Every grading plan in Dallas-Fort Worth is drawn on paper that assumes the ground will hold still. Much of the ground here does not. The dark clay under the metroplex takes on water and swells, gives it up and shrinks, and opens cracks wide enough to lose a boot in. The people who named it black gumbo were describing what it does to a wagon wheel. An earthwork contractor has to describe what it does to a pad, a swale and a parking lot, and build the grade so the clay's habits work for the site instead of against it.
This article reads the ground the way we read it before a bid: from the rock it weathered out of, through the soil series the USDA has mapped across the region, to the laboratory numbers an engineer uses to say how much a given site will move. Then it says what each of those numbers changes about the way the site is graded.
Where the clay comes from
The Blackland Prairie is a soil province before it is anything else. Texas Parks and Wildlife describes the region as underlain by Upper Cretaceous marine chalks, marls, limestones and shales, which gave rise to the characteristic black, calcareous, alkaline, heavy clay soils1. The soils are what is left of a sea floor after a hundred million years of weathering, and their chemistry, high in calcium carbonate and in the swelling clay minerals, is the sea floor's chemistry.
The map of Dallas County shows how close that rock is to the surface. The USGS geologic units for the county are the Austin Chalk on 41 percent of the county's area, the Ozan Formation on 17 percent, Holocene alluvium on 16 percent, Pleistocene and Holocene terrace deposits on 15 percent, and the Eagle Ford Formation on 11 percent, with the Woodbine Formation on a tenth of a percent along the county's western edge2. The county itself is 902 square miles of what the Handbook of Texas calls primarily flat, heavy Blackland Prairie, running from 382 to 850 feet above sea level, with an average rainfall of thirty-six inches a year3. Every one of those figures matters to a grade: the chalk is where rock excavation starts, the alluvium and terraces are where the river-bottom clays and the sand and gravel sit, and the rainfall is what the cracks let in.
Vertisols, the soil order that moves
Soil science has a name for ground that does this. Vertisols are the soil order defined by at least 30 percent clay in all horizons to a depth of 50 centimeters, a layer at least 25 centimeters thick that carries slickensides or wedge-shaped peds, and cracks that open and close periodically as the soil dries and wets4. The shrinking and swelling of the clay heaves the surface into the low undulations called gilgai, which a surveyor sees on a raw Blackland site as a pattern of shallow hummocks and hollows across a field that the plan calls flat4.
Vertisols occupy about 2.4 percent of the world's ice-free land, and roughly 2 percent of the United States, where they occur primarily in Texas4. That is the single most useful fact in this article for anyone bidding work here from somewhere else: the ground that is a footnote in most of the country is the default in North Texas.
The series under a Dallas-Fort Worth site
The USDA Natural Resources Conservation Service maps soils by series, and publishes an Official Series Description for each one that reads like a specification: the clay content of the control section, how wide the cracks open and how deep, how many days a year they stay open, how the soil drains, and what it weathered from. The seven series below account for most of what our crews grade between the Red River and the Brazos. The table is read straight from the descriptions.
| Series | Where it sits | Clay in the control section | Cracks when dry | Days open | Drainage | Weathered from |
|---|---|---|---|---|---|---|
| Houston Black | Uplands of the Blackland Prairies (MLRAs 86A and 86B) and the eastern Grand Prairie; slopes mainly 1 to 3 percent5 | 40 to 60 percent5 | 0.5 to 4 inches wide, from the surface to 12 inches deep5 | 90 to 150 cumulative days in most years5 | Moderately well5 | Clayey residuum from calcareous mudstone of Upper Cretaceous age5 |
| Heiden | East central Texas, MLRA 86A; slopes 0.5 to 20 percent6 | 40 to 60 percent6 | 0.5 to 3 inches wide at 12 inches or more6 | 90 to 150 cumulative days6 | Well, with very slow permeability6 | Clayey residuum from mudstone; densic contact at 40 to 80 inches6 |
| Branyon | Treads of stream terraces on river valleys; slopes 0 to 3 percent7 | 40 to 60 percent7 | 1 to 3 inches wide to 20 inches or more7 | 90 to 150 days in most years7 | Moderately well, very slow permeability7 | Calcareous clayey alluvium from mudstone of Pleistocene age7 |
| Ferris | East central Texas, MLRA 86A; slopes 1 to 20 percent, most areas once cultivated and eroded8 | 40 to 60 percent8 | 0.5 to 3 inches wide through 12 inches or more8 | 120 to 150 cumulative days8 | Well, very slow permeability8 | Clayey residuum from calcareous mudstone8 |
| Trinity | Flood plains of river valleys, MLRA 86A; slopes mainly under 1 percent; occasionally to frequently flooded9 | 60 to 80 percent9 | Up to more than 1 inch wide to 20 inches or more9 | Less than 90 cumulative days9 | Moderately well9 | Calcareous clayey alluvium from mudstone9 |
| Austin | Blackland Prairies, MLRA 86A; slopes 0 to 8 percent10 | 35 to 55 percent, with a calcium carbonate equivalent of 40 to 70 percent10 | None described10 | Not applicable | Well, moderately slow permeability10 | Residuum from chalk, mostly the Austin Formation; platy chalk at 29 to 30 inches10 |
| Eddy | A narrow band along the western edge of the Blackland Prairie; slopes 1 to 20 percent11 | Very gravelly clay loam, 35 to 60 percent chalky limestone fragments by volume11 | None described11 | Not applicable | Well, moderately slow permeability11 | Residuum from weathered chalk of the Austin Chalk; bedrock at 3 to 14 inches11 |
Read across the rows and the grading problem sorts itself into three kinds of ground. The first four series are the classic Blackland clay: Houston Black, Heiden, Branyon and Ferris all carry 40 to 60 percent clay and all crack open for three to five months of a normal year5678. Trinity is the river bottom, with 60 to 80 percent clay and a flooding frequency written into its description, which is a floodplain problem before it is a soil problem9. Austin and Eddy are the chalk country along the western edge of the prairie: shallow, well drained, with rock at 30 inches on Austin and as little as 3 inches on Eddy, so the cut that was clay on the east side of a site can be rippable chalk on the west1011. A single commercial footprint in Dallas or Collin County can cross two of these boundaries, and the county survey, published by the NRCS through the Web Soil Survey, shows where they fall before a boring is drilled12.
The numbers an engineer uses to say how much it will move
A soil description says a series has high shrink-swell potential. A geotechnical report has to say how many inches, and it gets there through a short ladder of laboratory tests that every earthwork contractor in this region should be able to read.
The Atterberg limits and the plasticity index. The liquid limit is the moisture content at which a soil goes from plastic to liquid, the plastic limit is where it goes from plastic to semisolid, and the plasticity index is the range between them: the width of the moisture window in which the clay behaves like modeling clay13. ASTM D4318 is the test, and its own scope says the limits are used, with other soil properties, to correlate with engineering behavior such as compressibility, permeability, compactibility, shrink-swell and shear strength13. A high PI is the first sign of a soil that will move.
The classification. The Unified Soil Classification System, ASTM D2487, plots liquid limit against plasticity index and assigns a group symbol; the boundary between low and high plasticity falls at a liquid limit of 50, so a Blackland clay reports as CH, high plasticity clay, and a leaner clay as CL14. The letters on a boring log are the fastest way to see how much of a site is fat clay.
The building code's definition. The International Building Code defines expansive soil by four criteria, and a soil that meets all four is treated as expansive: a plasticity index of 15 or greater by ASTM D4318, more than 10 percent of the particles passing the No. 200 sieve, more than 10 percent finer than 5 micrometers, and an expansion index greater than 20 by ASTM D482915. The code lets the engineer skip the first three tests if the expansion index is run15. The expansion index test itself inundates a compacted specimen with distilled water and reads its swell as an index number, which ASTM describes as a simple, sensitive way to evaluate expansion potential for practical engineering use16. Where the engineer wants inches rather than an index, ASTM D4546 measures one-dimensional swell or collapse of a specimen on wetting, on either a reconstituted specimen that simulates compacted fill or an intact specimen from the natural deposit, and the result is used to estimate heave or settlement of the soil profile17.
Potential vertical rise. The Texas number is PVR. TxDOT's test procedure Tex-124-E defines potential vertical rise as the potential of soils to swell in the vertical direction at a given density, moisture and loading condition when exposed to capillary ground or surface water, raising the elevation of the upper surface along with anything resting on it, and it is expressed in inches18. The procedure samples the soil horizon in layers, with greater emphasis on the top layers to a depth of 15 feet in most cases and as much as 20 feet when highly expansive clays are present, works the profile in 2-foot increments, and estimates the rise from each layer's liquid limit, moisture content, percent soil binder and plasticity index under the load of the soil above it18. When a geotechnical report for a North Texas pad states a PVR and a target, that is the calculation behind it, and the pad is built to bring the number down.
What each of those properties does to a grade
The cracks decide where the water goes. A Houston Black profile opens cracks half an inch to four inches wide from the surface to a foot deep, and holds them open for 90 to 150 days in most years5. Rain that would run off a graded pad instead runs into the cracks, saturates the clay from the top down, and swells it shut. The first inch of rain after a dry August does more to a finished subgrade than a week of drizzle in February. So the grade has to move water off the clay quickly, with positive slope and no place for it to stand, and the surface has to be closed with a compacted lift before the rain arrives.
Moisture is the variable, not strength. Because the volume change is driven by wetting and drying, a clay placed dry and then wetted will heave, and a clay placed wet and then dried will shrink and settle. The code's three answers to expansive soil say this in a sentence: a foundation designed to resist the differential volume change, removal of the soil to a depth sufficient to ensure a constant moisture content in what remains, or stabilization by chemical, dewatering, presaturation or equivalent techniques19. Two of the three are earthwork. Removal and replacement is the undercut and select fill on a pad; presaturation is moisture conditioning, bringing the clay to and slightly above its optimum moisture before it is compacted so it is already near its swollen state when the slab goes on.
Chemical treatment is chosen by the soil, and it is checked for sulfate first. TxDOT's guidelines for modifying and stabilizing soils lay out the order: classify the soil and run the Atterberg limits, measure the sulfate content by Tex-145-E, and if the sulfate is above 3,000 parts per million, stop and use the separate guidelines for sulfate-rich soils, because those soils can heave when lime is added20. Below that threshold, the additive follows the plasticity index. For a subgrade with 25 percent or more passing the No. 200 sieve, the guidelines list cement, lime-fly ash and asphalt for a PI under 15; lime, lime-cement, lime-fly ash, fly ash and cement for a PI of 15 to under 35; and lime, lime-cement and lime-fly ash for a PI of 35 and above20. Lime is on every list for the fat clays because it reacts with the clay minerals, but it is one additive among several, the mix design sets the rate, and the sulfate test comes before any of it.
The shallow chalk changes the equipment. On the Austin and Eddy series the problem inverts. Platy chalk sits at 29 to 30 inches under an Austin profile and partially cemented chalky limestone at 3 to 14 inches under Eddy1011. A cut that was scraper work on the Houston Black side of the site becomes ripping and, at depth, hammer work, and the material that comes out is chalk that will not compact like clay. The boring log and the county survey tell you where that line is; the bid has to have already crossed it.
How we grade on it
None of the above makes Blackland clay unbuildable. It makes it a soil with a procedure. Ours starts with the county soil survey and the geotechnical report before the takeoff, so the estimate is built on the ground that is there. Fill goes in at controlled moisture in tested lifts against the laboratory curve the report specifies. The finished subgrade is walked for standing water before it is turned over, because a low spot on Houston Black is a swelling problem the moment it rains. And when a site crosses from clay to chalk, the cut is bid and scheduled as two kinds of excavation, not one.
The field bears the numbers out. Our crews have graded sites in Frisco where the geotechnical report put the plasticity index in the fifties, well past the code's threshold of 15 and past the point where TxDOT's guidelines move every additive choice into the lime family1520. That is not an unusual North Texas number. It is the reason the geotechnical report, and not the plan sheet, sets how a pad here is built.
Three questions to settle before grading on clay
- Which series is under the footprint, and where does it change? The Web Soil Survey answers it for free before a boring is drilled12, and the answer decides whether the cut is clay, chalk or river bottom.
- What is the PVR, and what is the target? Tex-124-E gives the inches18; the difference between the two is the undercut, the moisture conditioning and the treatment the pad will need.
- What is the sulfate content? Above 3,000 parts per million the treatment plan changes before it starts20, and the test is the smallest item on the list.
Send us the geotechnical report and the site address, and we will read the ground with you before the takeoff is built.
