Agricultural & Environmental Letters最新文献

Accurate operational predictions of cereal rye (Secale cereale L.) biomass are critical for quantifying the agroecosystem services provided by cover crops and for guiding growers’ management decisions for subsequent cash crops. In this study, we developed machine learning-based biomass prediction models using two advanced gradient-boosted tree algorithms, CatBoost and XGBoost. A comprehensive dataset comprising cereal rye biomass and management information from 24 U.S. states, combined with soil and weather data, were used to train the models. Models relying solely on early spring weather inputs achieved moderate predictive skill (R² ≈ 0.74). Incorporating later-season weather data modestly improved mid-season fits but led to overfitting in late-spring predictions. Extending CatBoost to quantile regression enabled estimation of 10%–90% prediction intervals with moderate pinball loss. Overall, our findings demonstrate that publicly available soil and weather data, supplemented with limited management inputs, can support interpretable, uncertainty-aware biomass predictions suitable for optimal cover crop management.

Nematodes are important in soil food webs and are being considered in soil health frameworks. We evaluated nematode counts as a potentially rapid measure of land use effects on soil biology in a benchmark Mollisol. We assessed pedons (0–120 cm) from conventional, no-till, and grassland fields in eastern North Dakota in October 2022. Although bulk density was 25% higher at 5–10 cm and aggregate stability was 36% lower (0–5 and 5–10 cm) in conventional versus grassland fields, land use did not affect nematode counts above 20 cm (288–2498 100 g⁻¹ dry soil). Land use did mitigate nematode responses to soil properties. Nematodes more than doubled with increasing aggregate stability in no-till and grassland samples and were reduced by a third with increasing bulk density in grassland samples. Although nematode counts per se are not useful for assessing land use effects, soil properties can be used to predict nematode numbers in the series.

Agricultural systems are vulnerable to extreme weather, market volatility, and changing socio-cultural contexts. Despite efforts to create transformational solutions in agriculture to ensure economic, social, and environmental sustainability, there is often a disconnect between research findings and real-world experience. Co-production is a collaborative process that engages farmers, ranchers, and other community members as equals in research design and implementation, incorporates diverse knowledges, and includes community members as research decision-makers. While co-production requires more time, trust, and institutional support, it offers greater research impact and increased public support for science as a problem-solving tool. We share three case studies from our own research, and an introduction to the literature featuring best practices, to illustrate pathways for integrating co-production in research programs.

Many soil health indicators have been developed by researchers to aid farmer decision-making yet rarely incorporate farmer preferences in the presentation of that data. To fill this gap, we conducted focus groups with Midwestern row crop farmers to elicit the characteristics they want to translate data from soil health indicators into useable information that inform management decisions. Farmers were interested in the potential economic and conservation benefits of soil health, but current soil health report outputs are difficult to understand and put in practice. Farmers wanted clear management guidance that is tailored to their specific edaphic context. While they expressed ambivalence about specific indicators, they wanted to understand a full picture of their soil health. Moreover, they expressed interest in using the information to both affirm current management practices and adopt future practices. Findings suggest that improved alignment of current research questions with stakeholder needs can help harness the potential of soil health.

Manure applications can be a source of phosphorus (P) in runoff. This study evaluated whether RhizoSorb, a modified aluminum oxide soil amendment, could reduce water-extractable P (WEP) in the 0- to 7.5-cm soil depth after poultry litter applications. RhizoSorb was applied once at the beginning of the experiment at three rates followed by 3 years of poultry litter applied at three rates in split-plots of each RhizoSorb rate. Poultry litter applications increased the WEP of soil sampled in the fall of each year, with P concentration increasing as a function of litter application rate and years of application. RhizoSorb decreased WEP between 26% and 53% in the first and second year of the study. These results suggest that RhizoSorb may have potential to reduce soluble P losses from poultry litter applications, but further research to measure P losses from in situ runoff is necessary for confirmation.

Tropical weather events bring strong winds and rains to sugarcane (Saccharum spp. hybrids)-growing regions in the United States. Additionally, the peak of the Atlantic hurricane season coincides with the beginning of the processing season for the US sugarcane industries. In September 2024, Hurricane Francine made landfall in south Louisiana, resulting in acute impacts to the crop. Hurricane-affected cane samples were analyzed to quantify degradation and microbial activity. Marginal impacts on juice quality were observed immediately following the storm; however, these impacts did not persist across the season. Initial measurements showed a roughly 15% decrease in sugar content in storm-affected cane, with lower pH and elevated dextran content. Despite hurricane effects, the industry did not appear to have any major losses in productivity based on economic reporting and subsequent juice measurements taken at the end of the season. Regardless, the experimental techniques presented here for microbial degradation quantification represent potential opportunities for the industry to incorporate new modes of forecasting, monitoring, and mitigating losses attributable to microbial activity in damaged sugarcane.

Water management is becoming increasingly vital in the Southern Great Plains due to declining levels of the Ogallala Aquifer and the increasing drought conditions. As water scarcity intensifies, precision irrigation technologies, such as artificial intelligence (AI) with integrated ground-penetrating radar (AI-Radar), can be a promising solution for optimizing water use at the field scale in crop production systems. This on-farm trial evaluates the AI-Radar irrigation system compared to a conventional irrigation system (subsurface drip, SSDI) in central Kansas. Spatial analysis of the water deficit index (WDI) determined using high-resolution satellite imagery showed significantly reduced values of this index under AI-Radar–based irrigation (0.15–0.16) compared to SSDI (0.20–0.24; p < 0.07) during critical tasseling to dough (VT–R4) growth stages. Lower WDI indicates more efficient spatial water distribution with minimal crop water stress. Addtionally, the AI-Radar system applied 23.5%–25.1% less irrigation water than the SSDI system while maintaining lower crop water stress. Therefore, this case study demonstrates that adopting AI-Radar–based irrigation could support groundwater conservation and regulatory compliance in the Southern Great Plains region.

Quantifying soil organic carbon (SOC) and total soil nitrogen (TSN) at a cumulative depth of 0–30 cm under different land use and management scenarios remains a priority, particularly in the southeastern United States, where soil erosion can be intense with high rainfall and long history of intensive soil disturbance. Restoration of soil health is expected with conservation management, and sensitive indicators of change may be from SOC, TSN, and the active fraction of organic matter as soil-test biological activity (STBA). Cumulative frequency distributions were developed from data in published studies containing 1470 soil profiles that characterized SOC, TSN, and STBA under cropland, grassland, and woodland in the southeastern United States. Soil health targets were established at ratios of root-zone enrichment-to-baseline SOC and TSN of 2.0 ± 0.1 kg kg⁻¹; 4.7 ± 0.2 kg kg⁻¹ was the target for STBA. This database serves as a reference to begin classifying root-zone soil health conditions in the southeastern United States.