Last week, I was awestruck when I heard the weather forecast on the TV right at the end of the news: “Koshi Province will likely experience heavy rainfall, while Madhesh Province could see light rainfall, with other parts of the country remaining clear.” I was left wondering about the utility of such forecasting. I was also left thinking how people in developed countries receive and utilise weather forecasts.

In the developed world, one can simply check a weather app to determine the possibility of rain before leaving home. If there is a high chance of rainfall, they carry an umbrella or adjust their schedule accordingly. Such forecasts empower individuals to make informed decisions, making necessary arrangements before leaving home or office. Moreover, if you were outside and noticed clouds, you could open the app to check precisely when the rain is expected to start. You could even watch a simulated video in near-real time to see the location of rainfall, the situation of the wind, or the time the rain reaches your location.

The benefits of weather forecasting extend to various fields, including agriculture, tourism and outdoor activities, among others.

This modern approach is highly effective, and people can benefit from it. Providing detailed information on the possibility of rainfall can be extremely useful. However, the poor quality of forecasting and dissemination systems raises questions about the hundreds of engineers and meteorologists working with the public-funded Department of Hydrology and Meteorology (DHM). Are they solely collecting historical weather recordings and generating basic information? I remember when, 30 years ago, my grandmother checked the sky in the morning to predict rain and advise us about our daily activities. However, in today’s fast-paced life, relying solely on these traditional methods of forecasting and dissemination seems senseless.

Weather forecasting is not an easy task. In developed countries, meteorological agencies often have millions of dollars in annual budgets. Most of them host supercomputers for regularly modelling complex weather systems. For instance, the Japan Meteorological Agency (JMA) has an average budget of 64 billion yen (around Rs60 billion), while the Korean Meteorological Administration has 114,598 million won (around Rs12 billion in 2018). Both Japan and Korea considered challenging countries for weather forecasting due to their topography and location, achieve above 90 percent accuracy with the help of supercomputers.

In contrast, the DHM works with a much smaller budget. We cannot expect the underfunded DHM to achieve the same level of accuracy as the meteorological agencies in the developed world. It is an evolutionary process that requires optimising models, establishing a weather network nationwide, automating processes, and integrating data to enhance forecasting and dissemination, costing several billion rupees annually.

Despite these limitations, several meteorologists in the department have achieved a high level of skill in weather forecasting. The World Meteorological Organization and several other donor agencies support DHM’s capacity building and improvement of forecasting. As a result, they can run the Weather Research and Forecasting (WRF) model. Additionally, the DHM hosts three radar systems to enhance weather forecasting and acquire local datasets in near-real time. They also have access to weather satellite datasets. These valuable resources should be fully utilised in their weather forecasting efforts. At the very least, the DHM should ensure they have a dedicated research and development (R&D) team continuously working to improve the model with the available resources. By leveraging these capabilities and investing in its research and development efforts, the DHM can enhance the accuracy and reliability of their forecasts and develop effective dissemination systems, benefiting the public and various sectors that rely on weather information for planning and decision-making.

The problem is, the DHM is not known to have a dedicated R&D team to improve their models and dissemination, conduct regular model runs, and update their forecasts hourly to account for continuous changes in weather conditions. The DHM currently runs the WRF model every six hours; however, the public cannot access the forecasts obtained from the model. Many of us depend on the hourly weather forecasting available on our phones (smartphones) or web apps provided by various global weather agencies. Such forecasts are typically based on global models and may not effectively incorporate local information, resulting in lower accuracy. The DHM possesses the operational advantage of integrating local data, enabling them to offer more precise and localised forecasting even with limited resources.

The DHM should regularly evaluate its forecasts against measurements to enhance forecasting and dissemination. Currently, they only archive the forecasting data for a couple of months. It is crucial to have long-term forecasting and measurement datasets for their evaluation and model improvement purposes. The DHM should incorporate local information from various meteorological stations and regularly optimise the model. Every country follows this standard procedure for model improvement, and Nepal must do it independently. Adopting weather forecasting models from countries like the USA, China, South Korea, India, or any other country does not work fully for Nepal, as the influencing parameters differ. Therefore, the forecasting model should be developed and optimised locally using local data.

The DHM can establish a research and development lab in collaboration with academia or private research institutions. It has not published the accuracy of its forecasting, but there is a general feeling that its forecasting is better than global models. They should not be afraid of sharing historical forecasting archives, as this transparency can lead to continuous improvements in their model. However, the DHM has failed to disseminate and promote its competence and improve its public image. Given their resource constraints, the DHM’s accuracy rate of 70-80 percent is pretty commendable.

Accurate forecasts are just part of the triumph of the DHM, for it is equally important to disseminate the information to the users effectively. Unfortunately, the current dissemination approach falls short compared to the quality of the forecasts they generate. If the DHM cannot reach the public, the true value of their forecasting efforts diminishes. Consider the heavy rainfall that occurred in October 2021, severely damaging ready-to-harvest paddy in the Terai. The DHM had predicted that catastrophe in advance, but they failed to inform the farmers effectively. Consequently, many farmers lost crops that could have been salvaged if they had been adequately informed. The DHM must prioritise improving their dissemination methods to bridge this gap.

A possible solution is to develop mobile and web apps that can deliver location-based forecasting with hourly updates. Given that the DHM runs the WRF model, generating forecasting on a grid/location basis, providing localised forecasts should not be daunting. By developing a mechanism to feed these forecasts with filtering or interpretation to standard database systems in a semi-automated manner, they can enable automatic display in mobile/web apps. This user-friendly approach will empower people with near-real-time, accurate and accessible weather information, making them better prepared to handle weather-related challenges and mitigate potential losses.