Agricultural Horizons: Research, Technology and Sustainable Development Vol. 6

Personalised Agricultural Extension: Harnessing Artificial Intelligence, Farmer Data and Context-Aware Advisory Systems

2026-10-28T11:26:17+00:00

Agricultural extension is increasingly expected to deliver advice that is not merely digital, but timely, farm-specific, agronomically defensible and responsive to farmers' changing constraints. This expectation has accelerated interest in artificial intelligence (AI), farmer-generated data and context-aware advisory systems. This critical narrative review evaluates whether these technologies are strengthening agricultural extension in ways that can be distinguished from the effects of digital delivery or personalisation alone. The review integrates evidence on mobile and site-specific advisory services, decision-support tools, machine learning, generative AI, retrieval-augmented generation, farmer data governance, digital inclusion and human-AI collaboration. The strongest causal evidence concerns personalisation and site-specific agronomic information rather than AI itself. Randomised trials show that tailored nutrient and management recommendations can improve adoption and, in some settings, yields, profits or resource-use efficiency, but other trials find behavioural change without measurable productivity gains. Evidence specific to generative AI remains much less mature and is dominated by expert-rating studies, developer-led evaluations, observational studies and conceptual architectures. Large language models can expand conversational access and synthesise heterogeneous knowledge, yet their most consequential errors occur precisely where extension requires local specificity, including planting dates, input rates and timing. Personalisation therefore depends less on model sophistication alone than on the quality, provenance and granularity of farmer, field, weather and agronomic data; explicit decision rules; accountable human oversight; and institutions that sustain trust. Data governance is not an ancillary ethical issue because consent, control, interoperability, representation and benefit sharing influence both data quality and farmer participation. The review concludes that AI should be treated as one layer within a socio-technical extension system rather than as a substitute for agronomy or extension professionals. Future evaluation should test the incremental value of AI against simpler personalised digital services, measure distributional as well as average effects, and combine agronomic safety benchmarks with longitudinal farm-level outcomes.

Estimation of Reference Evapotranspiration, Effective Rainfall and Crop Evapotranspiration Using CROPWAT 8.0 in Puttur, Andhra Pradesh, India

2026-10-28T11:30:07+00:00

Efficient management of irrigation water requires reliable estimation of evapotranspiration and effective rainfall. The present study evaluated reference evapotranspiration (ETo), effective rainfall and crop evapotranspiration (ETc) for Puttur, Andhra Pradesh, India, using the CROPWAT 8.0 model. Meteorological data including maximum and minimum temperature, relative humidity, wind speed, sunshine duration and rainfall for the period 2019–2023 were obtained from the Regional Agricultural Research Station (RARS), Tirupati. Reference evapotranspiration was estimated using the FAO Penman–Monteith method implemented in CROPWAT 8.0. Effective rainfall was estimated from monthly rainfall data, while crop evapotranspiration was calculated using the crop coefficient approach (ETc = Kc × ETo) for paddy, sugarcane, groundnut and tomato. The results showed a distinct seasonal pattern in reference evapotranspiration, with relatively high values during the warmer months and lower values during the later part of the year. The maximum monthly ETo values were 252.03 mm in May 2019, 252.96 mm in May 2020, 249.55 mm in June 2021, 242.42 mm in May 2022 and 234.36 mm in May 2023. The minimum ETo values were 146.10 mm in November 2019, 142.80 mm in November 2020, 140.70 mm in November 2021, 140.43 mm in January 2022 and 139.19 mm in December 2023. Effective rainfall was highest during October in all five years, with values of 132.2, 133.1, 133.6, 134.1 and 134.5 mm for 2019–2023, respectively. Among the crops evaluated, paddy exhibited the highest crop evapotranspiration, reaching 290.67, 291.74, 287.81, 279.59 and 279.29 mm month⁻¹ during the respective study years. The results demonstrate the usefulness of CROPWAT 8.0 for assessing seasonal crop water demand and effective rainfall and provide information that can support irrigation planning and water-resource management in the study area.

Breeding for Yield Improvement: A Critical Narrative Review of Genetic Gain, Selection Strategy and Emerging Technologies

2026-10-28T11:36:24+00:00

Increasing crop yield remains a central objective of plant breeding, but yield is an emergent, highly polygenic phenotype whose expression depends on development, resource capture, reproductive success, stress response, management and genotype-by-environment interaction. This critical narrative review evaluates how breeding has improved yield, why rates of progress vary among crops and environments, and which combinations of quantitative-genetic, physiological, genomic and phenomic approaches are most likely to sustain future genetic gain. Literature was selected from major open scholarly sources, with emphasis on peer-reviewed evidence published from 1990 to 9 August 2026 and inclusion of earlier foundational work where necessary. Historical evidence shows that conventional selection, hybrid breeding, plant-type modification and adaptation to agronomic intensification have produced substantial gains, yet observed farm-yield trends cannot be attributed to genetics without careful separation of agronomic and environmental effects. The breeder's equation remains a useful organising framework because gain depends jointly on genetic variation, selection accuracy and intensity, and cycle duration. Genomic selection can improve selection efficiency for complex traits when prediction models are trained on relevant, well-phenotyped populations, but prediction accuracy is not a sufficient endpoint: realised gain per unit time and cost is the decisive criterion. High-throughput phenotyping, environmental characterisation, reaction-norm models, physiological breeding, crop growth modelling and speed breeding can increase the value of genomic information by improving trait measurement and shortening cycles. Genome editing and de novo domestication expand the design space but currently complement rather than replace population-based improvement for polygenic yield. The strongest evidence therefore favours integrated breeding systems that optimise the entire genetic-gain pipeline while maintaining genetic diversity, testing across representative target environments and explicitly managing trade-offs among yield, stability, quality, resilience and resource-use efficiency. Future progress will depend less on any single technology than on coordinated improvement of population design, phenotyping, prediction, crossing, testing and deployment.