Methods and systems for generating a neural network (NN)-based climate forecasting model are disclosed. The methods and systems perform steps of selecting a global climate simulation dataset from a plurality of simulation datasets each generated from a global climate simulation model; training the NN-based climate forecasting model on the selected global climate simulation dataset; and validating the NN-based climate forecasting model using observational historical climate data. Embodiments of the present invention enable accurate climate forecasting without the need to run new dynamical global climate simulations on supercomputers. Also disclosed are benefits of the new methods, and alternative embodiments of implementation.

An automated global weather notification system is provided. The automated global weather notification system is capable of obtaining observational weather data, including data form of forecast grids, and applying business rules and conditional variables to that data. Based on the business rules and conditional variables, notifications are generated. Relevant users are identified in particular geographic areas and notifications are delivered to those users via, for example, SMS, MMS, email, or other methods of electronic information delivery.

Surface modification control stations and methods in a globally distributed array for dynamically adjusting the atmospheric, terrestrial and oceanic properties. The control stations modify the humidity, currents, wind flows and heat removal rate of the surface and facilitate cooling and control of large area of global surface temperatures. This global system is made of arrays of multiple sub-systems that monitor climate and act locally on weather with dynamically generated local forcing & perturbations for guiding in a controlled manner aim at long-term modifications. The machineries are part of a large-scale system consisting of an array of many such machines put across the globe at locations called the control stations. These are then used in a coordinated manner to modify large area weather and the global climate as desired. The energy system installed at a control stations, with multiple machines to change the local parameters of the ocean, these stations are powered using renewable energy (RE) sources including Solar, Ocean Currents, Wind, Waves and Batteries to store energy and provide sufficient power and energy as required and available at all hours. This energy is then used to do directed work using special machines, that can be pumps for seawater to move ocean water either amplifying or changing the currents in various locations and at different depths, in addition it will have machineries for changing the vertical depth profile of the ocean of temperature, salinity and currents. Control stations will also directly use devices such as heat pumps to change the temperatures of local water either at surface or at controlled depths, or modify the humidity and salinity to change the atmospheric and oceanic properties as desired. The system will work in a globally coordinated manner applying artificial intelligence and machine learning algorithms to learn from observations to improve the control characteristics and aim to slow down the rise of global surface temperatures. These systems are used to reduce the temperatures of coral reefs, arctic glaciers and south pacific to control the El Nino oscillations.

The system as described collects and utilizes weather data sensor information in order to rapidly collect and update weather forecasts using real-time weather data collected at high rates of frequency, and use this collected high frequency weather data to rapidly correct and update the weather forecasts generated by the system.

Disclosed is a method for siting heat wave monitoring stations based on risk evaluation, including: acquiring historical meteorological data of a target region, and preprocessing the historical meteorological data to generate a gridded associated meteorological data set; identifying historical high-temperature heat wave events based on the associated meteorological data set, and calculating parameters and summary indexes of heat wave feature of grids; evaluating station building priority of the grids based on spatial distribution features of the summary indexes; acquiring multi-source data, and evaluating a heat wave risk to generate a heat wave risk map; performing iterative computation using an optimization algorithm based on current station building information, temporal-spatial distribution features of meteorological factors and the heat wave risk map to determine alternative station building positions; and acquiring on-site survey information of each alternative station building position, and determining a position where a station is to be built.

A reference devices system (720) for generating and managing calibration data for sensors used to obtain weather information is provided. The reference devices system (720) includes a plurality of reference devices, wherein a reference device (110) of the plurality of reference devices is configured to be calibrated against a weather station (115) and to serve as a calibration reference for a community device (130).