Arsenic (As) poisoning of groundwater in Bangladesh has become a major environmental and health issue. The extensive use of groundwater in irrigation of rice has resulted in elevated As in soils and crops. A study was undertaken to determine As concentrations in groundwater, soils, and crops in 16 districts of southwestern Bangladesh. Groundwater samples were collected from shallow-tube and hand-tube wells (STW and HTW) used for irrigation and drinking water. Soil and rice plants were sampled from the command area of the tube wells. Arsenic concentrations were determined using an atomic absorption spectrometer equipped with flow injection hydride generator. Groundwater samples contained <10 to 552 μg As L⁻¹. Arsenic concentrations in 59% of STW samples exceeded 50 μg As L⁻¹, the national standard for As in drinking water. Unlike groundwater, most of the surface water samples contained <10 μg As L⁻¹. Concentrations of As in the soils from the command area of the tube wells ranged from 4.5 to 68 mg kg⁻¹. More than 85% of the soils contained <20 mg As kg⁻¹. The mean As concentration in the rice grain samples was 0.23 mg kg⁻¹, which is much less than the maximum food hygiene standard. A positive relationship was observed between groundwater and soil As, implying that soil As level increases as a result of irrigation with contaminated water. However, irrigation water As did not show any relation with rice grain As. The findings suggest that surface water bodies are a safe source of irrigation water in the As-contaminated areas.

Rice production in coastal tidal floodplains is severely constrained by periodic tidal submergence, making modern semi-dwarf varieties unsuitable. While local landrace cultivars are widely grown for their adaptability, their adaptation mechanisms and response to nitrogen management under tidal ecosystems remain poorly understood. This two-year on-farm study (aman seasons, 2021 and 2022) evaluated landrace rice cultivars against a modern check (BR23) under different nitrogen-management practices: urea super-granule deep placement and prilled urea application. Landraces adapted better to tidal submergence than BR23, exhibiting greater plant height, culm strength, biomass, and nutrient accumulation. Cultivar effects explained most of the variation in grain yield (38–40%), followed by nitrogen management (18–21%). Deep placement of urea super-granule significantly enhanced growth, yield, and nutrient uptake compared to prilled urea. Shorna showed the highest grain nitrogen uptake (48.23 kg ha⁻¹), agronomic efficiency (20.81 kg grain kg⁻¹ nitrogen), and partial factor productivity (71.91 kg grain kg⁻¹ nitrogen). Bhushiara achieved the highest grain harvest index (0.54) and nutrient harvest indices, indicating efficient nutrient partitioning to grains. During grain-filling, nitrogen and phosphorus efficiently remobilized to the grains, while potassium remained mostly in the straw. Ultimately, pairing resilient landraces like Shorna, Bhushiara, Dudmona, and Lalmota with optimized nitrogen management, specifically urea super-granule application, substantially improves productivity, nutrient efficiency, and climate resilience. This study provides some of the first comprehensive evidence on how genotype-specific adaptation and deep-placed nitrogen interact to support sustainable rice production in Bangladesh’s coastal floodplains.