LA Chef’s editorial: The limitations of quantitative analysis

(Originally published December 27, 2015 on

Whatever one’s food politics or religion any study that attempts to quantify food production needs to be viewed with a degree of circumspection. Meta-analysis of prior studies and the interpolation (and re-interpolation) of data into mathematical models can’t account for all the variables involved in farming and ranching. Thus any unit of measure generated by researchers at institutions has limitations. Farming and ranching are often very land specific. The weather, water source, and soil quality matter. The methods of cultivation and livestock management also matter. So more often than not, how food is grown and raised, that is the methods of food production, are as important as what is grown or raised. Furthermore not all land is interchangeable. A lot of pasture or grasslands isn’t arable (suitable for crops). Thus the best or most appropriate use of any piece of given land for food production also needs to be considered. This is especially true in California, with its many micro-climates, where so much of the country’s produce and world’s specialty crops are grown.

In a new report, “Energy use, blue water footprint, and greenhouse gas emissions for current food consumption patterns and dietary recommendations in the US” by Michelle S. Tom, Paul S. Fischbeck, Chris T. Hendrickson, the researchers from Carnegie Mellon University took averages for blue water use, GHG emissions and energy inputs (on a per calorie basis) from different existing studies, plugged this data into their equations and generated the following outputs for three different dietary patterns:

  • For Scenario 1) a diet that reduced caloric intake, but with a mix of food types representing the average American diet, decreased energy use, blue water footprint, and GHG emissions by around 9 %.
  • For Scenario 2). A diet that did not reduce calories, but included the proportions of food types recommended by the USDA increased energy use by 43 %, blue water footprint by 16 %, and GHG emissions by 11 %.
  • For Scenario 3). A diet that reduced calories and included the proportions of food types recommended by the USDA increases energy use by 38 %, blue water footprint by 10 %, and GHG emissions by 6 %.

None of these three scenarios was specifically vegan or vegetarian, but the USDA, in general, recommends greater intakes of fruits and vegetables as well as dairy and seafood. So per the researchers’ analysis, the greatest increase in energy and blue water requirements came from increasing the intake of fruit (114 to 210 calories) and vegetables (127 to 234 calories) for the second and third scenarios. Higher energy and GHG emissions analysis came more from increasing consumption of fish (14 to 44 or 49 calories) for the second and third scenarios. The specific calorie numbers for USDA requirements were noted in the supplementary information provided with the report. The report thus attached adverse environmental consequence to increasing intake of foods more often construed as environmentally friendly as other alternatives.

The ensuing debate in the media about whether or not omnivore or vegetarian diets are better or worse for the environment was even more detached and abstract than the report itself. Needless to say debates about units of measure based upon averages and what’s good or bad in definitive terms fail to recognize all the variables involved in farming and ranching as well as what constitutes appropriate land use. The fact that fruits and vegetables can be blue water intensive really shouldn’t have been that big of a revelation. Though then again people regularly cite water footprint numbers without the slightest clue as to how those numbers are actually derived. What undermines or builds soil health is also missing from these debates. The Carnegie Mellon report itself in its section 4.2, Limitations and future work, acknowledges the shortcomings of such an ivory tower approach when it states the following:

“A major limitation of our study, thus, stems from this meta-analysis approach. For instance, differences in geography, climate, and culture may warrant different food production methods and resource requirements.”

Food production methods on any piece of land may determine whether farming or ranching practices are either beneficial or detrimental to the environment. Prevailing beliefs bantered about in the media are that meat production is bad especially for beef, but well-managed grass finished cattle in grasslands environments allow for the sequestration of carbon in well-aerated soils that contain methane oxidizing bacteria so cattle’s emissions in these intact systems are more than offset by carbon sequestration and methane oxidation. Plus the cattle are eating drought-resistant grasses that require little to no blue water. Soy grown with cover crops in a no-till system using a roller crimper will have a completely different environmental impact than monocrops of soy grown with synthetic fertilizers or in a no-till systems where cover crops are burnt down by glyphosate. Geography and climate may also determine whether any piece of land is suitable or not for certain kinds of crops or livestock. If one piece of land draws blue water from an aquifer four hundred feet deep and another piece of land draws water from an aquifer twelve hundred feet deep, the environmental impact may be drastically different for the same vegetable, fruit or nut (especially almonds and pistachios) crop being grown since the deeper well may pull up toxins that adversely impact wildlife and salts that adversely impact soil fertility. A large amount of energy is required to drill the wells and pump the water from these deep aquifers that took thousands of years to fill. Surface water diverted from rivers to irrigate fruits, nuts and vegetable crops also adversely impact fish like salmon in those river systems. None of these environmental impacts are measured via meta-analysis.

So one really can’t make any sense of what’s good or bad or somewhere in-between for any piece of land just crunching outputs from data entered into equations. Considering how critical soil health is to any environmental assessment, any such a remote report and ensuing debate just further reinforces the obvious disconnect between number crunchers and reporters from farmers and ranchers. Again regarding soils, repeated tillage for and harvesting of crops undermine the soil’s capacity to function as a methane sink, releases carbon into the atmosphere and causes erosion. So yes, annual vegetables can be detrimental especially considering the rates at which top soil is being lost. Turned over soil kills beneficial soil microbes. No-till systems with chemical inputs like glyphosate also adversely impact beneficial soil microbes. Synthetic fertilizers also lead to compaction and less healthy soil with greater water runoff and pollution due to nitrates as well as emissions from nitrous oxide. Again all these variables and varying methods aren’t accounted for with averages plugged into equations. So whether a specific report supports or contradicts one’s food politics or religion really doesn’t matter much since any such report is so limited with its analysis as to be a somewhat meaningless pedantic exercise.

Now in this specific report from Carnegie Mellon’s supplementary information, minimums and maximums are provided for different foods for different energy inputs and blue water use. Plugging in the minimums and maximums rather than averages may have provided a more meaningful basis for dialogue.Knowing which methods of food production produce the least or most emissions- plus are the least or most resource intensive- is potentially more informative since such an analysis may provide a better perspective as to what methods of food production reduce emissions, blue water usage and energy inputs. So again the how and where food is raised and grown matters as much or more than what is raised and grown. Such a more detailed analysis though requires getting out of ivory towers and back onto the land.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s