Core Insights - The article highlights the advancements made by Hohai University in developing a self-controlled ocean-atmosphere coupling prediction system, which has significantly improved climate forecasting capabilities in China [1][2][3]. Group 1: Technological Advancements - The ocean-atmosphere coupling prediction system was developed to reduce reliance on foreign numerical models, addressing significant "bottleneck" risks in climate prediction [2]. - Key breakthroughs include the creation of multi-source ocean-atmosphere observation data assimilation technology, which efficiently integrates satellite, buoy, and vessel data [2]. - The development of ensemble filtering assimilation methods has enhanced the quantification of prediction uncertainties by converting single-value outputs into probabilistic ranges [2]. - A parameter estimation and correction system was established to optimize model parameters, significantly reducing model errors [2]. Group 2: Predictive Accuracy - In 2023, the system successfully predicted a moderate-strength El Niño event nine months in advance, with a prediction accuracy improvement of over 15% compared to international mainstream models [3]. - The system is capable of accurately forecasting other critical climate phenomena, such as the Indian Ocean Dipole, providing reliable climate support for major events like the Beijing Winter Olympics [3]. Group 3: Practical Applications - The prediction system has become an integral part of the national marine environment forecasting center, contributing to decision-making for flood prevention, drought relief, water resource management, food production, energy supply, and major engineering projects [4]. - The system has received national recognition, including certification from the China Meteorological Administration and a first-class award from Jiangsu Province for marine science and technology [4]. - Over the past decade, the team has transitioned from reliance on foreign models to developing a complete set of assimilation technologies, achieving a leap in China's marine environment forecasting capabilities [4].

Report Industry Investment Rating No relevant content provided. Core Viewpoints of the Report - The report uses satellite remote sensing, meteorological data, and historical yield models to estimate the yields of key global agricultural products in August 2025 and monitor their growth. The estimated yields of key monitored crops have increased to varying degrees. - La Niña is expected to appear in September, with a weak intensity and lasting until January 2026, with a probability of 50%-60%. The Indian Ocean Dipole has turned negative, which together promotes more precipitation in Southeast Asia, making it difficult to form a long-term drought. - The weather in South America is less affected by La Niña. Periodic droughts may affect southern Brazil and northern Argentina, but the rhythm is earlier than in 2024, and the overall intensity is similar [2]. Summary According to the Table of Contents Global Key Agricultural Product Yield Estimation - **Varieties, Time Window, and Method**: The monitoring cycle covers the growth period of crops in the Northern Hemisphere in August 2025. The monitored varieties include US soybeans, corn, cotton, Canadian rapeseed, Australian rapeseed, and Southeast Asian palm oil. The time - cycle spans 20 years from 2005 to 2025, using current and historical data. The monitoring uses 24 key indicators from satellite remote sensing, meteorological data, and field observations, and a self - built yield model based on a deep - learning algorithm [6][7][12]. - **Yield Estimation Results**: Overall, the growth and development of key crops in each region are in good condition, and the yields are generally on the rise. The US soybean and corn regions are likely to set historical records. Cotton yields have increased compared to the previous month. Canadian rapeseed yields are expected to reach 2.27 tons per hectare, and Australian rapeseed yields have been raised to 1.79 tons per hectare [13]. Global Key Agricultural Product Growth Monitoring - **Malaysian and Indonesian Palm Oil Producing Areas**: Vegetation indices in the Malay Peninsula and Sumatra have increased, while those in Kalimantan have declined. Only precipitation in the Malay Peninsula has increased, and temperature and humidity have fluctuated moderately [16][19]. - **US Soybean and Corn Producing Areas**: Vegetation indices in most states have increased significantly. The growth of soybeans and corn in the Midwest has reached a new high, and the eastern region is also above the historical average. Precipitation shows regional differentiation, and soil humidity in most states has increased significantly. Temperature fluctuations are moderate [25][30][31]. - **US Cotton Producing Areas**: Vegetation indices show a differentiated trend, with Oklahoma showing overall growth and the southeast experiencing a decline in LAI. Precipitation varies greatly, and soil humidity fluctuates slightly. Temperatures in the southeast have decreased, while those in Oklahoma and Texas have slightly increased [40][41][45]. - **Canadian Rapeseed Producing Areas**: No detailed content provided in the report. - **Australian Rapeseed Producing Areas**: No detailed content provided in the report. Analysis of the Trends of La Niña and the Indian Ocean Dipole in the Second Half of the Year - La Niña is expected to appear in September, with a weak intensity and lasting until January 2026, with a probability of 50% - 60%. The Indian Ocean Dipole has turned negative, which together promotes more precipitation in Southeast Asia, making it difficult to form a long - term drought [2]. Analysis of the Future Weather Trends in South America - The weather in South America is less affected by La Niña. Periodic droughts may affect southern Brazil and northern Argentina, but the rhythm is earlier than in 2024, and the overall intensity is similar [2].