Introduction

Agriculture, particularly rice cultivation in northern Iran (Gilan and Mazandaran provinces), is highly vulnerable to weather fluctuations and climate change. Despite the availability of meteorological data and warning systems such as the "TAHAK" platform, rice farmers continue to suffer substantial economic losses annually. Reports from the Agricultural Insurance Fund (2018) indicate that over 63,000 rice farmers in Gilan province alone received compensation for damages caused by cold spells and heavy rains during the 2016 crop year. This paradox—abundant meteorological information yet persistent high losses—highlights a critical gap: the disconnect between data production and practical, actionable use by farmers.

Previous studies have identified barriers such as farmers' misunderstanding of forecasts, lack of spatial relevance, weak institutional support, and limited access to extension services (Sharifzadeh et al., 2010; Forozani et al., 2018). However, little attention has been paid to the lived experiences of leading farmers who have successfully integrated weather forecasts into their daily decision-making. This study aims to fill that gap by exploring how agrometeorological services and weather forecasts contribute to loss reduction and productivity enhancement among rice farmers in northern Iran.

Methodology

This qualitative study employed a grounded theory approach to capture the depth and complexity of farmers' lived experiences. Participants were 15 leading rice farmers from Gilan and Mazandaran provinces, selected through purposive sampling. Inclusion criteria included: at least 10 years of practical farming experience, familiarity with weather forecasts, registration in the TAHAK system, and willingness to participate in in-depth interviews. Snowball sampling was used to complete the participant pool.

Data were collected through semi-structured interviews lasting 45–60 minutes. Interviews were conducted in the farmers' fields, homes, or local mosques, creating a comfortable environment for open dialogue. The interview guide focused on: access to meteorological services, use of weather information in farming decisions, impact on loss reduction and productivity, and factors enabling effective use of forecasts.

Data analysis followed Strauss and Corbin's three-stage coding process (open, axial, selective) using MAXQDA software. Trustworthiness was ensured through member checking, peer review, and triangulation of sources (interviews, field observations, meteorological documents). Ethical considerations included informed consent, confidentiality, and the right to withdraw.

Findings

The analysis yielded 225 coded references organized into eight thematic categories. The most frequent code (53 references, 25.2%) was "optimization of operation timing," indicating that the primary practical function of weather forecasts is helping farmers schedule plowing, seeding, fertilizing, weeding, and harvesting. "Trust in meteorological data" followed with 47 references (22.4%), revealing that trust is not an abstract attitude but emerges from three concrete components: spatial accuracy, actionability, and timeliness.

A significant finding was the "temporal-spatial and content gap" (32 references, 15.2%), with farmers explicitly stating that weekly forecasts are ineffective and that they need 24–72 hour forecasts with village or farm-level resolution. The "triple knowledge integration" pattern (28 references, 13.3%) showed that successful farmers do not replace indigenous knowledge with scientific data; rather, they integrate three knowledge streams: formal scientific knowledge (weather data, agronomic principles), experiential indigenous knowledge (cloud and wind signs, animal behavior), and situational monitoring knowledge (field thermometers, record-keeping).

Seven structural barriers were identified, totaling 82 references: weak dissemination of TAHAK services (21 references), information being difficult to understand (18), lack of credibility due to non-localized forecasts (32), extension agents' unfamiliarity with farmers' real needs (18), weak trust due to occasional inaccurate forecasts (11), impractical information (11), and lack of timely access (15).

Discussion

The findings reveal a paradigm shift from reactive to proactive risk management. Farmers who received direct, ongoing training moved from "reacting to events" to "anticipating and preventing based on data." This aligns with global studies (WMO, 2024; Boon et al., 2022) but adds a critical insight: the temporal-spatial mismatch between weekly, station-based forecasts and farmers' need for daily, field-level predictions.

Trust emerged as a three-layered construct: instrumental trust (based on repeated accuracy), relational trust (built through direct interaction with a trusted expert), and systemic trust (confidence in institutions). This multi-layered understanding goes beyond previous conceptualizations of trust as a simple psychological variable.

The triple knowledge integration pattern challenges the false dichotomy between "traditional" and "modern" knowledge. Farmers do not choose one over the other; they create a dialectical synthesis that leverages the strengths of both. This finding has profound implications for extension programs, which often dismiss indigenous knowledge as folklore.

Based on these findings, a four-layer model is proposed:

1. Information layer: Accurate, localized (village/farm level), timely (24–72 hour), and actionable forecasts (translated into specific agronomic recommendations).

2. Trust and interaction layer: Direct communication with a trusted local expert via simple, low-cost channels (phone calls, text messages in local dialect).

3. Education and empowerment layer: Practical, field-based training on interpreting microclimate, understanding cloud and wind patterns, managing diseases based on humidity forecasts, and integrating indigenous knowledge with scientific data.

4. Institutional and support layer: Policy integration across the Meteorological Organization, Ministry of Agriculture, Insurance Fund, and extension services; defining "agricultural meteorological liaisons" in each extension center; linking weather data to crop insurance incentives.

Conclusion

Agrometeorological services in northern Iran can achieve maximum effectiveness in reducing losses and enhancing productivity when designed not as simple "weather news" but as a "contextual decision support system." The four-layer model provides a practical roadmap for achieving this transformation. As one participating farmer stated: "Meteorology doesn't just tell me if it will rain tomorrow; it tells me when to plant, when to spray, when to harvest—so I avoid losses and harvest better rice. That's what makes a smart farmer."