Background: Aroma is an important grain quality trait in rice, controlled by mutations within the BADH2 gene. The trait is simply inherited, and its importance in variety development makes it a practical target for marker-assisted selection in applied breeding programs. The predominant functional mutation within BADH2, an 8-bp indel, can be reliably detected using a PCR-based assay, but the available assays and associated genotyping platforms are insufficient for large-scale applied molecular breeding applications and are not compatible with outsourcing genotyping services. Results: We first characterized SNP diversity across the BADH2 gene in a collection of 2932 rice varieties to determine the number of gene haplotypes in O. sativa. Using 297 gene-based SNPs, 11 haplotype groups were detected, and subsequently identified a minimal set of nine informative SNPs that uniquely identified the BADH2 haplotypes. These nine SNPs were developed into KASP assays and used to examine a panel of 369 U.S. rice accessions. The panel represented modern breeding germplasm and included all known aroma pedigree sources in U.S. rice. Six haplotypes were detected within the U.S. panel, of which two were found in majority (85%) of varieties. A representative set of 39 varieties from all haplotype groups was evaluated phenotypically to distinguish aromatic from non-aromatic lines. Conclusion: One haplotype (Hap 6) was found to be perfectly associated with the aromatic phenotype. A single KASP SNP unique to Hap 6 was demonstrated to reliably differentiate aromatic from non-aromatic rice varieties across U.S. germplasm.

Nitrogen (N) fertilizer management in rice (Oryza sativa L.) varies with production practices. In drill‐seeded, delayed‐flood production systems, the most common production practice in Louisiana N fertilizer is applied at two application timings. The first application timing is just before the permanent flood is established. The second application is at midseason. Nitrogen fertilization before flooding is critical for maximum N uptake, nitrogen recovery efficiency (NRE), and yield. Field experiments were conducted from 2017 to 2020 to evaluate N timing effects on N uptake, NRE, and rice yield. The rice cultivar CL153 was drill‐seeded into a stale seedbed at a seed rate of 85 kg ha⁻¹. Fertilizer‐N was applied utilizing multiple application timings and rates adding up to a seasonal rate of 155 kg ha⁻¹ across treatments. A single N application 1‐day before flooding significantly increased grain yield in all trials, ranging from 8523 kg ha⁻¹ in 2019 to 11,322 kg ha⁻¹ in 2018. Compared to post‐flood applications, preflood N increased plant height, N uptake, and NRE. Split N application rates and timings after flooding did not impact rice yield or its agronomics, such as height, aboveground biomass, and time of heading. NRE and yield were significantly correlated (r = 0.805; p < 0.001). Our results indicated that a single N application before flooding has the potential to be an alternative option for N management in the drill‐seeded, delayed‐flood rice system.

Soybean [Glycine max (L.) Merr.] producers in Louisiana began shifting to an early soybean production system (ESPS) in the early 2000s by planting earlier maturing varieties in April and May. Although this shift in planting practices has been supported by research elsewhere beginning in the mid‐1990s, there is minimum data focusing on the ESPS across Louisiana. The overall objective of this research was to evaluate the effect of planting date, maturity group (MG), and their combination across a comprehensive set of yield environments in the state of Louisiana to determine the combination of optimum planting date and MG for soybean production. We gathered field data from the 2013 to 2020 seasons originated from three Louisiana State University AgCenter research stations and four seed companies. A total of 428, 926, and 331 observations were analyzed from the Northeast, Central, and Southwest Louisiana regions, respectively. When including all data from across Louisiana, the optimum planting date was April 30. Breaking by regions, the average optimum planting date for the Northeast Louisiana region was April 9. The Central and Southwest results were divided by MG section, and the approximate optimum planting dates were April 15 and May 15, respectively. These results support the ESPS for the Northeast and Central Louisiana regions, but not for the Southwest Louisiana region. Optimizing planting recommendation is a critical component for supporting the development of varieties suitable for the southern production systems of the United States.

Food engineering, food chemistry, and consumer segmentation were used to evaluate ready-to-eat rice. The aromatic Louisiana Clearfield Jazzman (CJ) and Thai Jasmine (TJ), and a non-aromatic parboiled (PB) rice were hydrated during the first 10 min of processing with reciprocal agitation followed by static retort processing. The aroma compound, 2-Acetyl-1-pyrroline (2-AP) was more heat-stable in CJ than TJ rice but decreased 15-fold compared to the rice cooker method. Pareto analysis indicated that rice type and agitation had the main effect on amylose and total starch and chroma and hue. Color differences of rice agitated during hydration and between rice cooker or static retort processed rice, indicated only slight differences for each rice variety. Hydration of dry rice during retort cooking and similar starch, color, and aroma quality were achieved with reciprocal compared to static or rice cooker methods. Survey responses categorized consumers into three, mindsets driven by rice consumption, convenience, or packaging.

Accurate in-season rice (Oryza sativa L.) yield prediction is crucial for improved nitrogen management and climate-smart decision making, yet rigorous comparative benchmarking of machine learning (ML) models using multi-temporal UAV spectral data with independent temporal validation remains limited. This study systematically evaluated four ML algorithms (Random Forest, XGBoost, Neural Network, and Linear Regression) and two Bayesian model averaging ensembles for rice yield prediction using UAV multispectral imagery. Field experiments spanning three growing seasons (2023–2025) at Louisiana State University comprised 9–10 varieties and six nitrogen rates (0–235 kg N ha⁻¹; 576 plots). Vegetation indices and spectral bands from three growth stages were extracted as predictors. Models were compared using 300 random train–test iterations with systematic hyperparameter optimization, followed by independent validation on 2025 data. Among the individual models, XGBoost achieved the highest internal accuracy (R² = 0.87, RMSE = 0.85 t ha⁻¹), substantially outperforming Linear Regression (R² = 0.66, RMSE = 1.32 t ha⁻¹), while reduced BMA reached R² = 0.89. XGBoost demonstrated robust temporal generalization (R² = 0.62, NRMSE = 8.47%) despite environmental variation. The Enhanced Vegetation Index and Normalized Difference Red Edge at 90 days after planting (reproductive stage) were the most stable predictors across 300 iterations. Tree-based ML models substantially outperform traditional linear approaches, providing reliable pre-harvest yield forecasting for operational precision rice production.