Optimizing spring cover crop management for productivity, soil health and climate resilience

Cover Crop Biomass Production in Fall 2023 (pre project funding)

In early Fall/late summer 2023, cover crops to receive spring management treatments were planted on collaborating farms and at the  Central Maryland Research and Education Center (CMREC), in Beltsville, MD.  At CMREC, the cover crops were broadcast overseeded into standing corn canopies in late August 2023 and into standing soybean canopies at leaf yellowing in early September. Due to nearly a month without any significant rain, the cover crops didn't germinate until mid-September. Different rates of side-dressed nitrogen had been applied in June to corn plots as part of a prior agronomic experiment on the site.  During the first year of the project cover crop experiments, this differential N could theoretically affect the cover crop performance in spring 2024 due to the effect of corn vigor in competing for light during early cover crop growth before corn harvest and due to possible residual N remaining on the soil for use by the cover crop plants.  To assess this potential carry-over effect at CMREC, fall biomass was hand-harvested in December 2023 for the radish-crimson clover-rye mixture (3-Way) cover crop that was intersown into the corn before harvest. This data was collected for all four replications at the sandy soil site, and two of the replications at the silty-clay soil site.  A statistically significant difference in aboveground cover crop dry biomass was observed between the high (160 or 180 lb N/acre) and low (0 lb Nacre) side-dress rate (Figure 1).  The range of biomass values from about 600 to 1000 kg/ha is lower than expected for an early intersown cover mixture at this location, probably due to lack of rain during the early cover crop growth period. In addition, the N rate applied to corn in June differentially affected the rye and clover components of the cover crop, the higher N applied to corn in June increasing the relative amount of rye while decreasing the amount of clover (Figure 2).  These results would be expected since the main competitive advantage of the clover is its ability to fix its own nitrogen, while the faster-growing rye would be more competitive where more N was available in the soil.  We conclude that 1) N applied at the higher rate was not efficiently used by the corn crop since residual N was available to stimulate the growth of the rye cover crop in September-December.  This was true across both the sandy and silty clay soils.  In addition, the differential effect on species in the fall biomass suggests that we might see the effect of N applied to corn in one year affect the performance and N-fixation ability of the following cover crop in the spring of the next year.  Although we will not apply different rates of N to corn in 2024, we monitored the cover crop biomass from all termination date subplots at the time of early termination (early April) and determined that this effect did not carry over into the spring.

Cover Crop Root and Shoot Biomass Production 2024

Figure 3 illustrates observations on the root - shoot biomass. Roots and shoots were sampled using the soil cube technique (see Figure  , below). Shoot biomass changed very little with sampling dates that correspond to the early mid and late termination. However, the rye + clover vegetation (which is all that remained of the three-species cover crop) produced more shoot dry matter than the sole rye even though the latter had a more dense population. Root dry matter was affected by both the cover crop type and the sampling date. The greatest root dry matter from both the sole rye and the rye+clover occurred at the mid-termination time which was the end of April.  The decline in root biomass between the mid and late kill dates does not necessarily represent less contribution of organic carbon to the soils. Rather the decrease in root DM may reflect carbon added to the soil during the the sloughing off of old roots and and regrowth of young ones. Shoot-to-root ratios varied with the type of cover crop and with the sampling date. The shoot-to-root ratios were lowest for the mid-sampling date and more than twice as high for the early and late sampling dates. For the late sampling date, the shoot-to-root ratio was almost twice as high for the rye +clover as for the sole rye. These data illustrate the inappropriateness of modeling cover crop below ground biomass by using a given shoot-root ratio for a given cover crop species.

Cover crop performance in 2024.

Cover crops produced moderate levels of biomass in the field experiments at the Central Maryland Research and Education Center (CMREC) Beltsville facility. This was partly due to less-than-ideal conditions for inter-seeding in the Fall of 2023 and also due to very cool and very dry conditions in April 2024. There was little difference in cover crop biomass between early mid and late termination dates, but the rye cover crop tended to increase biomass with time, while the three-way cover crop dominated by crimson clover showed little change or even a slight decrease (Figure 4, left).  Nonetheless, cover crops and their residues did have an impact on soil moisture during the summer growing season, which was extremely dry and hot. The cover crop effects began to show up in the soil moisture tension data in late June, and by mid-July, there was significantly more water available in the soil under the three-way cover crop than under the rye cover crop (Figure 4, right). The no-cover plots were intermediate in soil moisture. These data are based on averages for plots growing corn and soybeans, for the sandy soil and the silty clay soils, and measurements at 15 and 30 cm deep, since none of these factors had a statistical effect on the soil moisture. 

Slug Activity, Crop Damage, and Yields as Affected by Cover Crop Management

Study on Ridgely fields of James Lewis' Farm.

Monitoring of soil water, surface temperature, soil temperature, and slug numbers under shingle traps before the imposition of cover crop termination treatments and planting in the adjacent Ridgely fields revealed that the south field was significantly wetter than the north field. Slug numbers were weakly associated with soil moisture. Once the crops emerged, we monitored conditions and crop stand density on three dates for each field. The average data for the north field planted to corn are presented in Figure 5, below. The late-killed strips were not sprayed out until three weeks after corn was 'planted green' into them. They had slightly drier soil in the upper 5 cm, probably due to continued water transpiration by the cover crop during the very dry period in most of April. The corn stand establishment and stand density averaged over the three observation dates were not affected by the cover crop termination treatments. Slug damage to crop seedlings was scored on three dates in each field. Seedling damage by slugs was scored from 0 (no damage visible) to 4 (damage so severe that the seedling died or will die, such as eating off the terminal growing point on soybean). The figure below shows the damage scores for corn on three dates. On all three observation days, there was significantly less slug damage in the late-killed (planted green) strips than in the conventional early-killed cover crop strips. However, the corn seedling dry weight per plant was significantly lower in the late termination strips. 

Unlike for corn in the adjacent field, later termination and planting green did not affect soybean stand population or slug damage at Lewis'' farm near Ridgley (Figure 6).

Figure 7. Effect of later termination and planting green on slug damage scores for soybean seedlings in Lewis farm field near Ridgely, MD on three dates.

Figure 8 shows the number of slugs counted per shingle in the corn and soybean fields at Ridgely as averaged on five observation dates. The last two dates were after cash crop planting. These data are averaged across the termination timing strips.  The slug infestation was only moderate and the cover crop termination treatment did not affect the number of slugs observed in either the corn or soybean fields. The Figure also shows the surface and soil (5 cm) temperature and volumetric moisture content on the same dates. slug damage score for corn and soybean seedlings averaged over three dates. In this trial, letting the cover crop grow until cash crop planting by planting green into the living cover crop did not affect slug numbers or crop stand density achieved, but did reduce slug damage to both corn and soybean seedlings. The much heavier cover crop biomass residue on the late-killed strips could enhance yields when the summer is dry and water conservation is a factor.

Figure 8. Slug study data for Lewis Farm fields at Ridgley, MD showing soil and surface temperature, soil moisture, and slug counts for the north field in soybean residue where corn was planted in 2024 and (right) south field in corn residue where soybean was planted in 2024.

Because the corn was clearly stunted, we sampled the V5-stage corn plants and determined their dry weight per plant and the nutrient content of the plant tissue.  The tissue concentrations of most nutrients were not affected by the cover crop management, except higher levels of Zn, B, and N in the corn planted green into late terminated cover crop hairy vetch, most likely due to less dilution as in the bigger plants in the early terminated plots.  The K levels are also shown because they were unusually high, probably because the field had been regularly fertilized with poultry manure, which is very high in available K. 

In any case, the early season stunting carried through to the final yield. Corn yields were quite good in this irrigated field, but the late terminated strips yielded 71 bushels per acre lower (172 bu/acre) than the early termination strips (243 bu/acre, Figure below). Although 2024 was a very dry year during the growing season, these fields had the benefit of center pivot irrigation and water was not limiting. We propose that the stunting was not due to effects of nutrients, slugs, soil temperature, or water.   We hypothesize that the timing of the hairy vetch termination with herbicides may have produced active allelopathic compounds just when the corn seeds were emerging. Soybeans in the adjacent field, in contrast to the corn, yielded well regardless of the timing of the winter wheat cover crop termination and planting green.  Although he did not use row cleaners or residue rollers as suggested by Penn State University researchers (Reed, H. K., Karsten, H. D., Duiker, S., Tooker, J., and Curran, W. S. (2019). Planting green 101: Penn State research summary Penn State University. https://extension.psu.edu/planting-green-101-penn-state-research-summary ), Jim Lewis said that he has had poor results in the past with planting green for corn, but this 71-bushel loss was the last straw. He liked the results with the soybeans and plans to continue to experiment with planting green to improve his soil health when growing soybeans, but won't again try planting green for corn. 

Study on the silty field (7E) at CMREC Beltsville Facility.

Of the two research fields at CMREC used in this project, Field 7E has fine-textured Christiana soil and a history of slug infestations.  In the spring of 2024, we monitored the slugs and their impact on corn and soybean crops in this field in plots with rye, rye+clover, or winter weeds only. The average damage score increased over time during the first 10 days after emergence for both crops, but the damage from slugs was consistently worse on the soybeans than on the corn (Graph below).

Slug damage to both corn and soybean seedlings was generally not made worse by allowing cover crops to grow longer. In fact, on some evaluation dates damage scores were lower for the late killed and planted green plots. These results are consistent with what we observed in the on-farm trials on the Eastern shore. In the graph below on May 20th 12 Days after planting corn and soybeans the slug damage to crop seedlings was significantly less severe in the late cover crop termination plots than in the nearly cover crop termination plots on average across all three cover crop treatments and both cash crops. 

In addition to concerns about increasing slug damage, farmers are often concerned about longer-growing cover crops using up storage soil moisture. When making the slug counts and damage assessments we also measured soil moisture in the upper 5 cm of soil. Generally, the termination date did not affect this significantly and there was only a slight trend toward slightly drier soil in the plots with the longer-growing cover crop as shown in the graph below.  Obviously, a much-needed rain was received between the May 13th observation and the May 20th observation date. 

Cover crops did have a major effect on soil moisture conservation during the dry hot summer months. The spring and summer of 2024 were exceptionally dry on these non-irrigated plots. By June 22nd, a trend was beginning to emerge of greater moisture content (lower soil water tension) with the multi-species cover crop as compared to the rye, with the no-cover (winter weeds only) plots falling in between. With almost no rain received, by July 11 this trend was accentuated and was statistically significant. The graph below shows the soil water tension averaged across two soil depths (15 and 30 cm), two soils (a loamy sand field and a silty clay field ), and two crops (corn and soybeans). The reason why the rye cover crop resulted in dryer soil in the summer is uncertain but may have to do with the larger biomass and water uptake during its growing season, especially in April. 

Cover crop effects carried through to final crop yields. As we have seen on other sites in dry growing seasons, corn yields with cover crops were significantly (p< 0.07) higher than in the no-cover control plots. Nonetheless, in this extremely dry year, non-irrigated corn yields were very low across the board, especially on the sandy soil. 

Averaged across the two contrasting soils at the Beltsville research farm, soybean yields were not significantly affected by cover crops and were not as extremely low as corn yields. However, there was an interaction between cover crop species and termination timing such that the early and late kill of the multi-species mix resulted in significantly lower soybean yields than the mid-kill which was planted green and terminated on the same day. This effect, we believe, was largely related to the weediness of the early and late terminated plots due to the dry weather at the time of herbicide spraying and the lack of moisture to ensure the herbicides were absorbed by the weeds. Therefore, summer weeds grew out of control and there was a significant negative relationship between weed biomass and soybean yield in these plots (data not shown).

Accomplishments during 2025, the project's second field season.

During 2025, we collaborated with three farmers and conducted experiments at two research station sites. The main objectives were to investigate the relationship between cover crop management and slug damage to crop seedlings, and to study the performance of cover crops when interseeded into standing canopies in fall and terminated at various times in relation to cash crop planting in spring.

Investigations of slug damage and crop stand establishment.

In 2025, we collected three site-years of data on slug damage and stand establishment in relation to spring cover crop management. These three site-years, in combination with the previous seven site-years provides a total of 10 site-years of data and allow us to draw conclusions on which farmers can base their decisions with considerable confidence. The three farmers we collaborated with in 2025 were Steve Kraszewski of Ruthsburg, Maryland, Steve Groff of Holtwood, Pennsylvania, and Jim Lewis of Ridgely, Maryland. All three also participated in the project Farmer Advisory Board.

Steve Kraszewski of Mason Heritage Farms proposed a comparison of planting soybeans green into rye to them planting green into a mixture of rye and crimson clover. Although we discussed having replicated strips, in the end the two cover crops were sown in two different fields. We visited these fields to collect initial slug and seedling damage data on May 1st, but the soybean seeds had not yet emerged, so no useful data could be collected on stands or damage. We also placed shingles overnight, but observed no slugs the next day. Conditions were very dry, and soils were quite sandy. We did observe a significant difference in moisture content and texture between the two fields. We concluded that there was no scientific comparison that we could make because the inherent differences between the two fields were confounded with the two cover crop treatments.

Steve Groff, at Cedar Meadow Farm in Pennsylvania, wanted to try an innovative use of cover crops for his industrial hemp cash crop. He set up four replicated pairs of test strips. In one treatment, he drilled the hemp seed by itself in the usual manner at a rate of 800,000 seeds per acre. In the other treatment, he mixed the 800,000 seeds per acre with cover crop seeds, mainly brassica species. The hemp and cover crop seeds were drilled together in the same rows on the same date as the pure hemp seeding. We visited the site on May 1st to set out slug trap shingles, and again on May 2nd to count slugs and score damage. Conditions were rather dry, and we found no slugs under the shingles. The hemp seedlings and brassica cover crop seedlings were just emerging. We were able to make stand counts, but recognized that these would be incomplete because the cover crop and hemp had not completely emerged. We returned to collect additional data on May 21st, which was a rainy day. Although the cool and wet conditions were good for finding slugs, we found only a few. However, we did get useful assessments of hemp seedling damage and hemp stand counts, as well as data on the temperature and moisture conditions in the upper 2 inches of soil.

Although planting the mixed hemp - cover crop seed did not affect soil moisture, which was uniformly high, it did affect the temperature data. The land surface temperature measured with an infrared sensor was several degrees Fahrenheit warmer in the hemp cover crop mixture than in the sole hemp strips. In contrast, the 0-2 inch soil temperature was slightly but significantly warmer in the sole hemp compared to the hemp-cover crop mixture. We found significantly more damage to the sole hemp than to the hemp planted with a cover crop. We also found that the stand count for the hemp planted with a cover crop was about 60,000 plants greater per acre (474,418) than where the same seeding rate was used in the sole hemp (414,389). When we showed this data to Steve, he replied,

"We planted 800,000 seeds per acre, but a 50% stand establishment is fairly normal for him. This is interesting info - thanks for doing this!"

Steve ended up getting a bumper harvest of industrial hemp, but he said it would be logistically impossible to harvest the hemp separately from the research strips, so we do not have comparative yield data.

Jim Lewis established three replicated pairs of strips to make a comparison that was of interest to him. Jim chose a field that had been seeded by airplane in the fall with a mixture of wheat, triticale, and radish. This field has sandy coastal plain soils, but is equipped with a center pivot irrigation system to supplement the growing-season as needed. Jim compared two treatments: a mid-kill treatment that sprayed glyphosate to burn down the cover crop on April 22, eight days before planting soybeans. He compared this to three replicated strips of cover crop he left unsprayed and planned to plant green. We visited the farm on May 1st to lay out slug shingles and on May 2nd to count slugs and assess the soybean stand and damage. There was a small amount of slug damage on the cotyledons of the newly emerging seedlings. However, as the soybean seeds were still emerging, the stand counts were not conclusive. We returned to the site on May 21, after a rainy day, and found a significant number of slugs as well as slug damage to the seedlings. On this date, the cover crop in the mid-kill strips was completely dead and provided little groundcover, but the late-kill strips still had green, living triticale and wheat beginning to head out (See Figure 19). We measured a considerable difference in temperature with the still green late kill cover crop strips being 4° F warmer in the upper 2 inches of soil and about 2° F warmer on the surface. The mid-kill strips with the dead cover crop had wetter 0-2 inch soil with 22% water by volume, as compared to 18% in the still-green late-kill strips. There was no difference in soybean stand between the two treatments, with both having about 93,000 plants per acre. There was also no difference in the average soybean damage score, which ranged from 1 to 3 on a scale of 1 to 5. When we reported this data to Jim, he replied,

"I'm thinking about not spraying the planted-green beans. It's only going to be three weeks, and I'll put herbicide on the entire field anyway. The (cover crop in) the green-planted beans isn't that thick. The glyphosate application is around $15 per acre. That's worth 1.5 bushels of beans. If both treatments yield the same, I'm $15 ahead on the planted-green. What are your thoughts?"

After this discussion, Jim decided not to spray those strips to see how the soybeans would perform when the wheat - triticale cover crop was allowed mature and senesce. When Jim combine-harvested the soybeans in fall and sent me the weights, the two treatments were almost identical in yield: 80.2 bushels/acre for the mid-kill sprayed cover crop treatment and 81.3 bushels/acre for the ‘late-kill’ planted-green unsprayed cover crop treatment. So, Jim was right, he would make more money from the planted green unsprayed cover crop. To be clear, this would work where a cereal cover crop stand was not too dense and had reached boot stage while the soybeans were still young (V1-2), so that the cereal would mature and senesce soon enough to cause minimal competition. Figure 19 includes a photo to show what the cover crop looked like when we assessed soybean damage on May 21st.

On a slug -infested, somewhat poorly drained soil at the University of Maryland research farm near Beltsville, Maryland, we assessed slug damage as affected by three cover crop treatments and three cover crop termination times. The environmental conditions and slug damage were assessed on May 21, the same rainy day that others on the project team were assessing slug damage on the collaborating farms. Soil conditions were very wet with volumetric water contents averaging 38% in no-cover plots and 42% in the cover-cropped plots. Soybeans in this field were planted green on May 3. Overall, there was no effect of cover crop early, mid, or late termination timing on the soybean slug damage scores, which averaged about 2.5 on a 1 - 5 scale (Figure 20). There was a cover crop effect only for the early termination plots, where the damage score was slightly, but significantly higher for the mix cover crop (rye + clover + radish) than for the no-cover control. No other cover crop or termination timing treatment combinations had slug damage scores that differed from the controls.

Overall, in 2025, our project added three site-years of slug damage data to the seven site-years previously collected during and just before this project funding began. Across all 10 cumulative site years, we consistently found no evidence that allowing cover crops to grow later and planting green into them worsened slug damage to crop seedlings. In fact, in some cases, the plots with late-terminated cover crops had significantly less slug damage. There were also very few instances where cover crops were associated with more slug damage than control plots without cover crops. However, this is not to say that a no-till environment does not favor slugs. All of our plots, except those on Steve Kraszewski's farm, which is organic, were in long-term no-till. No-till soils are permanently covered with decaying plant residues that provide habitat for a wide range of faunal species, including many beneficials. However, they also provide habitat for slugs, which can be damaging to crops. There is no question that a bare, tilled soil is less hospitable to slugs, as well as to earthworms and other beneficial fauna. However, our data show that in no-till systems, there is little risk of greater slug damage being caused by allowing cover crops to grow up to or even beyond the cash crop planting date.

Cover crop performance in 2025.

As can be seen from the precipitation and temperature data in Figure 21, the 2024-2025 cover crop season was much drier than the previous year's, but temperature patterns were quite consistent from year to year. The figure shows that the cumulative growing degree days (GDDs, based on a minimum temperature for cover crop growth of 40 °C) in 2024-2025 were almost identical to those of the previous year. The arrows in the figure indicate the timing of such operations as the interseeding of cover crops into standing crop canopies before harvest in fall, and the planting of cash crops into standing cover crops before the mid- and late-termination in spring. In both 2023 and 2024, conditions in the fall were quite dry with almost no rainfall in October. These dry fall conditions affected cover crop establishment from broadcast interseeding and limited their fall growth. Nonetheless, the graphs in Figure 21 also show that approximately 500 GDDs were gained by planting the cover crop seeds into standing crop canopies and not waiting to drill them after crop harvest. An additional 350 to 450 growing degree days were gained by delaying spring termination until planting or a week or so after planting.

In spring 2025, we measured the above-ground cover crop productivity by collecting 50 x 50 cm (0.25 m²) quadrats of shoot tissue clipped 1 cm above the soil surface. We collected these samples within one or two days of the early, mid, and late cover crop terminations. We also sampled the volunteer weeds growing in the no-cover crop controls. We are interested in measuring the cover crop dry matter because the amount of plant dry matter produced is expected to directly impact many of the benefits provided by cover crops, whether the recycling of nutrients, protection from erosion, conservation of water, or enhancement of soil organic matter and aggregation.

Spring 2025 cover crop biomass was relatively low due to a slow start and uneven stands caused by the dry conditions in the fall of 2024. The highest biomass was produced by the mixed species cover crop in the sandy field (Field 39a, upper left panel in Figure 5 ). For that cover crop, which was dominated in spring by crimson clover, biomass increased about 1000 kg per hectare between each of the termination times, with the late-terminated cover crop averaging slightly more than 3,000 kg per hectare of aboveground dry matter. The rye cover crop in the same field followed a similar pattern, with the early and mid-terminations, but leveled off and did not increase between the mid and late-terminations. Weed biomass in the control plots did not change significantly between early and late termination timing and was approximately 500 kg per hectare in the sandy field. In the silt loam field (7e, lower three panels in Figure 22), the cover crop biomass was very similar to that in the sandy field, except that the mixed species dominated by clover did not increase between the mid and late-terminations. By the time corn and soybeans were ‘planted green’ into the cover crops, the biomass in both types of cover crops was approximately 2,000 kg per hectare.

Cover crop root dry matter and root-to-shoot ratios in 2025.

Only a fraction of published cover crop studies report the above-ground biomass produced by cover crops, despite the importance of biomass data for interpreting the effectiveness of cover cropping practices. A much smaller fraction of cover crop studies report the below-ground dry matter produced by cover crop roots. The root dry matter is important to understand because roots are about twice as effective as shoots in contributing to stabilized soil organic matter, and roots are responsible for directly feeding the soil biology through sloughed-off cells and root exudates. The chemical composition of roots is also expected to be quite different from that of the shoots, with roots being characterized by organic compounds such as suberin and by dramatically different concentrations and ratios of nutrient elements. In most plant growth and carbon sequestration models, the root biomass is estimated from measurements or modeling of shoot biomass using a set root-to-shoot ratios for a particular type of cover crop. One of the objectives of this project was to determine whether there was a root-to-shoot ratio characteristic of a particular cover crop or whether these ratios varied with conditions and stage of growth.

We used two different methods of sampling cover crop roots and calculating the root-to-shoot ratios (see Figure 23). In spring 2024, we sampled cover crop roots and shoots using the soil cube method, by which a cube of soil is excavated to a depth of 18 cm using a flat square spade. In the spring of 2025, we repurposed a large coring device originally designed to cut holes for golf course cups. This device obtained a core that was 20.3 cm deep and 10.2 cm in diameter. We obtained four such cores per sample, each core centered over an individual plant or clump of plants of either crimson clover or rye. One cube or four cores of soil and roots were soaked in soapy water for 24 hours to loosen the soil from the roots, and then washed under high pressure spray of water using screens and collecting the wash water in buckets. After washing, the roots and associated shoots from each sample were dried 65° C to a constant weight, and the dry weights were recorded. The dry root and shoot samples were ground through a 1 mm mesh for further analysis. The labor-intensive nature of this task probably accounts for why the literature has far less information about the below-ground parts of cover crops than the above-ground parts.

The spring 2025 dry matter data for crimson clover and rye roots are presented for each termination time in Figure 24. Note that the dry matter, expressed as kilograms per hectare, is much larger than the shoot dry matter displayed in Figure 22, where the shoot dry matter data were obtained from quadrant samples, which represented the uneven cover crop stands. The data in Figure 24 are based on soil cores centered over individual cover crop plants and therefore do not account for the plant density and uniformity. The shoot-to-root ratio data shown in Figure 24 were calculated from the shoot dry matter of the individual plants over which the soil core was centered. It is also important to note that the root dry matter at different stages of growth always represents roots from the same core volume, even though the plant's root system may have expanded to permeate a larger soil volume. Nonetheless, there were clear trends towards greater root biomass for the rye than for the clover, and less root biomass for both species as the cover crop plants matured.

The dried, ground tissue of roots and shoots associated with the core samples was analyzed for total carbon (C) and total nitrogen (N) content by high temperature combustion/gas chromatography at Penn State University Agricultural Services Lab. From these values, the C/N ratios were calculated. The tissue C/N ratio is a good predictor of how fast cover crop residues will decay after termination and whether they are likely to release or immobilize nitrogen in the soil. Residues with C/N ratios less than 20/1 reliably release nitrogen for crop plants to use, while residues with C/N ratios greater than 30/1 can be expected to immobilize nitrogen as decomposing microbes compete with crop plants for this element. Figure 25 illustrates how the C/N ratio varied between crimson clover and rye, between roots and shoots, and over time as the plants matured from early to late termination dates. The crimson clover shoot tissues had the lowest C/N ratios, increasing slightly from 13/1 at the early termination time to 16/1 and 17/1 at the mid- and late-termination dates. Interestingly, C/N ratios were considerably higher in clover roots, increasing significantly from 19/1 at early termination to just over 30/1 at the late termination. The rye shoot tissues had higher C/N ratios than the clover roots at all dates, starting near 30/1 at the early termination and increasing to 35 at the mid- and late-terminations. Most interestingly, roots of the rye plants had still higher C/N ratios, which were 35/1 for the immature plants in the early termination, but increased significantly to 44/1 by the mid- and 58/1 by the late-termination time. These root and shoot C/N ratios are similar to those reported in other studies, though the literature has very limited data on the C/N ratios of cover crop root tissue at different stages of growth.

The much higher C/N ratios of the rye roots compared to shoots may help explain why rye cover crops are notorious for tying up nitrogen and reducing corn yields unless extra nitrogen is applied. The relatively high C/N ratios of even the clover roots may help explain why the amount of nitrogen that subsequent corn crops typically receive from legume cover crops is often quite disappointing relative to the total amount of nitrogen measured in legume cover crop shoots. Terminating legume cover crops by non-chemical means, such as roller crimping, requires that they be in the full flowering stage. It seems possible that the high C/N root tissue of even the legume cover crops, when allowed to grow to the flowering stage, may immobilize enough nitrogen to make a difference.

Cover crop management effects on corn and soybean yields at CMREC, Beltsville.

In 2025, the Beltsville location saw ‘normal’ growing conditions for corn and soybeans. Corn yields were good for rainfed land and were significantly higher on the sandy field than on the silty field, probably due to greater weed and slug pressure at the silty field (see slug data above).
Across the two fields, corn stand density was significantly lower but still acceptable where rye was used as a cover crop compared to where there was no cover crop or the clover-rye-radish mixture was used (Figure 26, lower left). We measured corn grain yields using two methods: by hand-harvest of two 20-ft rows in the center of each termination date subplot, and by calibrated yield monitor during the combine machine harvest of the entire area of each termination date subplot. The results were similar and reasonably well correlated. The hand harvest values were slightly higher because the hand harvest avoided obvious anomalies like skips in the planting stand or groundhog mounds. The combine harvest yield in the sandy field was 153 bushels/acre in the mixed species cover crop plots, which was significantly higher than the 135 bushels/acre recorded for the no cover plots. At the silty field, yields were slightly lower, and there were no cover crop effects on the combine harvest yield data. At the sandy field, the hand harvest gave a similar pattern with higher yields in the mixed cover crop plots than in the no cover controls. However, in the silty field, the rye cover crop plots had significantly lower hand-harvested corn yields than the mixed species cover-cropped plots. The higher yields in the mixed species cover crop plots were probably due to a combination of somewhat better stands achieved by planting into the living clover and the advantage of some nitrogen provided by that legume cover crop. By either method of measuring corn yields, we detected no significant differences among the termination treatments in 2025.

Soybean yields were also measured by both hand-harvest and a calibrated combine yield monitor. However, the termination subplots were not long enough to get reliable yield monitor data, so only yields for the cover crop species main plot were measured by the combine. Soybean yields in the smaller termination date subplots were measured by hand harvest, but no termination date effects were observed. As with the corn, soybean stands were slightly but significantly lower in the rye cover crop plots compared to the no cover control plots. The yield monitor data showed significantly higher soybean yields in the mixed cover crop plots on the sandy field compared to the no cover plots on that field. On the silty field, soybean yields were significantly lower in the rye cover crop plots than in the no cover controls or the mixed species cover crop. The hand-harvest soybean yield data did not show any significant effects of either cover crop species, termination date, or soil texture.

Planting green in general did not negatively affect crop stands or crop yields. Cover crops, compared to no cover crops, more often increased crop yields than decreased them. Results from both the research station fields and the collaborating farms should help new adopters gain confidence that well-managed cover crops will not be a drag on their productivity or profitability.

Cover crop impacts on soil health
In fall of 2025 we sampled soil from the mid-termination plots of all three cover crop treatments at both the sandy and silty fields. The samples will be used in 2026 to determine the range of soil health parameters as affected by the previous 5 years of cover crop treatments.