Putting a price on the value of weather forecasts

Saturday, January 31, 2015

Not much research has been done to assess the benefits of accurate weather forecasting, but it has certainly been undervalued. And underinvestment has likely also contributed to the nation's poor productivity record

by PHIL CHADWICK

What is the value of a good weather forecast? The answer is a lot harder to quantify than you might think. How does one count the plane that didn't crash because the pilot monitored the weather forecast and took the appropriate action? What about the traffic accidents that were avoided when the travellers revised their plans to avoid dangerous driving conditions? These potential disasters and countless other plausible scenarios didn't occur and can't be tallied.

The forecast that was ignored with dire consequences is another matter. The media often refer to these as "surprises," Acts of God or freak weather conditions, implying that they were unavoidable. I don't think so. Even a great forecast has no value if the clients do not use it in planning their activities. My brother went driving even though freezing rain was predicted and the ensuing collision was no accident, although that's what it was called.

Canada is the second largest country in the world (after Russia) and has the longest coastline at 243,980 kilometres. Its population of 35 million has created quite a huge infrastructure over the generations. Canada has also become increasingly vulnerable to the extremes of weather and climate. Insurance claims from weather-related disasters totalled $3.2 billion in 2013 and have been above $1 billion a year since 2007.

One would think that the socio-economic value of weather forecasts would be quite significant, but we do not know for sure. There has not been enough research in Canada to answer this very important question. Even less has been done to attempt to translate the benefits of forecasts improvements in savings for specific economic sectors. We must look elsewhere for an estimate of the value of the weather forecast.

Two recent studies might shed some light on this. A 2010 study from Thomas Frei assessed the benefits of weather and climate information on Switzerland.1 Frei concluded that the value of weather services was of the order of $US400 million for households, agriculture and electricity production and distribution alone. Costs for the production and delivery of these services were in the order of $60 million (or a cost of $8 per citizen). Specific studies addressing the value of services to individual sectors (aviation, transport, marine, forestry) have not been included in this study. These sectors of the economy enjoy weather services funded out of the $60 million.

A 2009 study by Jeffrey Lazo et al. discussed the costs and value of weather forecasts to households in the United States.2 This extensive study concluded that the value of weather forecasts for households alone was in the order of $31.5 billion a year or 0.2 per cent of the country's GDP. The production costs for this information were about $5.1 billion a year, or $15 per citizen.

These "production" costs encompass a very large number of programs that provide services and information to the international community, including Canada. Thus, production costs for the U.S. programs are higher compared to those of other countries, such as Switzerland. The specialized sectors, such as energy, construction and transportation, are not included in this study but are provided with weather information out of the same funding.

For general households, the ratio of costs to benefits of weather information ranges from 1:4 to 1:6. When one considers that specialized sectors and services are provided with the same infrastructure and resources, the benefit/cost ratios are certainly significantly higher.

Based on current budgets, the cost of weather services in Canada is less than $10 per citizen per year, probably close to the $8 per year cost in Switzerland.  U.S. satellites and other data used extensively by the Meteorological Service of Canada are funded out of that higher cost of $15 per citizen in the United States.

It is very reasonable to assume that the sensitivity of the Canadian economy to weather is at least equivalent to that of the United States. Canada is large and occupies more diverse weather and climatic zones than the United States. However, using the U.S. value of weather services ratio of 0.2 per cent of the GDP, one arrives at a value of $3.1 billion for the equivalent value of weather forecasts in Canada. This Canadian figure is probably much higher, given our extremes in weather and climate.

A further complication is that it is highly likely that forecast utility and forecast quality do not change at the same pace. Reducing forecast errors by 10 per cent more likely represents an increased economic value of significantly more than 10 per cent. The 30 per cent reduction of forecast errors over the last 14 years, and the inference that this reduction corresponds to more than a $1 billion a year in economic value is quite in line with these recent studies by Frei and Lazo et al.

Figure 1 shows the trend in Numerical Weather Prediction Forecast Quality since its inception in 1958. The Quality Score (110-S1) along the left axis is sensitive to the forecast placement of the high- and low-pressure areas and the associated pressure gradients between them. The Quality Score does not explicitly measure the accuracy of the placement of the associated weather. That is a much more technical verification process but, for simplicity, the placement of forecast pressure patterns is very closely related to the weather.

A Quality Score of 100 would be a perfect match of the pressure centres and their associated gradients between what was predicted and what was later observed. The different coloured lines are for different prediction lead times. As would be expected, the Quality Scores are lower for predictions further into the future. The graph covers the prediction periods with up to five days (120 hours) of lead time.

A 2001 study3 commissioned by the Canadian Meteorological and Oceanographic Society of Canada addressed the economic value of meteorological services in Canada among many other issues. The following is an excerpt from this study:

"Meteorological services create value and economic benefits in two distinct ways: by improving productivity in many of the nation's key industrial sectors and thereby fostering growth in the Gross Domestic Product (GDP); and by facilitating fewer weather-related fatalities and injuries, less time stuck in traffic jams, less destruction of wildlife and habitat, and other prediction-related improvements. Society values such benefits but does not include their value in the accounting for Gross Domestic Product.  

"Benefits in the first category are called 'private' benefits (because they arise principally in the private sector of the economy). This study finds that each one per cent increase in the net value of the meteorological infrastructure (meteorological capital stock) leads to a 0.5 per cent increase in total factor productivity and a 1.8 per cent increase in the nation's GDP. The precipitous decline in the value of the meteorological capital stock over the past 25 years has thus contributed to the nation's sluggish productivity growth and the disappointing growth in living standards. HLB Decision Economics Inc. estimates that annual GDP would have been $15.02 billion greater had the value of the meteorological capital stock been maintained at its 1976 level.

"Benefits in the second category are called 'social' benefits because, despite having monetary value, that value typically is not reflected in private sector transactions.  This study finds that each one per cent improvement in weather prediction accuracy yields at least $1.02 billion in social benefits over a 30-year period. Since the deterioration in the capital stock has almost certainly meant foregone opportunities to improve forecasting accuracy, social benefits will have inevitably been foregone accordingly." 4

According to the statement in the last paragraph of the study, the social benefits for a one per cent improvement in weather prediction accuracy were of the order of $33 million a year in Canada in 2001. It is quite reasonable to expect that this figure is even higher in 2014.

Various meteorological measures for improvements in meteorological quantities suggest that predictions have improved by about 30 per cent since 2000. Using the Canadian Meteorological and Oceanographic Society social benefits factor of the order of $33 million a year for a one per cent improvement in weather prediction, the benefits associated with improving forecasts by 30 per cent should be of the order of $1 billion a year (i.e. 30 x $33 million a year). This is probably a serious underestimate of the true value of Canadian weather forecasts. It seems clear by looking out the window that the socio-economic value of weather and environmental prediction and services will increase even faster over the next decade.

Some may argue that the CMOS study is more than 10 years old, that it is over simplistic and that it does not provide highly detailed information as to how the results were obtained and the conclusions drawn. Agreed.

However, very little research on the economic value of weather forecasts has been done in Canada and not much more has been completed around the globe. We need to start somewhere. Specific sectors such as air transportation, tourism, construction, production and distribution of electricity, agriculture, water availability, all benefit from quality weather and climate prediction. What is the value of weather services to these sectors? What is the value of weather services to the farmer?

With answers to these and other questions, the solution to our weather woes might be reinvestment in the environmental sciences. I felt the cuts from my first day as a meteorologist in 1976 and spent my entire career trying to demonstrate the value of our services and the importance of weather.

Investments in weather services and infrastructure are paramount to ensure that Canada is well positioned economically in the world of tomorrow. An accurate weather forecast is better than gold if it is properly used. BF

Phil The Forecaster Chadwick has been a professional meteorologist since 1977, specializing in training, severe weather and remote satellite and radar sensing.

1 Frei, Thomas: Economic and social benefits of meteorology and climatology in Switzerland, Meteorol. Appl, 17: 39-44, 2010
2  Lazo, Jeffrey K., Rebecca E. Morss and Julie L. Demuth: 300 Billion Served: Sources, Perceptions, Uses, and Values of Weather Forecasts, Bull. American Met. Society, 90: 785-798 (June 2009)
3 HLB Decision Economics Inc. – Canadian Meteorological and Oceanographic Society, "Optimizing the Public and Private Sector Roles in the Provision of Meteorological Services, Nov. 19, 2001. Document available at http//www.cmos.ca/Privatesector/metstrategyappC.pdf
4 HLB Decision Economics Inc. – Canadian Meteorological and Oceanographic Society, "Optimizing the Public and Private Sector Roles in the Provision of Meteorological Services, Nov. 19, 2001, p. 7