Which Dynamic Accumulator Plants are Actually Helpful for your Garden, According to Science?

Jump to audio version of this post

Since the 1980s, permaculturists have been teaching about dynamic accumulators: long-rooted plants that can reach down to the subsoil and scavenge hard-to-reach minerals, which they store in their leaves and stems. When the plant is mulched or dies naturally, it releases those minerals into the uppermost soil horizon where shallow-rooted plants can access them.

Popular recommendations of dynamic accumulators include comfrey, amaranth, chickweed, dandelion, chicory, lamb’s quarters, nettles, and many trees; however, new research suggests that we have misidentified some dynamic accumulators. At the same time, new analysis of longstanding USDA data reveals as many as 340 plant species that could be contenders for the title.

For decades, the effectiveness of dynamic accumulators was supported by anecdotes alone, but in December 2021, for the first time, an early report from a Cornell University-supported field trial confirms that common nettle can be considered a dynamic accumulator of calcium, lambsquarters as a dynamic accumulator of potassium, and Russian comfrey as a dynamic accumulator of potassium and perhaps silicon. However, only nettle definitively scavenged nutrients from the subsoil, which is an important part of the picture. If comfrey is extra good at accumulating potassium but mostly takes it from the upper level of soil, it’s actually competing with surrounding plants rather than helping them.

While a single study can’t be conclusive about all of these factors, it did create an online database to simplify future studies. And, most useful of all, it established a measurable definition for what a dynamic accumulator is: a plant that accumulates a beneficial nutrient in its tissues at a concentration that is at least 200% of the average concentration for that nutrient across all plants.

Using known data about nutrient concentration from a wealth of other studies, 340 plant species seem to qualify as dynamic accumulators. The 2020-2021 study, which was conducted at Unadilla Community Farm in New York State, demonstrates the kinds of methodologies that can reveal if those species are indeed useful in practice.

The Study

The report now available on the Unadilla Community Farm website is comprehensive and easy for a layperson like me to read, but for those not prepared to read the 35-page document, I’ll set out the broad strokes here. 

Setting Thresholds to Define Dynamic Accumulators

The first hurdle to studying dynamic accumulators was the lack of an established way to define them. The researchers therefore set out to create baseline thresholds to identify what make dynamic accumulators distinct from other plants: exactly how good at concentrating nutrients do they have to be to earn the label?

Fortunately, an extremely useful resource existed already in the USDA-hosted “Dr. Duke’s Phytochemical and Ethnobotanical Databases,” which compiles thousands of data points from decades of research about the chemicals present in different plants. I won’t drag you through the specifics of calculating the biome-concentration factor (Bf) of each plant, but the gist is that it represents how plants match up against each other in concentrating a specific nutrient. A dynamic accumulator, they decided, would need to have a Bf twice as high as the average. Of all the plants in the database, 340, just over 10% qualified for one or more nutrients. This data has been made publicly available in the Google Sheet “Dynamic Accumulator Database & USDA Analysis”. 

Selecting the Plants to Study

To select the species for the trial, the researchers cross-referenced their 340 qualifying plants with a list of the plant species most commonly espoused as dynamic accumulators by permaculturists. From plants that appeared on both lists they chose five perennials or self-seedling annuals well suited to their climate in UDSA zone 4:

• redroot amaranth (Amaranthus retroflexus) 
• lambsquarters (Chenopodium album)
• dandelion (Taraxacum officinale)
• red clover (Trifolium pratense)
• common nettle (Urtica dioica)

To these five they added Russian comfrey (Symphytum peregrinum) because, although no nutrient data was available for it in Dr. Duke’s databases, this plant is hugely popular as a supposed dynamic accumulator, and it is known to produce an extraordinary amount of biomass.

The Physical Site

In unamended soil, two rows of 2.5 x 30 feet were planted for each plant, along with one fallow row for a control. In the first year the plants were allowed to establish themselves, and in the second year they were harvested. For each plant, one row was chopped and dropped to create mulch while the other was harvested to make liquid extracts.

To avoid skewing the data, the rows were cared for identically even when some of the plant species performed poorly in the given conditions.

The Tests

To prove the usefulness of dynamic accumulators, the researchers had to demonstrate three things:

1. That the plants were accumulating beneficial nutrients in their tissues
2. That they were doing so at concentration levels higher than those found in the surrounding soil
3. That they were drawing significant amounts of the nutrients from the subsoil rather than competing with with other plants for them at shallower depths.

To that end, the following tests were performed:

• Analysis of soil samples at depths of 0-6”, 6-12”, and 12-18” taken at three different stages: before planting, after the first year, and after the second year.
• Analysis of plant-tissue samples in the second year.
• Analysis of liquid plant extracts (since many people use these plants for compost tea) in the second year.

The Results

A full re-stating of the results here is impractical when they can be viewed in Unadilla Community Farms’ report, but I will highlight some of the most interesting take-aways:

• Many of the plants performed below dynamic accumulator thresholds even though existing data shows they are capable of exceeding them. This is a reminder that knowing how well these plants can concentrate nutrients in optimal conditions does not necessarily mean we can expect them to help improve poor soil.
• Common nettle was the most clear-cut champion, accumulating the highest concentrations of calcium in its leaves and in the upper two soil levels while simultaneously reducing calcium levels in the soil below 24 inches. This suggests that, by any definition that has so far been advanced, common nettle is a dynamic accumulator.
• Russian comfrey performed above dynamic accumulator thresholds for calcium and possibly for silicon; however, more baseline data for silicon is needed to determine what that threshold is.
• Lambsquarters surpassed the threshold for potassium
• In many cases the liquid extracts did seem to extract nutrients well; however, high acid levels in the rainwater used to make the tea complicated the results.

For Future Study

These field trials are just a first step in a broad, untapped field of study. The report’s thorough notes about how each plant performed in the trial’s conditions, compared to the aggregated USDA data for the plant, reveals clear paths for further research in several areas, notably:

• How growing conditions affect the ability of a plant to perform the function of a dynamic accumulator. For example, redroot amaranth exceeded the dynamic accumulator thresholds in the USDA data, but in the trial it grew poorly and also failed to meet the thresholds.
• Whether certain plants with a reputation for accumulating one mineral should be investigated for another; for example, redroot amaranth, known for concentrating high levels of magnesium and zinc, also showed the trial’s highest levels of aluminum, manganese, and sulphur.
• Whether a dynamic accumulator is, in fact, useful for scavenging minerals in poor soil
• How liquid extracts made from the dynamic accumulators and rainwater would be affected if the rainwater’s pH were corrected to an ideal standard for nutrient solubility
• Whether measuring nutrient concentrations in fresh leaves instead of dry would offer a more valuable comparison to the nutrient concentration in the extract
• How creating and testing multiple batches of liquid extracts for each plant could improve the accuracy of data

The authors also advise that generating BAFs for a large number of plant species in a wide range of growing conditions would help to better identify which results are due to the traits of dynamic accumulators and which have more to do with soil conditions or processes.

The Unanswered Question of Solubility

One aspect of this subject that I haven’t addressed is the question of when minerals from the tissues of dynamic accumulators become bio-available to other plants. It’s generally understood that these minerals are not likely to be entirely in soluble forms that another plant can immediately absorb. Instead, they’re bound up in larger molecules that must then be broken down by other members of the soil ecosystem. The authors of the Unadilla Community Farm report note that their tests measured total nutrient concentrations rather than available nutrients, so their field trials do not attempt to answer this question.

Interestingly, in Kourik’s take on dynamic accumulators, the question is less fraught. He discusses them as helpful plants to add to a compost heap, so he expects time and chemical processes to take place before the nutrients are returned to the soil. The permaculture community at large, however, has embraced dynamic accumulators as chop-and-drop plants, with the leaves dropped as mulch directly in the garden. How long, then, they wonder, will it take for those leaves to release their benefits? 

Personally, I can’t help remembering that even mulching for fertility takes time, and we all accept that. We know that a new garden needs thoroughly composted material to get started, and any mulch we add may take three years to break down enough for their nitrogen to become available to new plants. And so we amend in the first few years and then make sure to add a regular supply of new mulch to perpetuate the cycle. Certainly it would be helpful to know more about the timeline for other elements, but that alone shouldn’t determine the usefulness of dynamic accumulators. After all, we are designing for long-term systems. 

Regardless, for all of us who are anxious for a more evidence-based approach to permaculture, this study is a welcome first step. With the enticement of decades of anecdotes to make researchers curious, and with the foundations laid by the “Dynamic Accumulator Database and USDA Analysis,” I hope many more researchers will continue down this fascinating path. 

A Timeline of the Dynamic Accumulator Controversy

1986 Robert Kourik’s book Designing and Maintaining Your Edible Landscape—Naturally includes a chart on page 269 titled “Dynamic Accumulators.” 

1992 The USDA makes available “Dr. Duke’s Phytochemical and Ethnobotanical Databases.” These databases are drawn from the work of Dr. James Duke, former Chief of USDA’s Economic Botany Laboratory in the Agricultural Research Service, as well as data aggregated from studies around the world related to the chemical profiles of plants. Users can search by a plant name to find what chemicals it contains, or search by a chemical to learn which plants contain it (among other search functions). 

Relevant to the subject of dynamic accumulators, the databases contain hundreds of data points about the concentration of various nutrients in plants’ leaves, stems, roots, seeds, etc. 

2001 Gaia’s Garden by Toby Hemenway includes the chart of dynamic accumulators. It goes on to sell more than 250,000 copies in its first edition.

2005 The award-winning two-volume work Edible Forest Gardens by Eric Toensmeier and Dave Jacke includes references to dynamic accumulators.

2014 Robert Kourik publishes an article titled “Unstacking Functions: Dynamic Accumulators.” I have not been able to find the original location of this article, but a copy was anonymously uploaded to Scribd. I’m reluctant to assume that copy is the final, published version (some notes suggest it may be a late draft), but articles and forum discussions from the time corroborate the following: Kourik expresses his regret about including the original list in his book and explains that it was based largely on information from a number of books, including: 

• Weeds: Guardians of the Soil (Joseph Cocannouer)
• Practical Organic Gardening (Ben Easey)
• Stalking the Healthful Herbs (Euell Gibbons)
• Weeds as Indicators of Soil Conditions (Stuart Hill and Jennifer Ramsay)
• Weeds and What They Tell (Ehrenfried Pfeiffer)
• The Organic Method Primer (Bargyla & Gylver Rateaver) 

The books themselves depend on anecdotes and personal experience rather than formal evidence or exact measurements. 

The article also proposes using the nutrient concentration data of “Dr. Duke’s Phytochemical and Ethnobotanical Databases” to formally identify accumulators.

2015 Permaculture News publishes ecologist and microbiologist Dean Brown’s article “Qualifying Dynamic Accumulators: A Sub-Group of the Hyperaccumulators.” It identifies the value of studying dynamic accumulators (the leaching of nutrients from the topsoil to the subsoil is rampant, and we don’t have any other functional strategies to return them), and proposes to qualify them (plants that spend energy to accumulate nutrients at higher concentrations than the soil around them). 

Brown acknowledges Kourik’s proposal to use the nutrient concentration data from “Dr. Duke’s Phytochemical and Ethnobotanical Databases” to identify accumulators and describes an extra step that is needed to account for the fact that different soils will give different bioconcentration results. He proposes to divide a given plant’s nutrient concentration by the average-plant concentration of that nutrient. The resulting ratio, which he terms the biome-concentration factor (Bf), can be compared to the Bf of every other plant, and the plants with the highest Bfs will be contenders for the label of dynamic accumulator. The challenge is that someone will need to do all of these calculations. 

2015 The YouTube channel Alberta Urban Garden Simple Organic and Sustainable uploads a video with results from lab tests on dried comfrey leaves: Total NPK (nitrogen, phosphorus, potassium), immediately available NPK, and trace elements. While a single sample size is not definitive evidence and there is no data about how the nutrient levels in the soil changed, it’s the first widely available demonstration of such tests.  

2016 User Roberto pokachinni begins a thread on Permies.com discussing the history of the dynamic accumulator issue and the need for scientific study, including some proposed methodology (“The truth about Dynamic Accumulators: Science is needed!”). The discussion includes some relevant comments that cannot now be verified because the linked articles no longer exist, for example: “Hemenway is removing the table from future editions of Gaia’s Garden and is intending to replace it with a general note about accumulators, and Toensmeier says that he has stopped teaching about Dynamic Accumulators in his courses. They are both hopeful that studies will prove the merits of the accumulators.” (Roberto pokachinni) 

Robert Kourik joins the thread to give an example of the lack of data to support comfrey as a dynamic accumulator. 

2016 Robert Kourik publishes a YouTube video clarifying a common misapprehension that, because they have long taproots, plants identified as dynamic accumulators are necessarily pulling most of their nutrients from the subsoil. This belief ignores the plants’ lateral roots which may compete with shallow-rooted plants for nutrients in the topsoil.

2017 In the related and better-researched field of hyperaccumulators (plants that concentrate high levels of toxins in their tissues, enabling us to clean up polluted sites once the plants are removed), Reeves et. al. note in the publication New Phytologist that “spiked” soil is a confounding factor in the identification of hyperaccumulators. That is, if a plant can only achieve high concentrations of a metal if the soil has been amended with artificially high levels of it, that plant should not count as a hyperaccumulator.

2020 Unadilla Community Farm in central New York State announces receipt of a research grant from SARE (Sustainable Agriculture Research and Education, Northeast United States), in coordination with Cornell University, to establish thresholds to identify dynamic accumulators and to conduct the first field trials. A Permaculture News article titled “Breaking Ground With Dynamic Accumulators” describes the expected research.

With support from USDA-NIFA, Unadilla Community Farm creates the online “Dynamic accumulator database and USDA analysis” tool, using the data from “Dr. Duke’s Phytochemical and Ethnobotanical Database. The tool calculates average concentrations of 20 beneficial nutrients across thousands of entries for 340 plant species. The researchers use this data to set thresholds for the label of “dynamic accumulator”: to qualify, plants must concentrate approximately 200% more of a nutrient than the average plant does. In the same year, test plots are planted for six promising species. 

2021 In December, the report of the two-year field trial is made available on the Unadilla Community Farm website. 

Audio Version

Bibliography

Ben Tyler. (2020) “New research on Dynamic Accumulators (incl. the Mother of all Lists).” [discussion post]. Permies. https://permies.com/t/138059/research-Dynamic-Accumulators-incl-Mother

Brown, Dean. (2015) “Qualifying Dynamic Accumulators, a Subgroup of the Hyperaccumulators.”  Permaculture News. https://www.permaculturenews.org/2015/05/12/qualifying-dynamic-accumulators-a-sub-group-of-the-hyperaccumulators/ 

Kitsteiner, John. (2015). “The Facts about Dynamic Accumulators.” Permaculture News https://www.permaculturenews.org/2015/04/10/the-facts-about-dynamic-accumulators/ 

Kourik, Robert. “Unstacking Functions: Dynamic Accumulators.” Accessed on January 2, 2022 at https://www.scribd.com/document/376547571/Dynamic-Accumulators

Kourik, Robert. (2016). “Dynamic Accumulators?” Robert Kourik. YouTube video. https://www.youtube.com/watch?v=3l7Nm326QVc 

Roberto pokachinni. (2016). “The truth about Dynamic Accumulators: Science is needed!” [discussion post]. Permies. https://permies.com/t/53073//truth-Dynamic-Accumulators-Science-needed

Tyler, Ben. and Zarro, Greta. (2021). “A Dynamic Accumulator Database and Field Trials for Six Promising Species” https://unadillacommunityfarm.org/dynamicaccumulators/ 

Zarro, Greta. (2020). “Breaking Ground with Dynamic Accumulators.” Cornell Small Farms Program https://smallfarms.cornell.edu/2020/07/breaking-ground-with-dynamic-accumulators/