Can Dirt Become Soil Again?

Does soil become "dirt" when it's in an urban context? And if so, can it be made into soil again?

I’ve known Dr. Bryant Scharenbroch, a soil researcher at Morton Arboretum and assistant professor of soil science at University of Wisconsin – Stevens Point, for 11 years. In that time, he’s been involved in some excellent research. One of his most recent projects, done in collaboration with Katie Klaus (Morton Arboretum), James Urban (Urban Trees + Soils), and Mike Curry (Greensite) is a field trial of three different growing media to assess their effect on tree performance. I got a chance to talk to him about the study and their initial findings. – Peter MacDonagh

If you were in an old growth forest, natural area, or a property you knew had been continuously grazed (but not plowed) in the Upper Midwest, you could dig a hole about three feet (one meter) deep, take a handful of soil, and notice it smells pleasant.

It smells good because there is lots of air in the soil, about 25% by volume. In addition to plenty of air, real soil has structure, sand, silt, clay, organic matter, and lots of living micro-organisms. It also has “horizons” – layers – and the top layer of soil is quite a bit darker than the ones below due to staining from organic matter (leaves, branches, roots, worms, etc.). Here in this dark soil, organic matter feeds billions and billions of oxygen breathing micro-organisms that are cultivating this soil. Rains have been washing through the soil for hundreds, sometimes thousands of years, and the soils further down have been leached of their organics. These soils may also smell bad, because there’s almost no air (<3% oxygen) at that depth.

This all changes in an urban context.

Soil in the city has no real horizons, and it contains plenty of anthropogenic (manmade) materials. Bryant has encountered slag from iron smelting, buried original soils, burned clinkers, fly ash, construction waste, garbage fill, broken glass, wetland mucks, old foundations, and chunks of buildings. Is this even soil anymore? Not exactly. Bryant calls this material “A Harsh Dirt.”


H     Hot

I      Inert
R    Reduced
T    Tainted

Soil scientists have named the most common stuff you end up with in cities as “Urdothents,” but “dirt” will do for most people. The common perception of “dirt” from non-soil scientists varies widely. People tend to think it’s either all bad or all good. On the bad end of the spectrum, there is the mindset that “dirt” is useless, and needs to be disposed of and replaced with something screened and purchased from a supplier. At the other end of the spectrum there is the mindset that dirt is as perfect as a potting soil, and that everything planted in it will grow fine. Neither assessment is accurate.

Damaged urban soils (dirt) are salvageable, however. They can be repaired and that’s what Bryant and his team are working on. They’re researching how soil types affect tree growth response. One key question Bryant would like this research to help answer is, “can we design a better soil for trees in urban landscapes?”



The research project was set up as follows:

Bryant and the team planted 66 trees in 20-gallon pots in three different kinds of growing media. The trees were 3cm (1.2″) caliper, bare root Tilia “Shamrock” (Lindens). Prior to planting, the trees had half their roots cut off, both to fit into the 20 gallon pots and to provide as much room as possible for new root growth.

The media types were:

  • Native soil: 100% topsoil (not screened below 2 inches)
  • Tree soil: 60% topsoil, 15% compost, 25% sand (not screened below 2 inches)
  • Urban soil (sand based soil): 60% sand, 25% topsoil, 15% compost (screened below ¾ inch)

All growing media were compacted to 80% Proctor, about the same level of compaction as walking on soil. Two different cover types, wood chip mulch or bare soil, were also used. The wood chips were plain old utility line grade woodchips.


All in all, there were 22 trees per media type and 11 replicates. They were monitored for various growth indicators over the course of one year (including an entire Chicago winter growing at the Morton Arboretum). Three growth indicators were monitored:

  • Soil respiration: Carbon Dioxide and Oxygen (CO2, O2) levels to measure micro-organism activity;
  • Tree leaves: size, chlorophyll content, moisture content, stomatal respiration activity, insect herbivory;
  • Tree dimensions: trunk caliper diameter; tree height.

The findings thus far have been interesting and significant (meaning that the data results that are greater than the expected standard error). Bryant summarizes them as follows:

  1. Trees growing in native soil and tree soil exhibited significantly longer season of stomatal conductance, and higher leaf biomass production, and the soil itself had significantly higher organic matter, and lower salts and pH compared to urban soil (sand based soil).
  2. Tree soil exhibited significantly lower soil bulk density (easier for tree roots to penetrate), lower salts and pH, and longer stomata conduction compared to urban soil (sand based soil).
  3. Urban soil (sand based soils) had higher bulk density (harder for tree roots to penetrate), higher pH (more alkaline), and lower organic matter compared to tree soil and native soil:
  4. Native soil had significant higher moisture content and significantly lower pH compared to urban soil (sand-based soil)
  5. Trees that were mulched had more biomass, more moisture and soil respiration, lower salts, and less extreme temperatures in winter and summer in all three treatments (native, tree, urban).

What does this say about the different media being studied? Bryant believes that sand based soils (60% sand, 25% topsoil, 15% compost) are generally worse to a lot worse than tree soils (60% topsoil, 25% sand, 15% compost) and native soils (100% topsoil). Overall, tree soils performed the best for growing trees.

This is extremely encouraging, because it means that the “dirt” that is so common on urban sites may be transformed into effective soils for growing trees. The “tree soil” was 60% native topsoil. Wholesale replacement isn’t necessary. So then what should someone who wants to use the existing dirt on a site do in order to get it “tree ready”?

Bryant and I discussed this process, and I broke it down into several steps. You can salvage the dirt on a given site by doing the following:

  • Don’t run anything except track vehicles on it.
  • Don’t export existing topsoil off-site.
  • Move topsoil a maximum of three times.
  • Don’t screen it for clods, rocks, chunks of wood or peds below 2 inches (5 cm).
  • When moving it to its final home, mix in 25% sand and 15% compost (by volume). Don’t overmix!
  • Don’t use anything heavier than your foot to compress the soil in the planting hole.
  • Rethink your maintenance plan.

Bryant and his team will be publishing their results when the study is complete. Stay tuned.


L. Peter MacDonagh is director of science and design for The Kestrel Design Group.

Lead image: Randy McRoberts /CC BY 2.0
All other images: Bryant Scharenbroch


  1. In the irrigated west that 15% compost will quickly degrade causing some slight soil settling. Irrigation will flow into these lows causing irrigation to be even more uneven. It can easily lead to chronic wet spots
    Most clay in the west is geologically young and very electically active. 25% sand is almost trivial in soils with this kind of clay

  2. As a soil scientist, I can’t see how differentiating between soil and “dirt” is useful. Let’s stick with accurate soil assessments without the unnecessary step of assigning a derogatory term to it. Urban soils with included construction debris, refuse, and other artifacts will present management challenges, but there’s a fair chance that the typical Udorthent consists of clean fill material with a layer of decent topsoil. It may be a young soil but so is the sediment that was naturally deposited during last summer’s flood.

Leave a Reply

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