post http://smallfarmfuture.org.uk/?p=1102 I looked at the feasibility of feeding Wessex’s 6.3 million with essentially no mechanical aids. I think it would be possible in theory, but probably next to impossible in practice – and I think there’d be little room for horses in it. But your comments and Eric’s prompt me to try to look at this in more detail in a future post. Prior to the industrial revolution, the UK population was about a tenth its present number, and still importing food. I don’t see the point of assuming away 90% of the population. It could doubtless be argued that a large proportion of the current population is destined to ‘go away’ in the future whether we like it or not, but I just don’t see the point of sitting here now construing some kind of future agrarian society on that basis.

      On the issue of pre and post fossil fuel histories, it’s got nothing to do with intelligence. At one level, I’d say that post fossil fuel history won’t be the same as pre fossil fuel history in the same way that the post Roman history of, say, France, wasn’t the same as its pre Roman history. At another level, technologies develop even in the absence of exogenous inputs – which is perhaps what Clem has been saying in relation to plant breeding. The technological reach of England in 1700 was much greater than in 700, and that in turn than 300BC, and not because of any differences in people’s intelligence or substantially increased exogenous energy availability. I do think you’re tending to assume a priori that it will be impossible to carry any present technological knowledge or capacity into a post fossil fuel future – in which case, well yes, things then may end up looking quite similar in some ways to how they used to. But it’s an assumption, and in any case I’m more interested in looking at where we could try to go now in order to avoid that eventuality than at what that eventuality might look like.

      However, I do agree with you that humanity is currently putting a lot of its eggs in one rather threadbare basket, that it has wantonly lost a lot of older knowledges that we could do with retaining, and that the way it’s gone about these things doesn’t augur especially well for its ability to transition to a more sustainable future without pain.

      • At another level, technologies develop even in the absence of exogenous inputs – which is perhaps what Clem has been saying in relation to plant breeding. The technological reach of England in 1700 was much greater than in 700, and that in turn than 300BC, and not because of any differences in people’s intelligence or substantially increased exogenous energy availability.

        1700 involves far too much exogenous energy from coal to be included in your example. And was the technology of England in 700 really that much more sophisticated than in 300 BC even with the help of the Roman occupation in the intervening years?

        This “technological reach” issue is one that has long interested me, but I have found little research on it. It seems obvious that there must be some limit on technology, depending on how much surplus is available to support those who live as scientists, technologists, and other specialists, but I haven’t seen any good treatments of this subject except in very general terms.

        Is it really possible that given enough time and desire, even an agrarian peasant society could accumulate enough “technological reach” that they could gradually built a rocket ship and fly to the moon? The whole prospect seems ludicrous, but I can’t prove that it can’t happen, or find a reference that tackles this issue in detail.

        I think that there is indeed a limit to technology that only more and more exogenous energy can surmount, but I can only point to the fact that the ancients never developed this or that machine, not that it was theoretically impossible for them to do so.

        It does seem obvious that even if, say, the literate folks in a small village were given access to all the current knowledge of the universe via a library of books written in their language, only a tiny bit of that knowledge could ever be internalized or used in any way.

        A book that is never read has the same practical use as a book never written. As village readers die, everything they read and understood vanishes from their brain. A young villager has to start the learning process all over again, so only a tiny fraction of human knowledge could ever be understood by a small group, even with all knowledge given to it to use as desired.

        There must be only so much that can be done with a limited population, even with unlimited time to do it. My intuition tells me that history should be a pretty good guide to what is possible, but intuition can be wrong.

        Perhaps someone can point me to a reference that explains the relationship between energy surplus and technological specialization more fully and quantitatively. It is a subject that is very appropriate to our consideration of what possibilities remain after energy availability declines in the future.

        • I’m not persuaded that coal use in 1700 was hugely significant overall – more important was growing Dutch & British sea power and capitalist relations in the countryside, driven largely by renewable resources. People have a tendency to see increasing energy capture as the exogenous driver of social change – often it’s the other way around. Still, the capitalist and colonialist world inaugurated by early modern Europe likely has its own endpoint, which perhaps we’re reaching now. So I agree it’s interesting to think about the capacities of less energy-hungry societies. Those capacities are affected by how such societies connect. Near fully autarkic small-scale agrarian communities have scarcely been the global norm since…well, I don’t know…the Neolithic? The Bronze Age? I don’t see good historical reasons for thinking they’ll be the norm in the future. But it’s certainly interesting to think about the connections between energy, technology and social forms – something I hope to come onto in the second part of this analysis.

  9. NOTE: I noticed an error in my previous calculations, where I unaccountably overstated the amount of permanent pasture in Wessex. I’ve taken the liberty of correcting the figures in the post above from those included in the original text. The correction reduces the ratio of energy produced:energy needed from 1.7 to 1.5 – so I’d argue that it doesn’t alter the overall conclusions much.

    It’s also perhaps worth noting that the figure I’ve used for the total amount of grass available for AD does include the grass/clover mix in the 50% arable ley – about 20% of the total grass. So in relation to Clem’s points about crop improvement and reseeding, there’s perhaps some potential there that I neglected to mention in my comment above.

  10. While you’re correcting the figures, could you please also change GJ to PJ? The progression in factors 1000 is kilo Mega Giga Tera Peta, so a billion MJ is a PJ. You make the same mistake on both ends, so your conclusion still stands.

    • If you look at the text above, I think you’ll find it clearly says PJ…

      Aaargh, I hate making mistakes like that. Thanks Erik.

  11. Hi Chris, I’ve been lurking here for a while and not joining the debate, but I thought I should in this instance as I recently came across a similar debate amongst the members of the PFLA (Pasture Fed Livestock Association).

    It seems that Ecotricity are seriously considering producing “green energy” from grassland. Please see the following link:


    The flip side is that we all have to become vegan according to Biofuel Watch:


    It would seem that the concept is far from a straight forward green solution. I understand you are only looking to power the agricultural economy in your thought experiment and not the entire community but some of the same criticism applies.

    Biodigesters are becoming reknown in farming circles around me for being incredibly complex beasts and prone to digestive upset. Alarms constantly going off in the middle of the night and so on. Not a low tech concept at all.

    For what it is worth, my own view is that in the Wessex of 2039 central government will still exist and will be will be supplying farms with what is left of our dwindling diesel or we will be using biofuels from oilseed rape. (Any old tractor can be converted to run on biodiesel but not so sure about biomethane. The technology isn’t proven or widespread.) There will also be electric tractors charged using PV arrays. Food will become priority number one after the climacteric. If it gets so bad that none of this is possible, then that is a future I hope not to see.

    Full time farmer and ex-Philosophy graduate.

    • Thanks for commenting Dan, and for linking those documents.

      Ecotricity’s projections of 104 GJ/ha compare to mine at 68 MJ/ha but mine are organic so should perhaps be revised downwards in view of the assumed silage yield @ <50%.

      As you point out the big difference is that domestic consumption dwarfs agricultural consumption by a factor of 36, so I'd argue that Biofuel Watch's objections on land take don't apply to my analysis. Their document does rather seem to take current meat consumption levels as a given - if we're funding our economy out of biotic resources then something's got to give, and meat is an obvious target. If we choose to avoid biofuels then that surely takes us back to PV and/or nuclear as per the discussion under my previous post. Of course, the level of domestic consumption you've highlighted does raise the question of how my Wessexers will cook and stay warm. Which could be a problem - I agree that current levels of natural gas usage just aren't feasible with biogas. I'll try to write about that in a forthcoming post.

      Regarding the other downsides of biogas, I'm unsure. CO2 off-gassing may not be ideal, but surely it would be within the annual carbon cycle so not a forcing emission as such? Methane leaks would be more problematic - though I suspect the total emissions would be much lower for an agricultural sector based on biogas than at current levels of natural gas production. On the complexity of the technology, well the basics of AD are pretty simple, as attested by millions of impecunious Chinese peasants under Mao, but I don't doubt there would be a complexity involved to a modern plant aiming to optimise output. Perhaps I need to build in some margins for that, but there's methane in the bank from the above analysis.

      If you twist my arm, I could re-run the analysis with rape biodiesel - but we'd be losing a lot of our cropland, so the margins would get tighter.

      • “…does raise the question of how my Wessexers will cook and stay warm”…

        Don’t know about the cooking, but they could stay warm all bundled up in rabbit fur coats. Just sayin’ 🙂
        [full disclosure – I have zero financial interest in rabbits, rabbit breeding, hutches, or any other rabbit infrastructure. But that could change quickly if all the doom and gloom forecasts start to gain traction in the wider economy…]

        I have to second Chris’ hat tip to Dan. The Ecotricity piece is interesting, if overly optimistic. The BioFuelWatch response is also interesting from the perspective that a lot of the grief they take from their opponents is recycled and aimed at Ecotricity. If good for goose, then good for the gander I suppose. But one of BFW’s citations (#5) from the Soil Association really caught my eye: https://www.soilassociation.org/media/4671/runaway-maize-june-2015.pdf

        There are some problems in the degree to which the Soil Association overplays their complaint(s) – but I think their argument is fairly strong and worth consideration by a larger audience. And I find myself quite conflicted even admitting this much – as I would more often be siding with maize production (if for no other reason than it makes a suitable rotational crop for soybean).

  12. Eric, Chris:
    The value of any sort of breeding activity – plant or animal – will depend upon the target audience if you will (the technology available, the climate, soils, common pests, and so forth). To this extent then Eric makes very salient and worthwhile points about limited availability of some resources having an influence on yield improvements. Responses of grasses to N inputs are large and it is easily demonstrable that yield increase to higher levels of N is greater in the short term than increases in yield due to breeding or any other husbanding effort (save perhaps irrigation if incident rainfall is limiting).

    Having acknowledged that some agronomic practices (irrigation if needed, N fertilization if needed or practical) will provide faster responses than breeding an already adapted species, I will argue that further breeding efforts can be worthwhile – and even benefit a peasant grower. Paying for the effort is a real issue, and Eric has alluded to this challenge, so I hope to make some progress along that line as well.

    First, I’m not convinced that the marginal costs of increased N come at an exponentially increasing rate. I do agree that increased N amounts will increase costs, just that exponentially seems wrong to me. But for the moment that seems to me a digression. Lets work within the limitations of Chris’ present scenario.

    Hybrid corn is a phenomenal technological change in plant productivity. There is plenty of evidence from common garden type research to show that hybrids outperform open pollinated corns in all sorts of production systems. It is accurate to point to larger performance gains in higher yield environments – and that one aspect of higher yield environments is higher N availability. But higher yields are still obtained in more restrictive situations as well. The next question then becomes – is the juice worth the squeeze? In a resource poor environment are there breeding improvements worth the time and effort it takes to realize them? This is a more difficult problem to approach. And Eric rightfully asks “What kind of plant breeders would even consider a farming framework of N with increasing marginal costs?” Which, if I may, I’d like to reword slightly to ask what kind of plant breeders would even consider organic or other marginal input agriculture. The two letter answer? Me. But that’s being too trite. There are other breeders involved in this type of work, but it is still worth mentioning that Eric does have a point. The larger share of effort is going to follow the money.

    At this point I might split my response in a couple directions. First would be breeding for adaptation, the second would be breeding for continued small incremental gains and defense against pests which will continue to evolve along side our continued peasant husbanding efforts.

    By breeding for adaptation I am thinking of breeding various species not currently in use in a given area or cultural system. Soy in the UK could be exactly what I mean here. Closer to where I actually work we might be looking at flax, barley, buckwheat, and a host of cover crop species… each of these can be grown in Ohio, but current varieties and market conditions limit their use. Adaptation breeding then seeks phenotypic changes that allow successful production under current market conditions. Adaptation (think domestication to a certain degree) can yield huge impacts when successful. On the opposite hand it can be quite speculative and many will be hesitant to risk capital in pursuit of it.

    Breeding for incremental improvements is more of a grind. Successes are small and while not particularly flashy they can be more predictable and thus can merit the attraction of capital to warrant their continued application. A general drift of Eric’s argument is that breeding for increased yields works in high input husbandry and not in low input efforts. It is easier to demonstrate higher yield increase in higher yield environments, but it is not true that with lower inputs plant breeding cannot be successful.

    Another avenue worth consideration – grower participation. If a new variety of some crop has a particular feature giving it some advantage that is context sensitive (and this is very often the case) then the variety will likely not have merit for those growers who cannot or will not husband the variety in the manner necessary to exploit its potential. On the first issue – growers facing some prohibition to use – the breeder has essentially failed her audience. On the latter challenge – growers unwillingness to try something new… well you see where that goes.

    Paying for breeding efforts is a very real issue, and I have a piece in the pipeline at GP to touch on this. It doesn’t specifically address organic ag (in fact it is more focused on industrial ag), but some of the issues raised are relevant on both sides.

    • Thanks for that Clem. I like you beating the drum for plant breeding, and I’ve learned a lot from it. To play curmudgeon, though, the kind of figures linked by Dan above for total gas use suggest to me that we do have a total energy supply bullet to bite which is unlikely to be solved by plant breeding…

      • Hey, curmudgeon is my shtick… but you’re the host, so I’ll allow it 🙂

        I think there are opportunities for plants to step up and help us plug some holes in future energy supplies. We have only scratched the surface of technologies where plants (or plant like life forms) might help directly or show us how we might fashion something like a better solar array.

        Algal biomass for biofuel, microbial nitrogen fixation so that we can have N fertilizer(s) at lower energy expenditure (avoid Haber Bosch) – these are just a couple things. How do they possibly play on a Wessex homestead? Perhaps they don’t, but if I might stretch the imagination a bit you could set up a pond or shallow water raceway to grow the high oil alga. The oil extraction step might not be a scythe level technology, but at the same time I don’t see it being cell phone level tech either.

        Breeding doesn’t drive either of these potential technologies… but like any other effort where plants are used, once we get the fire started we can make it burn hotter, or longer, with less smoke, or tweak it in some fashion we find desirable just by knowing something of the basic biology of the system and how the organism reproduces. Then we can start breeding.

        If I play with the algal biodiesel idea for a moment you can imagine that once you separate the lipid from the rest of the algal mass you can find other things to use the non-lipid components for. Animal feed jumps out at me. And animals like plants are readily modified through breeding efforts as well. Got rabbits that don’t particularly like the smell of your de-lipidized algal biomass? Breed for some that do. How hard could it be? [well… it could be pretty tough, but hey – you have time on your hands, none of your electronic junk works anymore, what else are you gonna do?? 🙂 ].

        • Interesting points – if I could breed a strain of rabbits that only liked eating algae and that could displace their wild cousins, I might ditch the electronic junk right now.

        • Algal biodiesel has several problems that have to date affected widespread deployment.
          – Algae bred to increase lipid production become less able to compete with wild algae when grown in open water.
          – This problem can be removed by growing the algae in sealed tubes but this is very expensive and there’s issues with ensuring the bulk algae get illumination as against a surface layer up against the surface of the bioreactor.
          – The lipids made by the algae have a tendency to wander a bit from the desired straight carbon chain. This affects the biodiesel conversion reaction and the quality of the fuel produced.

          Algal biodiesel at the moment is one of those technologies that clearly works from a technical perspective but faces issues in productionisation. Having said all that, if fossil diesel became unavailable or very expensive tomorrow I’d have a hectare or two of algal ponds in quick smart. Next to a few hectares of oil seed crops. Liquid fuels are so useful that even if they were expensive there would be a market.

          On a related technology deployment note, making biodiesel from biogenic lipids is a simple reaction. Many farmers already do this in Australia. Much easier to manage than an AD. I’ve been looking for commercially proven AD kit at relatively small scale that can handle a range of feedstocks. There’s not a lot on offer IMO.

          • I’ve watched the algal biodiesel research in the past and had come to the conclusion that breeding could help, but there are other issues to be worked out as well. Knew of the difficulty of high lipid algae having difficulty in open ponds and now wonder if there might be ways to manage the ponds so the high lipid genetics are favored (or competitor genetics suppressed). Changing pH; co-cultivating some other organism that preferentially feeds on competitors, that sort of thing. It may be possible to raise high lipid algae in a batch system – harvesting a whole pond and restarting a new batch (essentially racing the competitors until the ‘soup’ is no longer economic, harvest and start over).

            I’m guessing none of these approaches would be perfect or immediately realizable… but as David points out, in the face of no (or diminished) fossil fuel the difficulties these technologies present start to disappear.

  13. I suffer from the vice that has no name (Railway Enthusiast) which often leads to me pocking around other bits of vintage kit such as you find in rural shows – traction engines, old tractors, stationary engines etc

    Now from looking at them all its clear that our forebears managed to live their lives with a lot less power than we use now – and that’s despite the much greater efficiency of new technology for example LED rather than tungsten filament light bulbs, or modern cars compared with the Morris Minor. Clearly however we have squandered the efficiency gains with much larger/faster/complex/whatever kit.

    Clearly if we were to go back to the days of the (modern equivalent of) little grey ‘Fergie’ rather than todays massive four wheeled monsters our overall energy use would be much less

    • Yes, good point. I’d be interested if someone with more railway (or tractor) enthusiast skills than I have could run a little energy analysis of what a Ferguson TE20 world might look like.

  14. Earlier in the week Eric and I were discussing the value of plant breeding for small farms and in particular organic systems. Now I’ve come across the following research from Germany which is very relevant:

    Development of locally-adapted faba bean cultivars for organic conditions in Germany through a participatory breeding approach. Participatory breeding of faba bean for organic conditions

    This is open access, you can get the full article from the publisher at:

    Most of the discussion here is accessible without an inside knowledge of plant breeding. I will offer though, if anyone has a question about a technical term or other issue about the paper I’d be happy to answer it. The scientists in this effort have done a remarkable job. And if Eric hadn’t pushed his side as much above I might have missed this article all together. Thanks Eric!

  15. Well, the most powerful Ferguson was 36HP while the current Case range starts at 65HP & goes up to 620hp.

    Presumably our ’50’s farmer would not have even considered some of the things his grandson now does, or if he did it would have taken much longer.

Leave a Reply

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