The official start of winter is less than seven weeks away and weather forecasters in New York are preparing. The New York State Mesonet, headquartered at the University at Albany, is partnering with the National Oceanic and Atmospheric Administration on efforts to improve winter storm observations and the sharing of that data. The two-year project is supported by $600,000 in NOAA funding. WAMC's Jim Levulis spoke with Jerry Brotzge, the Mesonet’s project manager.
Brotzge: The New York State Mesonet is a network of weather stations, we have 126 weather stations across New York state, we have at least one weather station in every county. And we have an average station spacing of about 17 miles across the state. And what that means is each of those weather stations continuously, automatically records the weather. And every five minutes, it takes a snapshot of temperature, relative humidity, wind speed, wind direction, a lot of different variables. And we collect all that data in real time and then we create maps and products from that data.
Levulis: And now the Mesonet is partnering with the National Oceanic and Atmospheric Administration on a new project to enhance winter storm observations. What does this effort entail?
Brotzge: Right, right. So one of the toughest things that that you can do is measure snowfall, for example, snow depth. A lot of times we just don't know exactly where it's freezing rain or raining or snowing. And so we've tried to address that problem. And this new funding really allows us to do three things. First, it allows us to improve how we measure the winter weather data. If you want to drive a meteorologist mad, just asked them how to measure snow depth, and you get lots of different answers. It's really a tough thing to do. And so we're going to try and improve how we measure say snow depth, or how much water is in the snowpack, a really critical measure for hydrologists. So we're going to try to improve how we measure winter weather data. The second thing is we're going to make these data available in real time, meaning we're going to send this weather information to the National Weather Service, we’ll send these data to the local hydrologist and the State Department of Transportation. So we're going to try to make these more readily available than they are currently. And then the third thing is, it's one thing to just provide data, but it's another to provide useful information. And so the third thing we're going to do is apply machine learning, apply artificial intelligence to that data to help improve our warning operations both for with the National Weather Service as well as with state transportation, aviation, emergency managers.
You know, I guess one question that comes up is, is why in New York? You know, roughly half of all states have some form of a Mesonet, some form of a statewide weather network. But a lot of those networks are in the southern tier of the country. And they're primarily focused on emergency management for severe storms. Of all those mesonets, New York is well placed to monitor the winter weather. New York is among the snowy states in the country. And it's also what many would consider the freezing rain capital of the country. And you look at the cost to the state, our winter weather cost, sometimes over a billion dollars in annual losses due to disruptions to travel, in aviation and power grid. And so, you know, there's a tremendous need here, especially in New York to optimize how we respond to these winter weather events. And so this is a first step in many steps on helping to mitigate those losses and save community tax dollars.
Levulis: And so you mentioned the difficulty of monitoring winter storm events and snowfall. What are some of the avenues you're going to be exploring to enhance the ability to enhance those measurements?
Brotzge: Fortunately with the New York State Mesonet, we're already collecting these data, a lot of these data in real time, all across the state every five minutes. And for example, we have a snow depth sensor that tells us the depth of the snow. But there are a lot of issues with that. For example, currently, we measure snow depth with what we call a snow depth sensor. It uses sound waves to determine the height of the snow. But inevitably, under extremely windy conditions, you have a lot of drifting of snow. So one thing we're going to do is take a lot of field measurements to determine exactly under what conditions can we trust the snow measurement? Are there certain periods when you have a lot of drifting and how can we correct for those issues? How representative is a particular snow depth sensor measurement relative to say the larger area, say a field? And so those are some questions we'll work to measure and work to correct. Another question is, you know, precipitation type. And so we have a way to estimate freezing rain at our Mesonet stations. And so we'll be using a combination of other external data to validate how we estimate freezing rain or the freezing rain depths and estimates.
Levulis: And overall, how might this new project and the data that comes from it and that hopefully increased sharing of that data, how might it help local governments, utilities, emergency responders, etc, prepare and then respond to winter storm events?
Brotzge: In a number of different ways, by basically improving what we call situational awareness, you're just providing more accurate information to the people that need it. For example, for the National Weather Service, if they're able to more accurately monitor what's going on in the field, say if they know where it's starting to freezing rain, if they know where the snowfall rates are exceeding say one inch per hour, that’s information they can then issue more correct warnings, faster warnings, faster advisories to the general public. One issue we have with winter weather is the impact on of course roadways. Roughly a quarter of all vehicle accidents annually are caused due to slushy and snowy roads. And so by providing greater accuracy, both spatially and temporally, we can then work with emergency managers, with the transportation folks on salting the roads earlier, faster, more frequently where it's needed most. And if you look at salting cost, it roughly cost $1,000 per mile to salt a road. And by providing more accurate weather information, you can optimize where the roads are salted, and how frequently and thereby reduce the salt usage you apply. And that of course saves tax dollars and also helps the environment.
