From Wikipedia, the free encyclopedia
Jump to: navigation, search
Article milestones
Date Process Result
July 18, 2007 Good article nominee Listed
October 14, 2007 Good article reassessment Delisted
January 20, 2013 Good article nominee Not listed
This article has been mentioned by a media organization:
Frequently asked questions (FAQ)
Information.svg To view an answer, click the [show] link to the right of the question.


Page views for the last 12 months


On land use[edit]

Regarding my edit on land use, which has already generated a couple of reverts and re-reverts, can I kindly ask why it's been removed from the page? There are three scientific studies that clearly support the sentence that I added, two of which were already cited in the page. It seems to me that the three main findings of those studies were correctly represented: (i) the vegan diet has the lowest absolute landuse, (ii) the vegan diet requires high quality croplands, (iii) other diets make a better use of the land overall, once one factors in that not all land available is high quality cropland. Why was point (iii) removed from the page? Let me point out that the title of one of the papers that were already cited in the page is "Diet for small planet may be most efficient if it includes dairy and a little meat". --Japs 88 (talk) 12:33, 10 August 2016 (UTC)

Point ii is actually an assumption of the paper, not a result. The issue is explained in the last paragraph of section 2.3. It's also put more succinctly later in the paper: "The existing ratio of cultivated to perennial forage crops was used to set an upper bound on the area of cropland considered suitable for cultivated cropping." It notes in section 3.4 that the carrying capacity estimates are highly sensitive to this assumption, with all but the vegan model diet plateauing, meaning in essence that it's entirely responsible for the widely-reported result that other diets make better use of land - which is not exactly what the paper says, either. Better use of land is subjective - what we are really talking about is more use of land, and this is neither an environmental nor a sustainability issue (hence doesn't belong in this section), as the paper found that the vegan model diet could feed well over twice the US population, while other options made modest improvements by exploiting more low-quality land which the vegan model diet made no use of, by the study's assumptions. --Sammy1339 (talk) 13:20, 10 August 2016 (UTC)
Also, I think there was only one revert? --Sammy1339 (talk) 13:30, 10 August 2016 (UTC)
Hi, thanks for the quick and precise reply.
First, I confused another revert with similar bites changed: you're right, only one revert.
Second, I saw you removed the reference about land use in the state of New York. Please notice that the only reference that claims that the vegan diet yields the lowest land use is limited to the state of New York, too (indeed it titles "Testing a complete-diet model for estimating the land resource requirements of food consumption and agricultural carrying capacity: The New York State example").
That said, in the study I cited, the assumption the authors make is not that the vegan diet requires high quality land (statement ii above). That statement is just a matter of fact: you can't cultivate crops on rocky/hilly terrain, although this might still be suitable for animals. The assumption they do make is on the fraction of high quality land over total productive land. Indeed, in sec. 2.3, they divide the available productive land into cropland and grazing land. In sec. 2.4.3, to model the available cropland, they add all the land used to feed humans and all the land currently cultivated to forage crops, thus obtaining an upper bound (i.e., an optimistic estimate) of the amount of high quality cropland fit to feed humans. In the USA, for which the authors crunched the numbers, this level, call it Q, is found to be Q=71%. This is the actual assumption.
As for the sensitivity of the results on this assumption, figure 5 comes to help. You can see that the vegan diet yields a better carrying capacity (call it C = people fed / land area) only for Q>90% ca. For Q=80% ca., the vegan diet yields a C equal to a diet with 20% calories intake from animal produce. For the conservative assumption that all lands used for forage could be converted to cropland (Q=71%), even a diet with 40% calories from meat yields a higher capacity than the vegan diet. Contrary to what you say, figure 5 shows that the claims are very robust with respect to their assumption: even if their estimate is off by 20%, i.e., if the real Q is actually 85% instead of 71%, the capacity of the vegan diet would still be more or less the same of a diet that sources 20% of its calories from meat. It may be that in the flat Netherlands you can reach the Q>90% required to make the vegan diet yield the highest C. At the same time, in places like Italy, Switzerland or France, where the orography makes large amounts of productive lands unsuitable for crops, the actual value of Q could be significantly lower than 71%, and make larger meat intakes even more favourable in terms of land capacity. Hope this clarifies the point. I'll wait for further comments before restoring my addition.--Japs 88 (talk) 16:24, 10 August 2016 (UTC)
To your second point, see table 1 from this study, showing that the vegan model diet used the least land per person.
As for sensitivity, it says that "Carrying capacity was shown to be highly sensitive to the starting assumptions about the proportion of land available for cultivated cropping." The estimate you call Q was deliberately conservative, as the authors assumed that current perennial cropland cannot be converted. ("Estimating the optimum combination of annual to perennial crops on U.S. cropland to control erosion and maintain adequate soil health lies beyond the scope of this paper. A baseline estimate of the limit on cultivated crops was made based on the current proportion of cropland under cultivation.") This assumption seems to be responsible for most or all of the effect, although I don't have any idea whether Q=92% is realistic. (That's the value at which the vegan diet outperforms the lactovegetarian diet.)
A reasonable summary would be that the vegan diet uses the least land of any option, and can feed many more people than necessary, but that there is some evidence that, in the United States, a lacto-vegetarian diet might support marginally more people, by maximally exploiting low quality land. Still, I wonder if that belongs in a section about environmentalism due to WP:WEIGHT. --Sammy1339 (talk) 16:46, 10 August 2016 (UTC)
I'm not saying that it is not true that the vegan diet uses less land overall, as you correctly point out from the table. The table also says that diet OMNI40 uses the same amount of cultivated cropland ("high quality" from above). Sensitivity means that the curves of fig.5 have a steep slope. However, this does not change the conclusion that diets LAC, OVO, OMNI20 and OMNI40 yield a higher capacity where it matters, i.e., for Q=0.71. Based on the paper saying that "the existing ratio of cultivated to perennial forage crops was used to set an upper bound on the area of cropland considered suitable for cultivated cropping", I understand that Q=0.71 is not conservative, as you state, but optimistic. They considered all the land currently used for perennial forage crops as "suitable for cultivated cropping". In any case, as I said above, let us assume that there is a 20% relative error on the value of Q: the result that other diets, including omnivore diets, yield a higher capacity does not change. This is what in science is called robustness. Finally, the increased capacity over the vegan diet is not marginal (almost 10% for LAC, almost 5% for OMNI20). I think writing about the reduced land use without also talking about the type of land needed and about capacity would be like cherry-picking the aspects of the study that "look good" on this page, trying to hide the rest under the carpet.Japs 88 (talk) 20:37, 15 August 2016 (UTC)
No, I think you have it backwards: they considered all the land currently used for perennial forage crops as unsuitable for cultivated cropping, and in fact, they cite earlier work which estimated the proportion of cropland actually used for perennial forages as low as 20%, but their analysis uses a more conservative figure of about 30%. In any event, it has to be understood that this is largely irrelevant to environmentalism, as the goal of environmentalists, as explicitly noted in section 1.2 of the paper, is not to use the most possible land or to produce the largest possible amounts of excess food. Note that the carrying capacity is well over twice the population; a vegan food system would not hit the limit of cultivated cropland in the foreseeable future. --Sammy1339 (talk) 21:16, 15 August 2016 (UTC)
I'm not going to pretend to have followed all the details of that, but I'm getting a strong sense that this information does not meet the standard of WP:PRIMARY: straightforward, descriptive statements of facts that can be verified by any educated person with access to the primary source but without further, specialized knowledge. Especially in light of concerns about "cherry-picking", I believe we should wait to integrate this material until an expert secondary source provides an expert summary properly accounting for the plausibility of different "Q"s, the significance of carrying capacity as a measure of environmental sustainability, the robustness of the studies' methodology, and the paper's appropriate WP:WEIGHT.
The recent study all but says as much: section 4.3 notes that essential details on variability are absent or unquantified, writes that "the results are perhaps best treated as a foundation for further hypothesis testing," and is confident only in the extremely minimal hypothesis that "dietary choices are important." That doesn't seem adequate for an assertion of fact in an encyclopedia.
This is especially true in light of the WP:WEIGHT of this 2015 systematic review, whose section 3.3 draws a different conclusion. Recall that primary sources should not be used to contradict the conclusions of secondary sources, unless the primary source itself directly makes such a claim, and this paper doesn't mention Hallstrom et al. FourViolas (talk) 21:30, 15 August 2016 (UTC)