Originally written as a paper for the course Global Economy and Public Policy at FLAME University.
Weather affects more than a third of the world’s total GDP, and forecasts are beneficial to everyone ranging from farmers to government agencies. Information on weather and forecasting has therefore become a global concern, and one of the driving forces that is changing how information is acquired, analysed and assessed in big data analysis.
Weather data has a high value and an extremely high demand. Last year, IBM acquired The Weather Company to meet the rising need for weather information by companies with its cloud-based software. However, although existing data services make information more accessible, they have yet to address the problem of information creation. Information is generated mostly by weather stations owned by the state or companies. This provision of information is restricted to the stations that belong to them and does not record areas outside of their coverage. This prevents the localisation of the data as it has a low resolution, which in turn affects the accuracy of forecasts.
This is a pitch for Governments to invest in basic, distributable technology. There exist elementary weather stations that provide simple information such as humidity, soil and atmospheric temperature, pressure, and air quality, including a simple anemometer and rain gauge. All of these functions can and have been produced in small, easy to use, and inexpensive single-chip weather stations that are easily sourceable and distributable. These weather stations can be distributed to any individual, household, or organisation, who can record weather in their own backyards and contribute to a global database in real time. This database is an open use, open access database that spans every continent. A farmer from coastal India could see the data generated by another in Maldives, allowing him to assess oceanic conditions and the arrival of the South-Western Monsoon winds. At an individual level, having a weather system on his own farm could also optimise irrigation systems and help save a lot of resources. A significant portion of the problems faced by rural farmers is caused by inaccessible information about the weather, a variable that their livelihood and survival depends on.
This will be an open science policy, where data is citizen-sourced, and quality as well as reliability is maintained through Government-provided tools. It is a system of globally democratised information, in which everybody has a part to play in data acquisition, analysis, and access.
Decentralisation has been a hot topic when it comes to the governance of the provision of public knowledge and resources. However, this global system would take a step further and make it not just decentralised, but distributed, wherein each individual, household, or firm becomes both, a receiver and a provider of information.
What’s in it for the Government? This dense flood of data comes together to form a precise and localised database that provides an in-depth understanding of the weather system, with its trends and fluctuations. Weather forecasts become more accurate, allowing for Governments to decide the best plan of action when anticipating a weather anomaly, saving scores capital, time, and labour. Better data is better resilience.
Additionally, weather forecasting plays a crucial role in the economics of renewable energy. Grid operators are constantly trying to meet supply and demand. On the demand side, significant advancements have been made in models that can predict how much energy consumers will use. On the supply end, however, predictions are difficult largely due to the intermittency of renewable energy supply, which depends almost entirely on the weather. Precise weather forecasts can save the grid several million dollars a year. In the US, Xcel Energy saved up to $40 million over four years, by improving their weather forecasting models. Given more inputs of information from across the world, this can be scaled up by several times.
It could, however, be argued that distributing the agency of creating weather data could lead to the wrongful manipulation of data. However, since the technology and software are provided by governments, and data is recorded and uploaded in real time, changes in the program can easily be detected and prevented. If the data is denser, it makes it easier to spot an anomaly. It is also important to acknowledge that the gestation period for such a system would be at least a year, wherein the technology and know-how are distributed and taught. The benefits would be seen in the long run, when the system becomes more efficient over time, as the knowledge and database grows.
In India, where agricultural occupations sustain most of the population, there is a constant need for more localised data that can be generated from farms to improve agricultural practices like irrigation. This would not only make farmers better equipped, but also give them more self-reliance as they can generate their own weather information and monitor the health of their farm. These devices would improve the connectivity in places where internet is unavailable, as they can act as individual servers, transmitting and receiving information. Additionally, in countries with such varying climates within the same administration, denser data could help in the formulation of more adaptive policies, that move beyond the one-size-fits-all approach. Big data can allow for individuals, organisations, as well as Governments to make smarter and more effective decisions.
Author: Shimul Bijoor