Climate change poses a significant threat to global food security, particularly in climatically vulnerable regions like China. China experiences frequent agrometeorological disasters, resulting in substantial crop losses and economic damage. Meteorological forecasting services play a crucial role in mitigating these disasters, reducing affected areas and casualties. While the positive impacts are evident, quantifying the agricultural value of these services remains a challenge. Previous research, primarily conducted in other countries, has shown the economic benefits of improved weather forecasting in agriculture, but a comprehensive study using household survey methods in China was lacking. This study aims to address this gap by using household surveys and regression analysis to explore the impact of weather forecasting services on agricultural income and costs in China.

Existing literature reveals several studies quantifying the economic benefits of improved weather forecasting in various agricultural contexts. Solow et al. (1998) demonstrated significant economic impacts of enhanced ENSO forecasts on US agriculture. Hyung et al. (2014) used climate-crop economic modeling to show substantial global benefits from Spanish climate projections. Petersen and Fraser (2001) estimated that seasonal forecasting could increase annual profits in Western Australia. Pierre and Jean-Michel (2005) highlighted the role of seasonal forecasts in EU agricultural decision-making. Rebecca et al. (2020) explored the varying usefulness of seasonal forecasts depending on technology and user application. Tesfaye et al. (2019) examined the economic value of meteorological services for Ethiopian smallholder farmers. However, no previous research has used household survey methods to assess the agricultural value of China's meteorological forecasting services, making this study novel and significant.

This study employed a household survey method to assess the value of meteorological services to farmers in China. A total of 5807 questionnaires were distributed across 31 provinces and over 2800 counties, with a final sample of 3771 valid questionnaires used in the analysis after removing invalid responses and outliers. The questionnaire collected data on respondents' demographics, farming practices, access to weather information, and perceived impact of weather forecasts on income and costs. The data was analyzed using multiple linear regression models with cluster-robust standard errors to account for clustering at the provincial level. Two regression models were constructed: one to analyze cost reduction and another to analyze income gains. The cost reduction model examines the relationship between cost savings and factors like land, labor, capital, planting area, crop price, information sources, and frequency of forecast attention. The income gain model investigates the relationship between income increase and similar factors, focusing on the number and type of information channels used, and frequency of attention to forecasts throughout different agricultural production stages. The models included control variables for occupation, province, and education level.

The analysis revealed a positive correlation between the frequency of using meteorological forecast information and increased agricultural income. Each additional type of access channel to meteorological information led to an average income increase of 11.11 yuan. Paying attention to weather forecasts at each production stage resulted in an average income increase of 5.755 yuan per mu. The study also found that access to information through agricultural experts and radio significantly boosted income, with increases of 18.22 yuan and 28.49 yuan, respectively. Regarding cost savings, the results indicate a positive and statistically significant impact of weather forecasts on labor, liquid capital, and fixed capital inputs. For each additional standard deviation increase in input, cost savings were approximately 10.75 yuan for labor, 10.07 yuan for machinery, and 5.47 yuan for agricultural inputs. However, land input, planting area, and crop price showed no significant effect on cost savings. These findings suggest the primary mechanism for cost reduction is through factor input optimization based on weather forecasts.

The findings confirm the significant positive impact of weather forecasting services on both income and cost reduction in Chinese agriculture. The substantial income gains associated with increased access to information highlight the importance of improving information dissemination channels and promoting effective communication strategies. The results suggest that targeted interventions focusing on radio and agricultural experts could be particularly effective in reaching farmers and improving agricultural outcomes. The cost-saving effects are also noteworthy, indicating that farmers can optimize resource allocation and reduce expenses through timely adjustments based on weather forecasts. The lack of significant effect of land input, planting area, and crop price on cost savings suggests that the primary benefits of weather forecasts come from operational efficiency rather than impacting these broader market forces.

This study provides strong empirical evidence supporting the economic value of meteorological forecasting services in Chinese agriculture. The findings underscore the importance of expanding access to timely and relevant weather information through diverse channels, particularly focusing on effective communication strategies. Future research could explore the potential of tailoring weather forecasts to specific crops and regions and investigate the effectiveness of various extension services aimed at improving farmer knowledge and application of weather information. Further research can also investigate the long-term impacts of these services and potential interactions with other agricultural policies and interventions.

The study relies on self-reported data from a household survey, which may be subject to recall bias and subjective interpretations. The cross-sectional nature of the data limits the ability to establish causal relationships definitively. While the regression models controlled for several factors, unobserved heterogeneity might still influence the results. The sample, while large, may not perfectly represent the diversity of agricultural practices and conditions across all of China. Further research utilizing longitudinal data and more sophisticated econometric techniques could strengthen the findings.