Large survey dataset of rice production practices applied by farmers on their largest farm plot during 2018 in India.

This dataset provides detailed information on rice production practices being applied by farmers during 2018 rainy season in India. Data was collected through computer-assisted personal interview of farmers using the digital platform Open Data Kit (ODK). The dataset, n = 8355, covers eight Indian states, viz., Andhra Pradesh, Bihar, Chhattisgarh, Haryana, Odisha, Punjab, Uttar Pradesh and West Bengal. Sampling frames were constructed separately for each district within states and farmers were selected randomly. The survey was deployed in 49 districts with a maximum of 210 interviews per district. The digital survey form was available on mobile phones of trained enumerators and was designed to minimize data entry errors. Each survey captured approximately 225 variables around rice production practices of farmers' largest plot starting with land preparation, establishment method, crop variety and planting time through to crop yield. Detailed modules captured fertilizer application, irrigation, weed management, biotic and abiotic stresses. Additional information was gathered on household demographics and marketing. Geo-points were recorded for each surveyed plot with an accuracy of <10 m. This dataset is generated to bridge a data-gap in the national system and generates information about the adoption of technologies, as well as enabling prediction and other analytics. It can potentially be the basis for evidence-based agriculture programming by policy makers.

Field-specific nutrient management using Rice Crop Manager decision support tool in Odisha, India.

The requirement of rice (Oryza sativa L.) for fertilizer can depend on crop and soil management practices, which can vary among fields within a rice-growing area. A web-based decision support tool named Rice Crop Manager (RCM) was developed previously to calculate field-specific rates of fertilizer N, P, and K for rice in Odisha State in eastern India. We compared field-specific nutrient management calculated by RCM with farmers' fertilizer practice (FFP) and a blanket fertilizer recommendation (BFR), which used a uniform 80 kg N ha-1, 17 kg P ha-1, and 33 kg K ha-1. A total of 209 field trials were conducted in two seasons (kharif and rabi) for two years across ten districts in six agro-climatic zones. Grain yield was consistently higher with fertilization recommended by RCM than with FFP. Higher yield with RCM was attributed to a combination of applying more of the total fertilizer N at the critical growth stage of panicle initiation, applying more fertilizer N in kharif, and applying zinc. The RCM recommendation frequently increased yield compared to BFR as a result of improved N management, which included the adjustment of N rate for a target yield set slightly higher than historical yield reported by a farmer. Fertilization based on RCM rather than BFR reduced the risk of financial loss. The effectiveness of an RCM recommendation relative to BFR and FFP was consistent across rice varieties with different growth duration, irrigated and rainfed rice, and three categories of soil clay content. The RCM recommendation failed to increase yield relative to BFR in one of the six agro-climatic zones, where a higher rate of fertilizer P and/or K was apparently required. The nutrient management calculations used by RCM can be improved as new information and research findings become available. Experiences with RCM in Odisha can help guide the development of comparable nutrient management decision tools in other rice-growing areas.