Final report for GS22-262
Project Information
Florida agroecosystems are mostly established on coarse-textured soils and subject to a hot and humid climate that favors rapid soil organic matter decomposition, water percolation, and nutrient leaching. Cover crops can replace fallow periods and help protect and improve soil conditions, but the adoption of cover crops is limited because they are usually not profitable. Chickpea is a short-season, cool-season legume crop that could be used as an alternative as it fits the rotation gap. Moreover, chickpea can increase soil cover and increase crop residues while being more economically attractive as it can be harvested, serving as a dual-purpose (i.e., cash and cover) crop. Replacing a bare fallow with chickpea could also benefit soil biology, as the greater amount of crop residues returned to the system should stimulate microbially-driven carbon and nitrogen cycling. Greater microbial activity is critical to soil health because microbes decompose residues, stimulate plant growth, and increase nutrient cycling and availability. As the microbial community is responsive to management practices and provides information on microbially-driven nutrient cycling, microbial indicators could help assess agroecosystem sustainability and predict possible crop benefits, such as improved yield and/or nutrient use efficiency. This project aims to assess how replacing winter fallow periods with chickpea in row crop agroecosystems of the Southeast US affects soil microbes. More specifically, our goal is to quantify how chickpea, other cover crops and fallow affect soil carbon mineralization, enzyme activity, and gene expression, to determine the impacts of different winter crops on agroecosystem sustainability.
Due to limitations in obtaining sufficient chickpea varieties, the original objectives were slightly adjusted to focus on a single commercial chickpea variety, two cover crops, and a fallow as winter treatments. The objectives were also expanded to assess corn performance under both recommended and limited N fertilizer rates. At project completion, the revised objectives were as follows:
- Evaluate how replacing a winter bare fallow with a legume cash crop will impact microbial activity, by comparing a novel legume cash crop for the area (chickpea) to common cover cropping options of rye (non-legume) and clover (legume). Our goal was to distinguish the effects of replacing fallow periods with a cash crop, with grains harvested and removed from the system (e.g., chickpea), from the effects of growing traditional cover crops where all the biomass is returned to the soil at termination (e.g., clover and rye).
- Quantify how a commercial chickpea variety, two cover crops (rye and clover), and fallow affect different soil microbial indicators: soil C mineralization, enzyme activity, and gene expression.
- Evaluate whether and how different winter management strategies (fallow, chickpea as a cash crop, rye cover crop, and clover cover crop) influence the productivity of the subsequent corn crop, and whether these effects vary under different nitrogen fertilizer rates (low versus recommended). We also aimed to determine if and how microbial indicators were correlated with crop growth and nutrition for all crops studied.
Research
At project completion (August 2025)
Compared to the original plan, the experiment was conducted in both years using a single commercial chickpea variety, rather than three, due to limited seed availability of the best-performing varieties. The experimental design was also modified to include two N management levels for corn: a low rate of 18 kg N ha-1 (starter fertilizer only) and a high rate of 270 kg N ha-1, which corresponds to the current UF/IFAS recommendation. The methods for assessing short-term carbon mineralization (Cmin) and enzyme activity were also adjusted. For Cmin, we adopted a simpler method, while enzyme activity assays were replaced with a more sensitive method suitable for the sandy soil conditions of Florida. Due to a small sample size, analysis of variance (ANOVA) and principal component analysis (PCA) were used instead of factor analysis. At project completion, the revised materials and methods were as follows:
Experimental design
The experiment was conducted over two years (2022–2024) at the University of Florida's Plant Science Research and Education Unit (PSREU) in Citra, FL. Four winter crop treatments were evaluated: a commercial chickpea variety, rye cover crop, clover cover crop, and a chemically maintained fallow. The trial was set up as a randomized complete block design with split-plot restrictions in randomization and four replications, for a total of 16 main plots and 32 split-plots. Chickpea and cover crops were set as the main plot during winter. Main plots were split into two plots receiving N rates of 18 kg ha-1(i.e., only the starter fertilizer) and 270 kg ha-1 (i.e., current UF/IFAS recommendation) during the corn season (split in three applications) to evaluate if N rates affect microbial dynamics observed under different winter crops. Rye was selected because it's a common winter cover crop option used in North Florida. Clover was used to compare chickpea to a winter legume cover crop, for which the biomass is incorporated into the soil rather than partially harvested (grains) like chickpea.
The winter crops were planted in 30 x 15 ft plots in December 2022. All crops received a starter fertilizer rate and were hand-weeded in February 2023. Irrigation was performed as needed with overhead irrigation. Because the rye cover crop was ready for termination sooner due to earlier flowering relative to legumes, this crop was terminated in mid-March 2023. Following chickpea harvest (early May in 2023), chickpea and clover were terminated, with the incorporation of post-harvest residues (chickpea) or the whole biomass (rye, clover) into the soil. After incorporation, a grain corn hybrid was planted in May 2023 in the same field, and plots were split into low and recommended N rates (15 x 15 ft plots), with the latter being applied at three split applications. Weed control was performed with glyphosate, and preventive pesticide applications and irrigation were performed as needed.
In November 2023, the winter crops were planted in the same experimental plots for the second season, aiming at an early termination and corn planting. However, chickpea had low germination, and seedlings were grown in a greenhouse and transplanted to the field in early January 2024. All crops received a starter fertilizer rate, chickpea and clover were hand-weeded in February 2024, and overhead irrigation was applied to all crops as needed. Rye flowered early again and was terminated in mid-February 2024. Reduced chickpea germination, growth delays due to transplanting, and maturity delays due to pest pressure resulted in chickpea being terminated as a cover crop (i.e., without harvesting the grains) together with clover in early-May 2024 to allow corn planting in the right window. Corn was planted after the incorporation of corn residues in early May 2024. During the 2024 corn season, southern corn leaf blight (Bipolaris maydis) was identified, and although fungicide was applied, most plants developed foliar lesions, leading to weakened stems. In mid-August, Hurricane Debby caused significant crop damage through lodging, broken tops, and ear losses, resulting in reduced yields compared to the 2023 season.
Chickpea, rye, clover, and corn aboveground biomass were sampled at termination/harvest using two random 0.5 m quadrats in each plot. Samples were dried at 65 ̊C and ground to determine nutrient concentration by digestion, followed by quantification using ICP-OES. Total C and N were determined by combustion. Chickpea and corn yields were also determined at harvest.
Soil samples were collected three times in 2022-23 and two times in 2023-24 per year (five times total): before chickpea and cover crop planting in 2022, at winter crop termination, and after corn harvest, to measure changes in soil microbial indicators through the rotation system. Ten cores per plot were collected from the top 20 cm of soil, and samples were split before processing: samples for gene expression were kept at -80 ̊C until processing, whereas samples for soil C mineralization and enzyme activity were air-dried and sieved before processing.
Short-term C mineralization
Readily-available soil C was measured with a 10 g sample of air-dried soil placed in a mason jar, then water was added to reach approximately 50% of the soil's field capacity (2.0 mL for our soil), the jars were sealed, and the CO2 concentrations were measured using an Infrared Gas Analyzer (IRGA) immediately after the addition of water (time 0) and after 24 hours of incubation. A standard curve was used with 2,000 to 10,000 ppm CO2, and the CO2 released in each jar was standardized by soil air-dried mass.
Enzyme assays
Due to minimal differences in N cycling genes and Cmin between N rates, and to reduce costs and sample processing time, enzyme activity was only assessed in the high N treatment. The activities of β-glucosidase (BG), N-acetyl-β-glucosaminidase (NAG), and acid phosphatase (AP) enzymes were determined using fluorescent substrates following a modified protocol from Bell et al. (2013).
N cycling gene expression
DNA was extracted from soils with Qiagen DNeasy PowerLyzer PowerSoil Kit, following the manufacturer's instructions. The bacterial, fungal, and archaeal populations were quantified using SYBR-based qPCR analysis. The relative abundances of selected functional genes quantified by qPCR were determined for N-fixation (dinitrogen reductase, nifH), nitrification (archaeal and bacterial ammonia monooxygenase, amoA), and denitrification (nitrous oxide reductase, nosZ).
Data analysis
Data were analyzed in R using mixed-model ANOVA to account for repeated measures across sampling events using the nlme package (Pinheiro et al., 2022). Least-squares means with Tukey's adjustment (α = 0.05) were used for multiple comparisons among winter crops and sampling events using the emmeans package (Lenth, 2021). Two principal component analyses (PCA) were also conducted using the FactoMineR (Lê et al., 2008) and factoextra (Kassambara and Mundt, 2020) packages in R. The first PCA analyzed total biomass (winter crops and weeds), the C:N ratio of these residues, and soil indicators from post-termination. The second PCA paired corn harvest soil indicators with corn yield and nutrient uptake, capturing variance more closely tied to cash crop performance.
References
Bell, C.W., Fricks, B.E., Rocca, J.D., Steinweg, J.M., McMahon, S.K., Wallenstein, M.D., 2013. High-throughput Fluorometric Measurement of Potential Soil Extracellular Enzyme Activities. Journal of Visualized Experiments. https://doi.org/10.3791/50961
Kassambara, A., Mundt, F., 2020. factoextra: Extract and Visualize the Results of Multivariate Data Analyses.
Lê, S., Josse, J., Husson, F., 2008. FactoMineR : An R Package for Multivariate Analysis. J Stat Softw 25. https://doi.org/10.18637/jss.v025.i01
Lenth, R. V., 2021. emmeans: Estimated Marginal Means, aka Least-Squares Means.
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., Heisterkamp, S., Van Willigen, B., Maintainer, R., 2022. Package ‘nlme’. Linear and nonlinear mixed effects models, version, 3(1).
At project completion (August 2025)
For the winter crop phase of 2022–2023, rye had the highest biomass among treatments, followed by clover and chickpea; rye also had less weed abundance than clover and chickpea. During the 2023–2024 season, clover produced the highest biomass, followed by rye and chickpea. Weed abundance was generally lower in 2023–2024, with minimal differences between treatments.
After winter crop termination in 2024, chickpea and clover had higher C mineralization than rye and fallow. Enzyme activity (BG and NAG) was generally highest in chickpea (and clover for one sampling event) but varied through time. AP activity did not differ between treatments.
Winter crop treatments generally had higher expression of N cycling genes relative to fallow, especially legumes. When significant (i.e., only at corn harvest), the low N rate showed higher expression of amoA-AOA, amoA-AOB, and nosZ compared to the high N rate.
In contrast to the soil responses, winter crop treatments did not significantly increase corn yield, grain N concentration, or whole-plant N accumulation compared to fallow. Corn variables were negatively correlated with C mineralization and enzyme activity, suggesting that these soil indicators do not explain corn responses. The limited corn response indicates that while winter crops may not directly boost yields through increased N availability, they may contribute to broader system sustainability by improving nutrient retention, soil health, and microbial activity.
Educational & Outreach Activities
Participation Summary:
In March 2024, we collaborated with the LS21-353 project ("Evaluating the Dual- Purpose of Chickpea: A Cash and Cover Crop for Agricultural Production Systems in the Southeast") by hosting and facilitating a grower assessment in our experimental trial, which involved both growers and scientists. Participants evaluated chickpea and cover crops based on criteria such as biomass production, canopy closure, and weed pressure in the field.
In July 2024, we collaborated once more with the LS21-353 project and facilitated another grower assessment ("Evaluating the Dual-Purpose of Chickpea: Effects on subsequent corn production"). For this assessment, participants evaluated corn production based on visual characteristics such as plant growth, ear size, and overall crop development.
After both field assessments, participants discussed their impressions about winter crop effects and addressed concerns around rotational fit and market viability. They identified the treatments that sounded more promising and provided insights on what they would like to see in terms of future research.
In addition, findings from this project are part of one submitted research article that is currently under review, and another peer-review article that is in preparation.
Project Outcomes
At Project Completion (August 2025)
Microbial activity was higher with winter crops, especially with chickpea and clover, suggesting that they can be a viable strategy to replace fallow periods and improve soil health. Increases in C mineralization, enzyme activity, and the expression of N cycling genes indicate that mineralization, and therefore N availability, may be greater when winter crops are used. Although we did not find significant effects of winter crop treatments on corn yield, numerical trends suggest that benefits from higher microbial activity may matter more when N fertilizer inputs are low. There was a correlation between winter crop biomass and C:N ratio and soil health indicators at termination, but these indicators were not correlated with corn yield, suggesting that higher microbial activity was not the primary driver of yield responses in this system. Therefore, adoption of chickpea may be more likely if the crop can be effectively managed to fit the winter gap of Florida corn systems, or if the benefits of winter crops on yield or nutrient management become clearer over the long term.
At Project Completion (August 2025)
As chickpea is a novel crop to the Southeastern US, some insights about chickpea management are still needed to guarantee a good agronomic fit for the crop, such as good weed management, a variety that fits local corn planting windows, and the need for longer rotations to avoid disease problems.
C mineralization seems to be a sensitive soil health indicator in Florida's sandy soils. C mineralization was also simpler and more straightforward to measure than enzymes and gene expression, confirming its potential as a practical option for soil health assessment, particularly when a standard control management is included (e.g., fallow).
Nonetheless, using a suite of indicators can provide a more complete picture of soil health responses to winter crops because each indicator can capture a different component of soil nutrient cycles. C mineralization reflects general microbial activity and mineralization, enzyme activity helps identify which nutrient cycles (C, N, and P, in our case) are being affected by residue inputs, while gene expression informs how N-cycling microbial communities respond to those inputs.
At Project Completion (August 2025)
Our experiment was subject to challenging conditions, such as limited seed availability, the consequent use of chickpea transplants, and extreme weather events. Because of that, it is premature to recommend replacing a winter fallow with chickpea. Still, winter crops, especially rye and clover, can be recommended to return organic residues to the system and help support nutrient cycling. Effective management to obtain these benefits will vary with context, including the fine-tuning of agronomic management for chickpea. Once we obtain more robust evidence of chickpea’s ability to grow in Florida winters, we will be better equipped to make recommendations about its potential in Florida agroecosystems.