1

Nutrient Density

This is a topic about which I have been speaking, and in these pages writing, first 10 years ago.

As a child in elementary school I remember telling my science teacher that I was an Organic Farmer. The distain with which she responded in front of my peers by saying “organic means contains carbon, so all farmers are organic farmers” sticks with me to this day. It was not actually that long ago that a group of back to the land homesteaders took a word that meant something in science and redefined it to create a cultural touch point which now has a significant meaning globally.

In much the same way, nutrient density has for quite some time meant something in the lexicon of food scientists. Specifically the nutrient density index of a crop is determined by its average level of nutrients per unit calorie. For example kale has on average a relatively high level of nutrients but a low level of calories. By this metric then kale has a high nutrient density score. Rice for instance has many more calories in it per unit nutrient and so would have a low nutrient density score. Soda, would be a “food” with very many calories and almost no nutrients and so has a very low score according to this definition of nutrient density.

This is an entirely different meaning than has begun to be understood by the broader food movement, focused upon food quality, which is well along the process of appropriating the term. What we mean when we talk about Nutrient Density is how nutritious one bunch of kale is in relation to another. Or one bag of rice to another. In the food scientists’ definition there is an implicit assumption that all kale is relatively nutritionally uniform, as is rice. This assumption is foundationally flawed and much of my work over the past 10 years has been to shed light on this topic.

This graphic shows how crops identified with a particular farming method compare, across all nutrients measured, to the same crops drawn from the samples the RFC received without identification as to the way the crop was raised. Thus the Biodynamic crops in the first box have 15% fewer antioxidants and 6% fewer polyphenols than the unidentified crops, as well as 18% less chlorine and 6% less iron. On the other 8 nutrients they exceed the levels in the unidentified ones.
The intensity of the color in each bar represents the confidence RFC has is the reported value — 25%, 50% or 75% certainty.

As the theme for this issue is Who Owns Science?, and what are the mistakes in thinking that are caused in the broader population by incomplete definitions of reality propagated by that community, I hope to share an inspiring story here about how we can use the process of science to more deeply understand and engage with the world around us.

As a farmer, my experience has been that some crops have greater sheen, vigor, vibrancy, pest and disease resistance, flavor and shelf life than others — carrots to carrots or squash to squash. Also, it seems that crops with these attributes cause the soil they grow in to increase in its health. My understanding is that there are some profound nutrient variations in these crops that are directly connected to these other physiological attributes, and that that is really what I should be striving to accomplish rather than a label like Organic or Regenerative or local or Non-GMO etc. None of those correlate to the inherent nutritional quality of the crop and are effectively process standards, not quality standards. They are black and white binaries. A crop is either organic or not, local or not. There is no subtlety or nuance in them. That would be like saying someone is healthy or not. Or strong or not. We all know that our health or strength exists on a continuum and is not actually able to be understood with such a simplistic summary.

Perhaps this binary way of thinking is part of the colonized mindset that is foundational to many of the existential crises that we appear to be facing at the moment, but again, this is a digression that can wait to be elaborated upon.

For a decade, the Bionutrient Food Association, which I helped to found and serve as Executive Director of has been working to expand upon our deeper understanding of this new definition of Nutrient Density, and for the past 4 years has been actively engaged in using the scientific process to do so.

Before commencing we tried to work out a process to use the scientific method to describe a profoundly multifaceted concept — nutrition. Also, we thought, this could become a new standard to potentially displace the foundationally binary nature of most of our current assessments of food.

To that end we devised a strategy to move this agenda forward:

  • Identify the nutritional variation that exists in crops so that an empirical definition of the Nutritional Density of any individual crop could be accomplished. Specifically then we would be able to say that this carrot would be at the 20th quality percentile for instance while we could say that that carrot is at the 80th quality percentile.
  • Identify the environmental conditions and causal factors that relate to those varied nutrient density readings so that we may provide guidance to growers hoping to increase that quality in their crops.
  • Build and calibrate a hand-held nutrient density meter that anyone from grower to consumer could use to get a real time assessment of whatever food they wanted so that there would be no need or role for a bureaucratic certification system that took direct perception away from anyone who wanted to have it.

As a relatively small educational non-profit organization we understood that the capacity to accomplish all of these objectives within our skill sets did not exist, and so the final piece of this strategy was to engage as many aligned partners as possible in this endeavor — our thinking being that we could not claim to have answers without engaging those with the critical subject matter expertise, and that to succeed it would require those perspectives to build the reputation and trust behind our process regardless. We call this partnership the Real Food Campaign.

Foundational in this process were a couple of key allies, Greg Austic now of Our-Sci, and Dorn Cox, now of openTEAM.

To this end we began in 2017 with the creation of our first generation “Bionutrient Meter” and presented it at the Soil and Nutrition conference that fall to great fanfare.

In 2018 we established our first lab to begin identifying the nutritional variation in food, and chose two crops to begin with: Carrots and Spinach. We reached out to our community for volunteers and asked for samples of these two crops to be shipped in to the lab. We received samples from gardens, farms, farm stands, grocery stores and farmers markets. We got local and organic and not organic, and the geographic range ran from Maine to Iowa. We looked at 16 different elements in the crops like calcium and potassium and copper and zinc, as well as polyphenols and antioxidants, two well defined plant secondary metabolites associated with flavor and nutritional value. The results we found in this first year of assessment were nothing short of astounding.

The ratio of variations for mineral levels, depending on the crop, were from 3:1 to 18:1. As in, this carrot has as much copper as those three carrots, and that leaf of spinach has as much iron as those 18 leaves of spinach. When it came to those higher order nutritional compounds’ antioxidants and polyphenols it was 75:1 to 200:1. As in this leaf of spinach has as many antioxidants as those 75 leaves of spinach and that carrot has as many polyphenols as those 200 carrots.

This variation was found across the board and not correlating to local or organic or any other labeling or marketing type. As in, some non-organic carrots in a grocery store had much more nutrition in them than some organic carrots from the local farmers market. This is a preliminary validation of our hypothesis that nutritional variation in food cannot be correlated with specific isms like organic or local.

In 2019 we deepened the work by adding lettuce and cherry tomatoes and kale and grapes to our lab assessment process along with soil from 35 farms across the country where those crops were grown, plus management and environmental conditions data like cover cropping, crop variety, soil minerals, tillage practice, soil carbon, fertility amendments and fertilizers, irrigation type, soil biological activity, mulching etc. With the processing of the 2019 data we now have the ability to begin to overlay all of these different dynamics in relation to each other. After reviewing this information, although from a relatively small data set of 35 farms, it became clear that another one of our hypotheses seemed to be verified. It is that no one factor like type of seed, or no till, or a certain fertility product, correlates with nutrient density variation. It seems that it is a combination of these factors.

Also in 2019 we were able to verify that the dramatic nutritional variation in crops, both in the mineral levels as well as the higher order compounds is present in the broader spectrum of crops assessed. The most significant variation we found was the antioxidant levels in spinach. 364.5:1. That means if you ate one leaf of the highest antioxidant level spinach on January 1 of a year, you would have to eat one leaf of the lowest antioxidant level spinach every day for the entire year to get the same level as you received on January 1st of the most nutritious one!

Most impressively perhaps we were able to build a calibration on those 6 crops for our first generation Bionutrient Meter. That means that anyone who has one can use it in the grocery store or farmers market to get a red / yellow / green answer in real time after flashing a light at any one of those crops. The calibration is not perfect, and stands in for quality. What we are using is level of antioxidants and polyphenols, but most importantly we have proved that we can build a hand held spectrometer at a consumer price point that can be used in real time to non-invasively give readings about nutrient density in food. Building a meter is one thing. Calibrating it so that it gives meaningful readings in real time is a whole other accomplishment. For context I must say that we consider this a proof of concept accomplishment. For those who get the metaphor, lets say that we have built an Apple II, and we are hoping to build an iPhone.

Now in 2020 we have increased the number of crops to 20, from 6 in 2019 and 2 in 2018, and broadened our base of labs from the primary lab in Ann Arbor to the first satellite lab at Chico State in California and our first European lab in partnership with Valorex in the Normandy region of France. We have also gone beyond fruits and vegetables to include oats and wheat.

In addition we increased the number of farms we are getting management data from to 125+ and so should have much more meaningful environmental conditions data to begin to be able to build correlations from management to quality in a way giving significant insight to growers about what limiting factors they can change most efficiently to increase overall function in the biological system of the operation.

As well on the engineering front much work has been accomplished to begin the hardware improvements so that we may begin building the “iPhone” now that we have our “Apple II”.

This work has been accomplished through charitable donations solely and so all information, raw data, hardware engineering, software code, comments, etc. remains open source and will into perpetuity. We are committed to this process of discernment being managed so that free access to the best information is available to all globally regardless of resources. As we continue to succeed in our stepwise process, more and more interest is following this work, and more significant donations are flowing to continue to develop it. Much more is needed, however to bring it to fruition, and in many ways it is access to capital which is the time limiting factor. Will we have a comprehensive definition of nutrient density with cutting edge spectrometers and a deep understanding of how to do better in 2 years or 5? It will likely be in that time window. The level of support is really the primary variable.