Sea Surface Temperature Prediction Using Lunar Orbit Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing weather forecasting systems struggle to reliably predict climate events beyond a few weeks or months due to limited visibility and high computational requirements, necessitating the development of a lightweight system that can accurately forecast sea surface temperature fluctuations and extreme climate events.
Innovation Solution
A machine learning-based system utilizing lunar orbit characteristics and historical data to train models for predicting sea surface temperature fluctuations and extreme climate events, integrating with existing weather forecasting systems to enhance accuracy and extend forecast horizons.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing weather forecasting systems are used to predict climate events, then short-term forecasts (up to two weeks) are reliable, but long-term forecasts (beyond six months) become unreliable
Solution Approach 1:
The patent introduces machine learning models as intermediary components between historical climate data and future predictions. These models process complex patterns from historical data (including lunar orbit characteristics) to generate predictions that extend beyond the traditional forecast horizon while maintaining reliability through data-driven pattern recognition rather than direct physical simulation
Solution Approach 2:
The system performs preliminary analysis by training machine learning models on historical climate data and lunar orbit characteristics before making predictions. This pre-training process enables the system to identify long-term patterns and relationships that can be applied to future predictions beyond the traditional two-week forecast horizon
2Measurement precision
If traditional weather prediction systems are used, then computational requirements are high, but the system complexity increases making it less adaptable
Solution Approach 1:
The patent replaces traditional mechanical/physical weather prediction systems with machine learning-based computational models. These ML models process historical climate data and lunar orbit characteristics to generate predictions, substituting complex physical simulations with data-driven algorithms that achieve comparable or superior accuracy with reduced computational overhead
Solution Approach 2:
The system changes the fundamental parameters of prediction by incorporating lunar orbit characteristics and historical climate data as input features. This parameter transformation enables the machine learning models to capture long-term climate patterns and extend forecast horizons beyond the capabilities of traditional systems
Data Source
AI summary
Systems and methods for predicting decrease in average sea surface temperature (SST) are disclosed. The method includes setting a value of a first Boolean flag when at least one of an eclipse season start date and an eclipse season end date in a particular calendar year occurs in a period of thirty days preceding a timestamp for a particular partition in a plurality of partitions segmenting the particular calendar year. The method includes inputting to a trained machine learning model the first Boolean flag. The trained machine learning model is trained using a plurality of time series data using lunar orbit characteristics. The method includes predicting, using the trained machine learning model, a probability of decrease in the average SST for a predefined time interval in the particular calendar year. The method includes invoking a weather forecasting system to generate weather forecast in dependence on the predicted probability.