Updated May 10, 2025:
Lay Language Final Results- Enhanced Rock Weathering - Weil
Project Background
Maryland soybean farmers and landowners have recently been offered silicate rock powder spread for free as a substitute for limestone powder that farmers normally have to purchase. The funding comes from selling carbon credits. Before having this rock spread on their land, farmers understandably want to know if silicate rock will raise soil pH without any negative side effects. Limestone costs about $100 per ton delivered to Maryland farms. Normally applied at 2-3 tons per acre every few years, this is a significant expense for soybean farmers and a source of greenhouse gas emissions because carbon dioxide gas is released into the atmosphere when calcium carbonate, the main mineral in limestone, reacts with acid soil in a process called rock “weathering.” A major acid that promotes rock weathering is carbonic acid, which forms when carbon dioxide in the atmosphere dissolves in water. This weak acid is highly concentrated in soils because microbial and root respiration produce high levels of carbon dioxide in the soil pore spaces.
In contrast to limestone, silicate rocks like basalt, metabasalt, or olivine do not contain carbon and therefore do not release carbon dioxide as they weather. Instead, carbon dioxide from the atmosphere turns into carbonic acid, which reacts with silicate rocks to form dissolved bicarbonate ions. Natural silicate rock weathering can neutralize acidity and raise the pH, but at rates far too slow to reverse soil acidification. However, rock weathering can be tremendously sped up through a practice called Enhanced Rock Weathering (ERW) whereby a calcium or magnesium silicate rock is finely ground to increase its reactivity and applied to carbonic acid-enriched soils, primarily on cropland. Once the silicate rock powder reacts with carbonic acid in the soil and the bicarbonate produced drains to the ocean, this biogeochemical carbon dioxide sequestration is essentially irreversible. If carried out over millions of acres of land, enough carbon dioxide could be removed from the atmosphere to significantly fight climate change. This process would also provide the co-benefit of neutralizing excess ocean acidity that is currently destroying coral reefs.
With the final carbon sequestration occurring in the oceans, this practice allows farmers to get credit for carbon sequestration efforts (spreading silicate rock powder) without having to pledge to conduct certain practices far into the future as is the case with carbon credits based on increasing soil organic matter, such as growing cover crops or using no-till. While carbon sequestration by increasing soil organic matter does provide soil health benefits, it is beset by uncertainties about the measurement of soil carbon stocks, the prediction of carbon sequestration rates, and limits due to carbon saturation in the topsoil. Enhanced silicate rock weathering is different in that once the rock is applied and the weathering takes place, it cannot be undone because the carbon is being sequestered in the ocean, not on the farm. There is no long-term commitment involved.
The idea behind ERW carbon sequestration is that the bicarbonate produced percolates with drainage water through the soil profile to the groundwater, then to surface streams, and eventually to the ocean, where it is permanently locked up. However, field evidence to evaluate this theoretical process is largely lacking. Several reactions could occur in the soil to reduce the amount of carbon sequestered.
Large corporations attempting to reduce their carbon footprint or go carbon neutral are willing to pay for “carbon credits” if it can be proven that sequestration of carbon dioxide out of the atmosphere through the ERW process has actually occurred. In this research, we partnered with a company (Lithos Carbon) that is offering, for free, to spread a finely ground metabasalt rock waste product from a mine in Pennsylvania near the Maryland border. To many farmers, a free liming substitute sounds a bit too good to be true. Before agreeing to have this rock powder spread on their land, they want to know that the metabasalt rock will actually raise their soil pH to the level recommended for their crops. They also want assurance that there will not be any negative side effects from applying the rock powder. There could even be positive side effects, like improved crop growth from nutrients such as potassium, magnesium, and micronutrients in the metabasalt. Currently, there is little published field research to test the efficacy of powdered metabasalt as a substitute for ag lime or its effects on soybeans and other crops.
If the application of silicate rocks like basalt can effectively substitute for aglime in raising the pH of acid soils, the adoption of ERW practices would be greatly accelerated. Because the infrastructure for, and farmer familiarity with, applying rock powders already exists, scaling up the adoption of basalt application could be relatively easy. Currently, very little research has tested the efficacy of silicate rock as a liming material, its effects on soybeans and other crops, or the assumed carbon sequestration through bicarbonate leaching.
We conducted an on-farm field experiment on two types of soils in Central Maryland with four rock powder treatments applied to replicated strips 60 ft wide x 485 ft long. Our objectives were to 1) compare powdered metabasalt at 7.5 and 15 tons/acre to limestone at 2.5 tons/acre for their capacity to neutralize soil acidity; 2) measure the pH change over time and with soil depth and across two soil types; 3) measure any effects on soil physical condition such as aggregate stability, infiltration, and water holding capacity; 4) monitor any positive or negative effects on soybean and wheat crop yield and composition (heavy metals); 5) quantify bicarbonate produced by reactions with soil carbonic and trace its leaching through the profile to confirm carbon sequestration. 6) Compare the reactivity and acidity neutralization capacity of several rock powders in a range of acid soils using moist aerobic incubation.
Results
Crop Yields and Composition
Across all treatments, the 2024 yields of wheat were quite good for the region (about 100 bu/acre or 6,700 kg/ha) . But the soybean yields were low (about 30 bu/acre or 2,000 kg/ha) due to severe summer drought (Figure 3). Yields of both crops were not affected by the rock treatments. However, yields were slightly higher on the limestone-derived soils in Block D than in the shale-derived soils of Blocks A-C. The wheat grain was analyzed by X-ray Fluorescence, and no treatment differences in heavy metals or nutrients were detected. There were no rock treatment effects on any elements, but block D (limestone-derived soils) produced grain quite different from the rest of the field with shale-derived soils, especially concerning sulfur and calcium, as suggested by the Factor 1 loadings in Figure 4. Toxic elements (Pb, Cr, Cd, Ni) were below detection limits. The soybean seed from 2024 and the 2025 crop will be similarly analyzed.
Water samples
Soil pore water samples were collected approximately every 2-3 weeks as weather allowed between 90 cm depth from February – July 2024 and from 60 cm and 90 cm depths from February – March 2025. The 2025 water samples were still under analysis as of the end of the reporting period for this document. The pH of both the porewater and the stream water varied over time during 2024, ranging mainly from pH 6.5 to 8.5 (Figure 5). The above neutral (alkaline) pH suggests that the water was charged with bicarbonates and hydroxyl from biogeochemical reactions, but there was no indication of greater bicarbonate where the metabasalt rock powder had been applied. Thus, no treatment effects were observable in the pH of the porewater.
Soil pH
We measured the soil pH on samples from three depths, 0 to 10 cm (0-4 inches), 10 to 20 cm (4 to 8 inches) , and 20 to 30 cm (8 to 12 inches). We also used three methods of measuring soil pH, namely making a slurry with pure distilled water, adding enough calcium chloride to that water to make its salt concentration similar to soil solutions (0.1 molar), and adding enough potassium chloride to that slurry to make it a much stronger salt solution. The soil pH measured in a pure water slurry is what most soil test labs in the US perform, but this pH measurement may vary considerably as soluble salt concentrations change through the year. Making the weak calcium chloride solution evens out that variation and creates a constant background so that the pH readings are more reflective of the true acidity and not variations in salinity due to drying or wetting, organic matter decomposition, and fertilizer application. The pH measured in the strong salt solution includes most of the exchangeable acidity held by the clay and organic matter, and is probably the best measure of changes in total acidity or alkalinity of the soil. Typically, the pH measured with a strong salt solution is about one unit lower than when measured in pure water.
Analysis of samples taken in November 2024, about 1 year after the rock powders were applied, showed that very little change in pH with treatment occurred below the 4 inch depth, probably because the rock powders were disked into the soil to about 4 inches. The lack of any treatment effect below that depth suggests that there has been little movement of rock weathering products into the deeper layers during the year since the rock application. Only the agricultural limestone showed a significant increase in pH(KCl) compared to the untreated control plots, and only in the upper 4 inches. The pH(water) did not show any significant treatment effects in the November 2024 samples, even from the limestone. Perhaps because the variability was a bit lower in July and September, samples from those months did show a slight pH increase from the limestone.
Since no treatment effects were detected on any date below 4 inches, we focused on the upper 4 inches to examine the change in pH over time. The data for all three methods of soil pH measurement suggested that there were some seasonal changes, with maximum pH reached in spring of 2024. The only treatment effect that stood out was the effect of agricultural limestone, which had the highest pH on all dates after the treatments were applied. The effect of agricultural limestone was statistically significant on more dates when measured in the salt solution slurries than in pure water because of the greater variability in the latter due to changes in ionic strength from various soil processes. We conclude that limestone was effective in raising the pH of this mildly acidic soil during the first few months after application, but 5 and 10 times as much of the metabasalt rock powder had no significant effect on soil pH, even after one year.
Total Dissolved Substances (EC)
We determined the electrical conductivity of a 2:1 slurry made with the rock-treated soils as a measure of total dissolved substances and therefore the rate of rock dissolution. There was a general downward trend over time as salts that may have accumulated during the very dry summer and fall in 2023 were washed out of the system during winter and spring of 2024, lowering the electrical conductivity. The only rock treatment that significantly affected the EC was the agricultural limestone, which showed higher electrical conductivity on most sample dates. This indicates that the agricultural limestone was weathering faster and releasing more soluble products than the metabasalt, even though it was applied at 1/5 to 1/10 the rate. The electrical conductivity at three soil depths indicated there were no treatment effects below 4 inches. Again, only the limestone had significantly higher electrical conductivity than the control. These data suggest that the dissolution of the metabasalt rock was much slower than the limestone and produced considerably less soluble weathering products.
The limestone was applied at only one rate while the metabasalt was applied at two (higher) rates. The zero rate was on the unamended control plots. Limestone raised the pH by approximately 0.5 units while the metabasalt had no significant effect, even at the higher 15 ton/acre rate.
Lab Incubation Study
The proprietary biogeochemical model used by the company that supplied the metabasalt in our study, Lithos Carbon, appears to have greatly underestimated how much silicate rock would be needed to make the pH change equivalent to the 2.5 ton/acre rate of agricultural limestone that was recommended by the University of Maryland Extension methods. Since this comparative liming effect would be very important for farmers using the metabasalt rock powder, we decided to conduct a laboratory incubation to get more detailed information with which to estimate how much of this or another silicate rock (called Olivine) being offered to farmers would be needed to make changes in soil pH comparable to limestone. We used 3 rates of each rock plus a no amendment control. The highest rate of limestone applied was equivalent to 10 tons per acre, while the two silicate rocks were applied at the rates equivalent to 10, 20, and 50 tons per acre. We used four soils and two layers from each soil, making a total of eight soil materials. The upper 4 inches of each of the soil profiles was much higher in organic matter than the 8-12 inch deep layer, but the texture and mineralogy were nearly identical since the soil had originally been plowed and mechanically mixed at least 30 years before being put under perennial grass vegetation.
This incubation study is still underway as of the end of this reporting period, but results were obtained for the day zero measurements made on air-dry soil amended with air-dry rock powder and then shaken in deionized water for 1 hour. The data suggests that there is a very different relationship between pH and EC (dissolved substances) for the high organic matter surface soil than for the much lower organic matter deeper soil. That is, the organic matter content of the soil had a great influence on the solubility of the rocks in the short term, as well as the amount of dissolved substances initially in the soil. The short-term impact of the limestone was dramatic, with a rapid increase in pH over a small range of rock additions. There was much less if any change in pH with increasing rock amounts for the silicate rocks. All soils and rocks showed a steeper rate of pH increase with the amount of rock added in the high organic matter surface soils. This can be partially explained by the much higher organic matter providing greater buffering capacity and resistance to change in pH. Secondly, the higher organic matter was probably associated with numerous acids, ranging from weak organic decay products to stronger acids formed from nitrate and sulfate.
The change in electrical conductivity of the slurry with the increasing amounts of rock applied showed a strikingly opposite relationship between the high and low organic matter layers as compared to the change in pH. For all of the rocks, the electrical conductivity was much higher in the high organic matter surface layer than in the lower organic matter deeper soil. The rate of increase with the amount of rock applied was also greater in the high organic matter surface soil, except for the olivine. For the deeper low-organic matter soils layers there was much less soluble material and much less change with the amount of added rock. This suggests that organic matter reactions and microbial reactions fueled by organic matter are very important in weathering rocks in the short term. We plan to repeat these analyses as well as do analyses for bicarbonate ions after 30, 90, and 180 days of incubation, but those results are beyond the scope of this reporting period.
The results from this static aerobic moist incubation are expected to provide a baseline of reactivity data to be able to better estimate the amount of silicate rock needed to match the soil neutralizing effect of high calcium agricultural limestone. These results will help to broaden the interpretation of the field study results to other rocks and soil conditions.
Effects on soil structural stability
Finally, we also examined the surface soil for possible changes in physical soil health parameters such as water-stable aggregation. For the samples taken in September 2024, we measured the water stability of 1-4 mm macroaggregates using the wet sieving method. Although soil from all the field plots had rather stable aggregates, soil amended with the high rate of metabasalt, and the limestone had significantly more stable aggregates than the control soil or the soil amended with the low rate of metabasalt. This corresponds to the electrical conductivity effects, showing that the high rate of metabasalt and limestone had weathered sufficiently to increase the ionic strength of the soil solution, which in turn tends to increase flocculation and aggregate stability. Importantly, there was no indication that adding the high rate of basalt, which includes more magnesium than calcium, had any detrimental effect on soil structure
Conclusions
The results to date show that the rates of metabasalt that we applied in the field were not able to significantly raise the soil pH and neutralize soil acidity during the first year after application. Drainage water also showed no treatment effects, even from the limestone, which did increase the soil pH, but only in the upper 4 inches of soil. We found no effects on the yield of soybeans and wheat grown on the field, and no detrimental effects on soil quality. Our continuing incubation study should yield results that allow more accurate prediction of silicate rock rates required to manage soil acidity comparable to agricultural limestone.