An Open Letter To Miami University & Its President: Please Share Your Airport Weather Station Data

To Miami University and President Crawford:

I have great respect for Miami University. I applied to your MBA program in late 2015 because I have great respect for your business program. My mother is a three time graduate of your school – a bachelors and two masters degrees in education – so I was raised to respect Miami University. I have been to your campus, and it is beautiful. I was there a couple of years ago to visiting your geothermal plant; as a meteorologist and someone who cares for the environment, it is wonderful to see a university not only take steps to become more sustainable and green but also improve their image through actions over words.

While I recognize your primary focus as a university – as your mission statement highlights – is to have an “unwavering commitment to liberal arts undergraduate education and the active engagement of its students in both curricular and co-curricular life,” your university is among a select few in our area to have an own and airport. It’s 3 miles west of campus and 2 miles west of downtown Oxford:


It’s a very convenient airport for those who want quick access to the university by airplane. With the drive being 56 minutes from CVG and 62 minutes from the Dayton International Airport, it’s ideally located between two major airports and two large cities.

I’m certain you know of this airport, but what you may not know is that there is a valuable piece of equipment sitting just north of your runway:


That fenced in area (with a path leading to it) contains an AWOS – or Automated Airport Weather Station. It’s actually one of the nicer ones out there. More formally known an AWOS III P/T, it can measure and report barometric pressure, altimeter settings for pilots, wind speeds and directions, the temperature, the dewpoint, visibility, sky conditions, cloud coverage, liquid precipitation amounts, precipitation types, and the direction of thunderstorms relative to the airport. This equipment creates valuable information for pilots but also to meteorologists.

This piece of equipment isn’t cheap. While the cost of these varies depending on what an airport wants, can afford, whether they can get nearby parts, and the year purchased, an article in The Leader-Herald in Gloversville, New York from 2017 claims a local airport manager purchased an AWOS for $240,000. I have heard costs for an AWOS closer to $70,000 or “above $100,000.” Regardless, the AWOS at your airport came at a significant cost to the university. Additionally, there are variable costs associated with keeping the AWOS up and running, including electricity to the system and replacement parts.

When I returned home and first started working in the Tri-State as a meteorologist in 2011, I began researching and understanding the processes that allow weather data to flow into various systems, including to the National Weather Service. In the years that followed, I worked with local airport managers to see if funding could be secured to share their weather station data out with the world by using a vendor. I am thankful that the Warren County, Ohio Airport; the Clermont County, Ohio Airport; the Fleming-Mason, Kentucky Airport; the Madison, Indiana Airport; the Fayette County, Ohio Airport; and the Batesville, Indiana Airport have all started sharing their airport weather station data with the Federal Aviation Administration (FAA) through a vendor. We have more local weather data than ever before.

But we don’t have the Miami University Airport’s weather station data. I believe it was 2013 when I first reached out to your airport manager about sharing your weather station data. He informed me at the time that he was interested in sharing the airport weather station data but also addressed cost concerns. I believe he spoke to his manager at the time, but the idea didn’t get much traction from the manager. After some time – perhaps a year – had passed, I called the airport manager again to express my continued interest in getting a feed of Miami University’s weather station data. His position had not changed, but I asked for his manager’s contact information this time. He obliged, and I reached out to his manager by phone. I never received a phone call from that manager, and I believe through an Internet search some months later that he had retired.

In the spring of 2017 and as I mentioned above, I came to your campus to see your geothermal plant and other nearby sustainability operations. I was impressed to see such forward thinking from a university. I also respected the business angle; being more sustainable benefits the bottom line.

While I was there, I connected with your marketing team and also your Director of Planning, Architecture & Engineering. I asked both parties about the possibility of sharing Miami University Airport’s weather station data out with the world. The marketing department responded to me in the days that followed with news that the university was not interested in sharing your airport weather station data. I followed up with the Director of Planning, Architecture, & Engineering in the days after I met him, and he responded saying that any additional operating expenses [at the airport, I assume] would make any additive costs prohibitive.

That takes us to today. I feel the need to clarify, publicly, the importance of your airport’s weather station data:

  1. Safety: Your weather station as measured severe wind gusts many times over the last few years. These measured gusts led to Severe Thunderstorm Warnings, which are issued to keep people safe and informed about dangerous storms. Many of those people placed in these warnings are your students and staff. Many of those people are also residents of Oxford. The National Weather Service has ways of accessing your weather station that the rest of the world does not; they have this ability in an effort to keep people safe, but know that because they do not have a stream of your data, they have to go looking for it. As a meteorologist, I want you to know that meteorologists depend on real-time weather station from local airports, especially during times of active and severe weather.
  2. Quality: Backyard, personal weather stations are not as reliable or accurate as airport weather stations. State-owned weather stations networks – like the Ohio Department of Transportation’s – are not as accurate as airport weather stations. I recognize that you can get real-time data from Oxford area personal weather stations or the ODOT weather station northwest of Oxford near the Ohio/Indiana border. While “close enough” may work for temperatures, visibility and wind measurements are not as reliable or accurate for non-airport weather stations. Airport weather stations also observe precipitation types and intensities; state and personal weather networks don’t measure that kind of weather data. Your airport weather station does. That’s why it’s important to share out your data.
  3. Marketing: Imagine media meteorologists sharing weather conditions from the Miami University Airport. Imagine pilots seeing weather conditions in the cockpit on their iPads and landing there because they know it is safe to land…and then they refuel and increase your revenues.
  4. Cost: There seems to be a misconception of the cost of sharing airport weather station data out to the world through a vendor. Yes, it true; you must use a vendor to share the AWOS data out through either an FAA-certified method or non-FAA-certified method. This process does not cost thousands of dollars. In fact, I’ve done the research on costs on sharing the airport weather station data through a vendor, and I want to share my research:
    • Stanwyck Avionics (non-FAA-certified method, but still flows to NOAA’s secondary data stream where the National Weather Service and media can use it): $1,500 one-time software charge/cost and an annual fee of $300.
    • RSINet (FAA-certified method): $60/month with a one year contract. No upfront costs through this vendor that I’m aware of.
    • AnyAWOS (FAA-certified method): $70/month. No upfront costs through this vendor that I’m aware of.

Based on the price of Jet A and 100LL fuel on the Miami University Airport website, you could fund sharing your airport weather station data each month by selling about 15 gallons of gas. A Cessna 172 – a very common general aviation aircraft – has a fuel capacity of 42 gallons.

I looked at Miami University’s 2016-2017 operating budget, and found $70 a month would take 0.4% of the airport’s annual budget. Unfortunately, the revenue and expenses for the “airport services” segment of the operating budget are equal:


The Director of Planning, Architecture & Engineering was correct in saying that additional costs would make this segment unprofitable. Of course, the nominal cost of sharing AWOS data out with the world through a vendor can be budgeted. I am quite surprised – as a business person – that the “airport services” segment is the only one in the “Auxiliary Enterprises” umbrella that doesn’t generate a profit.

The marketing budget for the Farmer School of Business alone in 2015-2016 was $3.78 million. With an annual budget in the hundreds of millions, can Miami University not afford $840 a year to spend on sharing airport weather station data out through a vendor? This small expense is an investment in weather safety and awareness while also increasing marketing for the university. Your asset that costs tens of thousands of dollars next to the runway holds valuable data that only a select few can see now. For a fraction of the cost of that asset, you can share that quality data that comes from that asset with students, staff, the community, the media, meteorologists, pilots, and everyone in the world…and they will know what the weather is like at Miami University.


Scott Dimmich


Posted in Uncategorized | 2 Comments

A Meteorologist’s Guide To Get More And The Best Quality Weather Station Data

Data matters in meteorology. Some of the most underrated data in the country, in my opinion comes from personal weather stations, groups of weather stations called mesonets, and airport weather stations. While satellite, radar, and model data can highlight and suggest threat areas, weather stations can directly observe, quantify, verify, and sample conditions.  Surface observation data that is shared with NOAA (the National Oceanic and Atmospheric Administration) and the NWS (National Weather Service) is also used in computer forecast models and for verification of weather conditions – including rainfall totals and wind gusts in active or severe weather situations.

It can take years to get existing and new points of information on the map integrated into one place, but it can be done. Official weather records are great, but there is often an incredible amount of space between these sites, and daily, comprehensive weather records (including those with wind, snow depth, and snowfall totals) are almost entirely limited to airports in larger cities.

This blog post focuses on three things: 1) how the data flows, 2) how do get data into the data streams, and 3) how to fix errors in data points flowing into the data streams.

All of this information is publicly and freely available on the Internet. There is nothing propriety here, and I link to public sites throughout the article. The data streams themselves are public, and that’s the way they should be. The more weather data that is public, the more that is accessible to those who need it. My hope is that more data will be added to the streams over time so that the weather can be better understood and verified.

How The Data Flows

The ultimate goal is to get weather data to NOAA and NWS.

Before we start pushing data into the streams, we must first understand how the streams work. There’s are two main NOAA data streams: the ADDS and MADIS. It is important to know the difference:


ASOSes are controlled cooperatively in the U.S. by the NWS, FAA, and DOD, and the data from them flow to ADDS. AWOSes can be controlled by the FAA, local or state governments, or private groups. Data from an AWOS can go to ADDS or MADIS, and not all AWOS data goes to MADIS or ADDS. ASOSes are superior in some, but not all, ways to AWOSes. You can verify which ASOSes and AWOSes share their data with ADDS by putting in an airport code or ICAO here:

MADIS’ website shows the flow of data through MADIS looks like this:


This process is very similar to that of ADDS data except FAA certified data is the only input to ADDS. MADIS takes just about everything else that’s left (that doesn’t flow to ADDS). Here are some example inputs to MADIS:


While not “official,” MADIS is very important. As of the writing of this blog, there are 34,191 points of surface weather data flowing through MADIS from 74 different networks or mesonets. About 25.4% of those observations come from the Citizen Weather Observer Program (CWOP), where personal weather stations share their data to NOAA (more on this below).

NOAA’s MADIS map shows all of the publicly available MADIS sites and ADDS sites. MADIS sites are the ones that do not appear on the ADDS map. You can verify this by looking at the MADIS and ADDS maps. Not all data collected by MADIS is redistributed to the public; you can review what MADIS data goes where here.

Much of the discussion below focuses on CWOP, Roadway Weather Information Systems (RWIS), Weather Underground, AWOS, and WeatherLink IP data. These are all low-hanging fruits compared to growing the other weather networks.

How To Get Data Into The Streams – CWOP

First, let’s focus our efforts on CWOP and how it works. Some, but not all, personal weather stations – through either hardware, software, or both – will share data to CWOP.

The basic steps to get personal weather station to CWOP are:
1) Get a CWOP ID unless you’re an amateur radio operator
2) Configure the weather station software to share data with CWOP
3) Verify CWOP is getting the data
3) Register the station with NOAA by e-mail

Visually, this is the goal:


It’s important to note that just because data goes to CWOP doesn’t mean it goes to MADIS! The weather station owner must register their station with NOAA for the data to flow to MADIS. Please see the note on checking to see if their are unregistered sites for a given state in the “How To Fix Errors In Data Points Flowing Into The Data Streams” section below. 

Like I said above, not all weather stations can share data with CWOP. These can’t:

– AcuRite Pro Weather Center
– Neatmo
– Ambient Weather WS-1000
– Ambient Weather WS-1001

To repeat, if an owner of one of these weather stations wants to share data with CWOP, it can’t be done and you should give up trying immediately!

With this considered, let’s move to step two. How do you share data to CWOP? It’s easy if a weather station owner has a Davis WeatherLink, WeatherLink IP, or popular weather station software (MeteoHub, MeteoBridge, Virtual Weather Station, Cumulus, WeatherSnoop, or WxDisplay) because all of the directions, produced by the National Weather Service in El Paso are here:

There are the step-by-step directions for sharing the data with the world. Add it to your bookmarks if you need to! Click here for more information on sharing data from weather station software.

Once we’ve configured the software, we need to verify that CWOP is getting the data. CWOP should have an inventory of the data in no more than 30 minutes, but it may be there in as little as 5 minutes. Weather reports on CWOP’s page will surface here:, where XXXXX is the CWOP ID or amateur radio handle. Also, once the data starts showing up on this page, please check to make sure the weather station latitude and longitude are correct here:, where XXXXX is the CWOP ID or amateur radio handle. A minus sign may be needed on the longitude to get you into the western hemisphere.

If the weather data makes meteorological sense and the location on the map is correct, we are in good shape. If these items are wrong. check the settings in the weather station software. See my note on the minus sign for the longitude above.

Important: we are not done yet! We still need to register the weather station with NOAA. When the settings are right, have the weather station owner send an e-mail to CWOP support ( requesting to register the  weather station with CWOP; in the e-mail,  include the CWOP ID, the closest town to the weather station (use a specific community), zip code, and the weather station’s elevation in meters (not feet). If the weather station owner gets an e-mail from NOAA CWOP support, verify the date that they will add you to the MADIS map…along with the CWOP ID. This is an important step because registering the station ensures the data will flow to the world through MADIS.

It’s important to note that the siting of the weather station matters! More on proper siting for this can be found here. You can share this PDF with weather station owners to help them correct weather station data biases.

How To Get Data Into The Streams – RWIS

Roadway Information Systems (RWIS) are weather networks owned and operated by state departments of transportation (especially in colder climates). Some networks are robust, and others are weaker. The data that comes from these networks are still very important!

Odds are good that the state is already sharing the data from all of their RWIS sites, but you should check because you might be surprised! MADIS don’t always configure their redistribution systems to share all RWIS data as new RWIS sites come online.

The goal here is to:
1) Find the state’s official RWIS website
2) Compare the sites on it to NOAA’s MADIS map and make sure none of the RWIS sites are missing

If all of the RWIS sites are there, you’re good to go. If not, e-mail and bring it to their attention. They are there to help you and can work with a state’s RWIS contact about sharing data and troubleshoot errors.

If there are problems with the data coming from a specific RWIS site, that falls under the responsibility of the state department of transportation. Contact them directly to fix individual RWIS site issues.

How To Get Data Into The Streams – AWOS

Most AWOSes are located at airports, and some AWOSes don’t share their data with MADIS or ADDS. The first step here is to find the airports that have an AWOS that aren’t sharing their data with MADIS or ADDS. You do that by:

1) Using the FAA’s ASOS map to see where there are ASOSes and AWOSes
2) See if those same sites are on NOAA’s MADIS Map

Once we’ve found a site that is on the FAA’s ASOS map and not on the MADIS map, we need to get the airport identifier (ICAO) for the site. Next, we need to get in touch with the airport manager. Legally, the airport manager’s name, address, and contact number must be listed in FAA records. You can get this contact information by putting the airport ID or ICAO into this website:

It is now your goal to convince the airport manager to share their AWOS data – through a vendor – to either ADDS or MADIS. Googling “AWOS vendors” will give you a list of the vendors. Sharing data with MADIS is cheaper, and sharing data with ADDS is more expensive (because it’s FAA certified). If you’re a member of the media, saying you’ll share their airport AWOS data with the community on-air helps persuade them to pay a vendor to share their data. If you’re a pilot, saying you’ll use the data for safety will help. If you’re member of the community, you can say you just want to put the community on the map.

Precipitation type and intensity sensors are not required on FAA certified AWOSes (all ASOSes and nearly all AWOSes themselves are FAA certified, but the feed out of them does not have to be). Having this precipitation information in a weather report is helpful, but the sensors cost money. This can be a hard sell to an airport manager, but it can be done. Keep in touch with them, and be persistent, and you’ll eventually get the sensor.

How To Fix Errors In Data Points Flowing Into The Data Streams

– As I alluded to earlier, there are registered CWOP stations and unregistered CWOP stations. The data from the latter does not flow to MADIS. Think of the latter as a TV that doesn’t have the cable connected correctly into the back of it; you just want to complete the transmission.
– The easiest way to view registered and unregistered CWOP site is on the CWOP website and by state. You can find the page for a given state by clicking on the map under the “Data Time History Displays” here. Clicking on Texas, for example, will bring you to list of registered stations hereNotice the e-mail contact information for each weather station owner is listed on this page. Clicking on “WxGraph” for each site will allow you to see recent trends in weather data or show you that CWOP hasn’t received data from that site in a while; if there’s no recent data for a site, you can reach out to that weather station owner via e-mail to notify them they haven’t been sharing data in a while, ask them to check their weather station, or ask them to look at a sensor if their data is inaccurate. If you scroll down the page, you’ll see a link that says “UNREGISTERED XXX MEMBERS WITH UNVERIFIED STATIONS,” where XXX is the name of the state. An example for the unregistered Texas page is here. This is a list of people that registered for a CWOP IP but they didn’t register their station with NOAA. Just because they didn’t register their station with NOAA doesn’t mean they aren’t sharing their data with CWOP! Clicking on the “WxGraph” link for a site will show you recent data CWOP has received from that weather station owner. You might find a site that has site that you want to flow to MADIS; if you encounter this, e-mail the weather station owner and ask them to register their station with NOAA (see e-mail to directions above). Even if a unregistered weather station owner isn’t sharing data to CWOP, you know they are at least interested in sharing data with CWOP and probably own a weather station; you might as well get that data! E-mailing them and seeing what their situation is will helpful. You might find they just hadn’t gotten around to getting their weather station setup and may be excited you’re using their data (especially if you’re in the media or another public setting). The CWOP site gives you their first name, which is a good start for relationship building and networking.

Weather Underground
– Weather Underground (WU) data does not flow to MADIS. They are actually kind of jerks about it when you do some research.
– There is, however, a workaround for most but not all weather stations: contact the weather station owner and ask if they will share their data to MADIS and WU (actually, I prefer they do because WU has good weather archives). First, find a site on Weather Underground that you would like to see on MADIS. I’d suggest using the Wundermap. Here’s an example from Kaufman, Texas:


If I click on the Station ID link (the KTXKAUFM4 part), I’m taken to this website. If I click the “About this PWS” link here in the top of the page, I see this window pop-up:


The first thing to notice here is that it not on my list of forbidden, dead end weather station brands and types above, so there is hope we can get this weather station into MADIS. So how do we contact them? We could message them if we have a WU account, but this is a dead end. We want to convert their latitude and longitude to an address and mail them a letter. Go back to the Wundermap and zoom in down to street level on the weather station; also, turn the “Map Types” in the upper-right corner to “Hybrid”:


Once these settings are set, we see the map looks like this:


There’s a really good chance here that that homeowner owns that weather station. If this was a densely populated neighborhood, we might have to get in closer and try to figure out if the weather station is in the backyard of one house or the front yard of another.

So how do we get their address? Take their latitude and longitude and put into Google Earth and see where they are and what county they are in. This station is in Kaufman County, Texas. The easiest way in the modern era to get an address is through GIS, or Geographic Information Systems. If you Google the name of the county, name of the state, and “GIS,” you’ll likely get a government website that can use to get you to an address. In the example, Google searching “Kaufman County Texas GIS” gets you to an interactive map where you can get an address.
– Have the courage to contact someone. I know it’s weird to send someone a letter in the mail out of the blue, but you can do it…and if you want their data, you’ll have to.
– Once the person e-mails you, you can greet him or her, and explain that you’re trying to direct their weather station data to MADIS, and send them step-by-step directions to configure their software using this page that I linked to above. You may have to Google directions for other weather station software types.
– Most weather station owners, with the right approach, will share their data with MADIS. Some will not; for those that don’t want to share, you might try asking again in a year or so.

– When there’s an error in data, call the airport manager. They don’t want bad data from their airport being broadcast to the world. 
– It’s worth noting that the clocks inside AWOSes tend to drift. If the clock drifts into the future, NOAA will reject the observation (observations don’t come from the future), and you won’t get the data through ADDS or MADIS. You fix this issue by calling the airport manager, telling them this, and convincing them that their AWOS technician needs to check the AWOS clock once a month (they can adjust this remotely).

Davis WeatherLink IP Stations

Once you get the hang of how to get WU stations into MADIS, try your hand at getting Davis WeatherLink IP weather stations online. Davis’ WeatherLink IP website is here:

There are many weather stations on here that you won’t really be able to find anywhere else. Here’s a snapshot of Davis WeatherLink IP weather stations (dots) in the Tampa area:


Before we start contacting everyone on this map, remember that:

1) WeatherLink is not the same as WeatherLink IP. In a WeatherLink setup, data flows from the Davis weather station to a computer and then to CWOP. In a WeatherLink IP setup, data flows from the Davis weather station to CWOP through Davis’ servers; WeatherLink IP owners still must have a CWOP ID so they can push their data to CWOP!

2) Some WeatherLink IP owners are already sharing their data with CWOP and/or MADIS. You can verify whether the data are flowing to MADIS on NOAA’s MADIS map.

3) There will be obvious cases when the position of the weather station is wrong (in a field, over water, etc.). Clicking on each of the dots on the map may give you clues about a location (a business or organization name), an address (123 Easy Street), or someone’s name (J. Richardson). If you only get a name, use the location (latitude/longitude) on the Davis WeatherLink IP map and county GIS webpages to find the address in the latter and see if the name makes sense. You might find a John Richardson two houses down from where you thought on the GIS maps, and that is likely the person you want.

Human Processing Still Required

There will be times when weather stations go offline or report information that is clearly wrong. It is up to you to find ways to – over time – remedy the problems or find ways to quality control – either in your own mind or with a computer – the errors included in the data. Don’t get burned presenting a 0.00″ rainfall total from MADIS on a rainy night or a -40° temperature on a spring day!

Here To Help

Questions? Comments? Issues? Concerns? I’m here to help. You’re welcome to contact me at the “Contact Me” link near the top of the page.

Posted in Uncategorized | Leave a comment

10 (Relatively) Easy Steps I Recommend For Being A Better Meteorologist And Communicator

As I transition away from a chapter of my life centered on meteorology, I’ve realized I’ve learned a lot about what methods work in weather and which ones don’t. There are no hard-and-fast rules for success in meteorology or weather communication (and I’m continuously learning every day). After more than 10 years of weather, I wanted to share what I believe are 10 relatively easy ways to be a better meteorologist and communicator

1. Look to see what is going on. It is a huge mistake to just jump into the model world without reviewing recent weather trends. Where are the clouds, which way are they moving, and what altitude are they at? Where are the upper-level troughs and ridges? Where is their surface convergence and upper-level divergence of air? How fast and in what direction are showers and storms to the west moving? Computer forecast models frequently suck, so blindly depending on them will get your feet to the fire frequently. You should be spending at least several minutes each time you forecast looking at radar, satellite, water vapor, weather balloon, and surface report data before you even look at a computer model. While it is a mistake to expect models to match reality (they shouldn’t match), you can better gauge a given model’s strengths and weaknesses if you know what has occurred recently. As my father told me when I was a child, “You’ve got to know where you’re at to know where you’re going.”

2. Stop being everyone else. Just because there are many meteorologists in the country using 4-panel computer forecast guidance plots doesn’t mean you have to. Just because there are many meteorologists in the country that frequently say “those temperatures” and “these temperatures” does not mean you need to. Just because there are many meteorologists in the country that make heavy use of the terms “tracking,” “watching,” and “alerting” doesn’t mean you need to. Just because someone found success in forecasting an event or communicating the impacts of that event effectively doesn’t mean you’ll have the same success nor does it mean he or she will have the same success during the next event. There’s nothing that says you can’t dress for the job you want even if everyone else is wearing business casual. There’s nothing that says you can’t use forecast tools that the veteran in your office refuses to use. There’s nothing that says you have to put your forecast slides in a specific order. You don’t have to and you shouldn’t forecast and communicate like everyone else. You have to respect the spirit of the previous forecast (see below) and the ideology of where you work, but part of what you gets you a meteorology gig is your technique and individuality (at least I hope that’s how you’re measured). Your time is best spent maximizing your potential – whether it’s telling a weather story through explainer graphics, using long-range forecast analogs to give your customers insight beyond the week ahead, or developing unique data sets that can augment weather coverage – and not comparing yourself to other people. If everybody looks the same, success (pick your metric) is more likely to be evenly distributed.

Think of it this way: if you really like a store that’s 10 miles away, and a new company opens nearly the exact same store 5 miles away, odds are good you’ll start going to the closer store and – over time – both stores will have about the same number of customers (all other factors equal). A third and fourth store may open up, and the pattern repeats. Now suppose the store that’s 10 miles away has the special sauce, the secret weapon, and the sustainable competitive advantage that their competitors can’t reproduce; odds are good you’ll continue going to or start going back to that store. The moral of the story here is that the store 10 miles away from you dared to be unique and not like everyone else and got the customer base, attention, and income they earned and deserved.

3. Look at the previous forecast. If a previous forecast is available, you should be looking at it every time you forecast. The last forecaster – hopefully – spent hours looking over information before you showed up; they probably saw a few things in model guidance that you don’t see in the first few minutes of being in the forecast chair. Additionally, when every forecaster in the group reviews the previous forecast, there is more consistency, which benefits the customer and reduces the amount of work (in the form of updating) for forecasters. Refusing or simply not being interested in the current, persistence forecast signals arrogance and ego, and these elements of your personality will surface in time whether you try to hide them or not.

4. Make an actual forecast, and don’t be wishy-washy. Models suggest fog late tonight; odds are good many meteorologists will forecast “patchy fog” or “fog possible.” One computer model says showers are coming Thursday, but another says we’re dry. What’s the forecast? “Showers possible.” Showers are on the far outer-edge of an area of low pressure are approaching. In these scenarios, I often hear “there’s a slight chance for a few scattered showers.”

These are not strong forecasts. These are the meteorological equivalents of a punt in football. If you spend some time in the models or know your climatology, you’ll know where the fog will be favored and – with time and the right education – you’ll be able to forecast what the visibility will be at several spots. Fog is more than possible; it’s something every meteorologist should be able to forecast. When it comes to clouds and showers, you have more than just the latest model runs to look through; you can look the runs before them and the previous forecast. You can review the model trends for a given forecast hour. You can look at ensemble models. You can look at individual ensemble models members. You can look at satellite and radar trends.

Regardless of what you use, you should always make a forecast and run the risk of being wrong just like every other meteorologist in the world. You’re far more likely to gain credibility and respect from a customer if you give them the information they want to hear and be as specific as you can.

In a world where all the forecasts are “chances,” “possible,” “something to watch,” or some other form of hedging a bet, you want to be definitive, honest, and to the point. If you want to stand out from the crowd, don’t be wishy-washy.

5. Be conversational. It’s easier said than done, but communicating a forecast is just talking…so just talk. Talk in the same way you would to a friend sitting next to you on the couch about the weather; if you do this, you’ll avoid all of the terrible forecast crutch phrases. I’ve never had a meteorologist talk to me about the weather and say “those temperatures are going to fall through the 50s overnight;” instead, they say “we will fall through the 50s overnight.” You don’t talk to your mother about “windy conditions,” but you will say “it’s windy, but the wind will relax this evening.” Have you ever seen two people describe the weather for the “nighttime hours tonight?” I haven’t, but I’ve seen meteorologists do it. Have you ever looked up outside and thought the skies were beautiful? You said the sky was beautiful. I still don’t know what a thundershower is. If a cell produces lightning, it’s a thunderstorm; if it doesn’t, it’s a shower. Did you ever learn about thundershowers when you first learned about the weather in grade school? I didn’t because they don’t exist.

Important parts of communication are subject matter knowledge and relaxing. If you’re using crutch phrases, you’re either nervous or not as familiar with current weather and upcoming changes as you should be. Treat the camera as a neighbor. Treat your customer in the forecasts you draft as a friend you’re standing next to.

6. Review recordings of what you communicate, and work to improve what you didn’t like. When I was early in my broadcast career, I would record weathercasts from different television markets off of TV station websites. Of the ones I liked most, I would transcribe what the meteorologist said to text. I found several great weather phrases just by reviewing and typing out these recordings. I got in the habit of doing that in my next two jobs for at least a couple of months. I learned a lot, and I realized something important:

In my opinion, the best weather communicators spoke in complete, full sentences that flowed together and connected their content back to the forecast.

When you speak to another person either in person or in writing, you write in full sentences that flow together and often tie back to a central theme. That’s what communicating a forecast is all about: flow and emphasizing key points with inflection.

Whether you produce written, audible, or visual forecasts, record it and watch it. See what you did well and what you could have done better. Do this process for your presentations until you strengthen the good and eliminate the bad. Use pauses in your presentation. Inflect your voice when you’re highlighting what is important. If nothing else, avoid cliches and replace them with something better.

7. Don’t just be a “latest model” forecaster. When you forecast only using the “latest models,” you’re forecasting more of the noise and less of the signal. There will be anomalies, there will be large model shifts, and there will be situations – especially when a system comes ashore and is finally sampled by a weather balloon – when the forecast needs to be changed. Realistically, though, changing a low or high temperature by a degree gives your customer no additional value in most situations. What is the difference between 61° and 62°? Absolutely nothing. 31° versus 32° is a different story, however. Even with this acknowledged, though, remember the previous forecaster – assuming he or she reviewed the last forecast – carried over elements of the forecast before him or her.

Being a “latest model” forecaster can prove costly. Here’s an example from this past Sunday using the NAM model. The latest model at the right shows a lack of clouds (tan, low relative humidity amounts) at 850mb or 5,000 feet above the ground. The model run before it had a significant amount of cloud cover (green, higher relative humidity amounts):


So which one is right? Of course, the real answer is a function of the previous forecast, what other models and their trends suggest, and what satellite and observed trends suggest (if anything at all). Blindly following the latest model here means you’ll likely remove and decrease clouds from the forecast. If this “latest run” isn’t a true trend, the next forecaster will just revert back to the cloudier solution; this will leave your customers are confused, show a lack of collaboration, and show your inability to make a reliable, accurate forecast.

8. Don’t be a “mandatory pressure level” forecaster. Weather doesn’t just happen at 850, 700, and 500mb. When you’re first learning to forecast and getting introduced to models, you’ll see the classic 4-panel map like this.


There it is: 500mb (18,000 feet) vorticity, sea-level pressure, 700mb (10,000 feet) relatively humidity, and precipitation. In reality, this is not that helpful for forecasting. How about boosting this up to 6-panels?


The lower-middle panel showing 850mb relative humidity and the lower-right panel showing accumulated precipitation doesn’t really help us beyond what we had above. What if a shallow layer of clouds forms at an altitude of 3,000 feet developing in the afternoon? What if upper-level winds to the west carry a copious amount of high-level clouds to your forecast location? You’ll likely never see them if you are just focused on clouds at 700mb and 850mb.

You need to know temperature and dewpoint profiles in the atmosphere. You need to know the magnitude of shear and instability to see if thunderstorms will form. You need to look at more than just the temperature at 850mb and the surface to make a snow forecast.

If you’re only focused on the main pressure levels, you’re missing most of the atmosphere. Find ways and tools that can give you a more comprehensive view of the atmosphere.

9. Stop saying – or never start saying – “um,” “uh,” “like,” “you know” and other crutch words. These words may be “conversational,” but they make you look unintelligent, unprepared, and immature in presentations. I have seen people “um” and “uh” through life-threatening weather situations, and it’s not just painful to watch; it’s suggesting the speaker can’t handle the situation and don’t take people’s safety seriously. When you present weather information, you should be able to pivot. If you speak a sentence with words in an order you didn’t intend, can you still complete the sentence? If not, I encourage you to master and refine this skill. Can you explain the same weather in more than one way? Doing this will help you avoid the killer crutch phrases. Instead of saying “uh” or “um,” just don’t say anything. There is nothing wrong with 1 or 2 seconds of silence. If you’re reading weather bulletins, make sure you know where the key information is so you are not tripping over words.

10. Don’t idolize communicators. Being the best forecaster you can be means being the best version of yourself. If you’re good at using metaphors when explaining how weather works, refine and perfect that unique craft. If you’re good at connecting current or future weather to climatology, make that your thing. If you’re good at tailoring the forecast to people’s lives and events, do that. With any forecasting job you have, there will be rules to follow and guardrails for which you’ll want to stay inside. If you’re a National Weather Service forecaster, you’ll have to make aviation forecasts to fit certain specifications. If you’re in radio, you’ll have to use company taglines and only give forecasts for certain times of the day. If you’re in television, you’ll have to cover certain types of forecasts and make relevant graphics. While you should follow the rules, you should do everything you can inside of the guardrails to make a unique, interesting, and captivating presentation that showcases your best talents and skills. If you look like every other forecaster and forecast the same way, where is your value? Do everything you can to do what you love and highlight your value add. What got someone famous isn’t necessarily what will work for you. Focus on building what you already have.

Posted in Uncategorized | Leave a comment

Scott Dimmich, MBA

I try to be a modest person, but today I want to share with you something for which I am tremendously proud and required months of hard work.

For the last 2 years, I’ve been very quiet about my accomplishment, but with a career transition imminent, now is the time to share my story.


I am a proud graduate of Xavier University’s Williams College of Business. From January 2016 through early July 2017, I began and completed a Masters in Business Administration with a concentration in Business Intelligence.

In the fall of 2015, I looked at my career options for the future. My choices were: 1) stay where I am at and hope to advance, 2) stay in broadcasting and look for advancement opportunities elsewhere, including in other cities, or 3) choose a new career path. Someone I knew suggested I consider pursuing a master’s degree – specifically an MBA – in late 2015. It took less than an hour of research to realize that I would need to take about 15 courses to complete an MBA, and if I wanted to get an MBA and both start and complete a job search in 2 years, I would have to get moving. I immediately applied to two MBA programs, and ultimately chose Xavier. I’m glad I did.

I took 2 to 4 classes (usually 3) per semester from the spring semester 2016 through the first summer session of 2017. For a year and a half, I essentially had one night each week away from classes and coursework. Most of my vacation days in 2016 were spent studying, writing papers, or in a library. Some classes – like marketing and managerial strategy – were manageable; others – like operations management, managerial accounting, and data mining/statistical analysis – were quite difficult, mathematically intensive, and time consuming.

Balancing a full time job and school was challenging, and it was even more difficult to mix in the other facets of life. After a while, I ran out of “100 percents” to give. Some people fell away as I pursued a degree that I knew would make me stronger and increase my likelihood of career options. I have a new level of respect for how difficult business decisions can be and how much impact they can have.

I didn’t have a concentration when I first started at Xavier, but a professor of mine – who I now consider a friend – saw my love for data and gathering insights from it. He encouraged me to pursue a Business Intelligence track, where I could learn everything I could about data, the systems that handle it, and the techniques used to understand it.

In the summer of 2017, I finished my last two classes and began a job search. At the end of that search, I found a job that was a perfect fit and allowed me to merge my business intelligence acumen with my work experience in communication and forecasting. This job will also enhance my quality of life and work-life balance; working regular business hours is a sharp contrast to waking up at 1am for work, working holidays, and working weekends. Lastly, there will be many opportunities to grow, advance, and learn in my new job; growing and advancing in a sustainable company and industry is important to me.

My job search confirmed the things I believed to be important to me two years ago; most importantly, I wanted to stay in Cincinnati where my friends and family are. I worked hard to come back home to Cincinnati in 2011 after years working elsewhere and at college. I’m thankful that I will be living in Cincinnati – my hometown – for the foreseeable future.

I think it is best that I not share specifically where I’ll be working. This is not for lack of pride, but – instead – in an effort to make elements of my life more anonymous. I’m not one that needs to be in the limelight to be happy. I have never been interested in being an on-air “talent” or “personality” (the terms make me cringe); I chose to be a broadcast meteorologist so I could forecast and communicate the weather. After all, it was the Blue Ash/Montgomery/Symmes Township tornado of April 9, 1999 that caused be to revisit my career path.

I make this career transition with my head held high and with a lot more self-confidence and self-worth that I had years ago. I am working in my hometown with more than 10 years of experience and as an American Meteorological Society Certified Broadcast Meteorologist. I have the honor of being recognized, interning with, and being hired directly by Cincinnati’s greatest meteorologist. I’ve worked alongside some talented people who genuinely want to make the people around them better. I’ve worked some severe weather events I will remember for the rest of my life – most notably the deadly tornado outbreak of March 2, 2012. As an MBA graduate, I see businesses differently. I appreciate the impacts of business strategy. I see the difficulty and value in creating a sustainable competitive advantage. Perhaps most importantly: I see the ethical side of decisions and know when to spot poor ethics, unfair practices, and caustic behavior.

The weather will always be a part of life. I assure you I am still very connected to meteorology and will not be abandoning it. I still love the weather, and I always will.

I am thankful for the relationships I now have as a Xavier MBA graduate and for the friends I have made in the program. I am thankful to my friends and family for being patient and supportive of me in my career transition. Having the people I love, trust, and respect standing with me means a lot. I hope my late father is proud of me.

This is not a goodbye. This is simply a fresh start, and I hope you’ll celebrate with me.


Posted in Uncategorized | 54 Comments

When A Tornado Debris Signature (Apparently) Isn’t A Tornado


This is a mind bender. Imagine you’re nearly certain something is happening and is a threat to someone’s life…then after an additional investigation, nothing happened.

There was a rotating thunderstorm in northern Brown County Sunday night around 1am. It needed a Tornado Warning, and it got one. Shortly after the warning was issued, there was a tornado debris signature, which essentially confirmed a damaging tornado was occurring (based on radar).

I put my own credibility on the line to highlight what I saw on Facebook:

I did the same on Twitter…once to highlight the original warning:

…and the second to emphasize the radar confirmation of a tornado:

This is the “usual” shower and thunderstorm mode of radar – formally called reflectivity – during the warning:


If you look carefully and near Buford (north of Mount Orab), you might see a hook-like feature, signaling a possible tornado. This isn’t easy to see, so let’s look at the Doppler part of Doppler radar, which will tell us how winds are moving relative to the radar:


In pink, I’ve circled what is called a couplet, or an area where winds are moving towards and area from the radar in close proximity. Red here means winds are moving away from the radar, and green colors means winds are going towards the radar. The environment around this storms has to be supportive of tornadoes in order for us to see this as a tornado; in this case, there was support for strong, severe, and tornadic storms. Let’s subtract out the overall motion (speed and direction) of the storm so we can see the storm relative motion. If you’re still confused, think of it like this: it’s easier to assess the rotation of a toy top on a table if it’s nearly stationary versus moving rapidly across a table. Here’s the storm relative velocity animation:


Again, I’ve highlighted a couplet with a pink circle. That’s strong rotation slowly weakening as it goes east.

Modern, dual-polarization radar can tell a meteorologist about the size and the shape of objects it scans, relatively speaking. If a radar scans a “bin” of the atmosphere and finds objects of varying shapes and sizes, that bin’s targets will have a low correlation. If a radar scans a “bin” of the atmosphere and finds objects of similar shapes and sizes, that bin’s targets will have a high correlation. To quantify this, we’ll turn this correlation into a correlation coefficient. Here’s the animation of this correlation coefficient just after 1am Sunday night:


I’ve circled a consistently low correlation area moving east. This area is basically the same area with strong rotation and a strong thunderstorm. As a meteorologist, this area low correlation coefficient within the area of a strongly rotating thunderstorm signals that a damaging tornado has occurred. The higher up in the atmosphere this low correlation coefficient goes, the more likely there is a damaging tornado and the more likely the tornado is to be strong. That low correlation coefficient area went up to about 5,200 above radar level:


The actual altitude is greater when you consider the height of the radar and the curvature of the Earth.

This signature (the combination of the storm, strong rotation, and low correlation coefficient) is called a tornado debris signature. Now you see the reasoning for my Facebook post and tweets above. This is clearly a dangerous situation.

The Tornado Warning ends in about 30 minutes. The storm weakens, and order is restored.

Let’s fast forward to this morning. The National Weather Service plans to survey the area for damage. The damage is surveyed, and the NWS has a verdict:


What? What happened here? We had a tornado debris signature last night.

I asked a radar expert for a second opinion:

Professor Matt Kumjian of Penn State makes a good point: it’s a blurry area. A low-level circulation with leaves in it is not a tornado. Another broadcast meteorologist – Ryan Hanrahan – suggested a drone may be helpful in a damage survey. Joey Picca of the Storm Prediction Center also has some thoughts:

I don’t know how “exhaustive” the search was by the NWS, but apparently they didn’t see anything of interest. Maybe it is a false positive. But the radar in this case is like a camera witnessing a crime where police later find nothing at the scene. Or the radar shows a series of facts but the jury says not guilty.

What do you think? I think I’m frustrated.

Posted in Uncategorized | 2 Comments

Let’s Talk About The Severe Weather Threat Late In The Weekend

It’s November in the Tri-State, which means we are entering a secondary severe weather season. 11 tornadoes have been confirmed in the Tri-State during November since 1950. On average, 7 Severe Thunderstorm Warnings and 2 Tornado Warnings are issued in the Tri-State each November. The jet stream is getting stronger this time of the year, and the amount of instability in the air is getting lower as temperatures and dewpoints drop.

There’s a threat for severe storms late in the weekend, specifically Sunday night and early Monday, as a cold front sweeps into and through the Ohio Valley. While the Storm Prediction Center does not issue severe weather categories (marginal, slight, etc.) 4 or more days out, they have highlighted an elevated risk for severe storms in the Ohio Valley Sunday night:


Think of the yellow area as a “slight risk” of severe storms and the orange area as an “enhanced risk” of severe storms. Cincinnati is – more or less – in a slight risk for severe storms Sunday night and early Monday.

The environment that these storms develop and evolve in is important. It will be warm Sunday with high temperatures in the 70s:


Low temperatures Monday will be around 60°, and it appears we’ll hit our low temperature at least a couple of hours after sunrise Monday:


Temperatures will likely fall through the 70s and 60s Sunday night. This is warm enough to support thunderstorms.

Moisture is another important ingredient for storms. Dewpoints will be in the 60s ahead of Sunday’s front:


This is sufficient moisture for thunderstorms, including severe storms. Wind shear, the change in the direction and speed of the wind with increasing altitude, supports organized storms and the threat of severe storms. Here’s what the Thursday morning’s NAM model thinks for effective speed shear Sunday evening:


Numbers over 40 (knots) here support thunderstorms and severe storms. How likely are storms to rotate? Here’s what the same run of the NAM thinks for helicity (storm relative, indicating the likelihood for storms to rotate):


Number of 200 (m2/s2) here are significant, but we need other ingredients present. We need bubbles of air near the ground to rise rapidly if severe storms are to form; one way to do this is having a high low-level lapse rate, or a fast drop in the temperature going from the ground to a few thousand feet above the ground. Here’s what the NAM model thinks of that for Sunday night:


Those are low values, working against the thunderstorm and severe threat. How are the mid-level lapse rates looking?


These higher numbers are supportive of storms and severe storms if bubbles of air close to the ground are able to get higher up in the atmosphere. 

How about instability and layers of stable air aloft? Here’s what the NAM model thinks:


The warm colors are instability, and the blue colors are stability. If severe storms are to form we want a lot of the former and less of the latter. Instability is modest, and stability is generous here. This works against the likelihood of storms and severe storms, but these ingredients are less important in the colder months of the year and more important in the warmer months of the year. 

For lower-instability, higher-shear scenarios, I review what is called the SHERB parameter. As I discussed back in 2015:

While instability can often have a big influence on the chance for thunderstorms, it isn’t as important this time of the year. If thunderstorms are likely […], the SHERB parameter or index can be very helpful to a meteorologist in the colder months when looking a threat for severe weather. The SHERB parameter is helpful for getting a handle on a severe weather threat in the colder months because it focuses on temperature changes near the ground, lift in the atmosphere, and wind shear instead of instability (instability tends to be low in the winter even when we get severe weather).

Why is SHERB important? Unlike summer severe weather events which are driven by high instability and less of everything else, cold season events are driven by everything else and not often by instability. SHERB is a special blend of “everything else” that is important when gauging a severe weather threat…which makes it valuable when we don’t have summer-like heat and humidity. When SHERB values are high and the chance for rain and storms is high, severe weather is often a concern.

So what does the NAM model think of SHERB Sunday evening?


I’m looking for values of 1 or higher, which are focused northwest of Cincinnati. The 0.5- 1.0 values west of Cincinnati at this will likely drop slowly and translate east later Sunday night.

Wind speeds 5,000 feet or so above the ground are important, too; heavy rain can drag these winds aloft down to the ground. What does the NAM model think of these wind speeds?


These are significant, but not off the charts. If you’re of the math variety, these values are about 2-3 standard deviations above normal. This is enough wind to support storms and severe storms.

The positioning and strength of the jet stream is very important in the colder months of the year. Where does the NAM have the jet stream Sunday evening?


This is not ideal for thunderstorms and severe storms. The highest wind speeds are to the west (but still moving east) at 8pm Sunday night. Divergence (rising air) is in purple and positioned west and north if Cincinnati. What does Thursday’s morning’s GFS model say?


It has a lot more lift Sunday evening, and it has it stronger compared to the NAM model as it progresses east. There are clearly some strength and timing differences to resolve.

In summary, here’s what I’m thinking for Sunday night and early Monday:


This is still a wishy-washy threat at this point. We have good mid-level cooling, plenty of wind shear, and sufficient lift…but the lift timing and positioning are uncertain, instability and stability forecasts work against the storm threat, and we’ll need sunshine (even if filtered) to get the storm threat maximized.

There is plenty of time for conditions to change. Stay tuned!

Posted in Uncategorized | Leave a comment

Why NWS Wilmington’s Radar Being Down For 2 Weeks Is A Big Deal

Bad news. The radar operated by the National Weather Service in Wilmington – used as a primary weather radar for the protection of life and property in northern Kentucky, southeastern Indiana, and the majority of Ohio – is going to be down for 2 weeks:


This is a significant change in the time of completion compared to what was expected Monday:


In simple terms, the radar was supposed to be down through Thursday but now will be down for maintenance through August 16th.

The bull gear, the gear that drives the radar antenna, has failed. Thousands of pounds of equipment and a team of radar meteorologists will be imported to fix the radar and get it operational again.

This is a big problem. This was a scheduled upgrade meant to extend the life of the radar. Instead, the technicians have uncovered a major problem with it.

Why is this outage such a big problem? Because we still get thunderstorms in August! It is worth noting, however, that the severity of storms typically drops during the summer. The average number of Severe Thunderstorm Warnings issued in the Tri-State drops from 25 to 21 to 12 from June to July to August, respectively. The trend in the average Flash Flood and Tornado Warnings from June to July to August, respectively, goes from 7 to 5 to 4 and from 2 to 1 to 0. You may say that with the Severe Thunderstorm and Tornado Warning risk dropping that there’s no concern, but flash flooding and flooding has been a more deadly and damaging concern recently.

Here’s the NWS Doppler radar network when all of the radars are working:


And here the network when the Wilmington radar is down:


That’s a large lack of radar coverage over the Tri-State! So are meteorologists blind? Not entirely, but kind of.

There are still Terminal Doppler Weather Radars near Dayton, Columbus, and Cincinnati. The one covering Cincinnati is in southern Kenton County:



So no problem, right? The TDWRs are higher resolution than the NWS radars, so we’re good to go, right? Not so fast.

The TDWRs have a wavelength that is basically the same as that of a raindrop. This means that if heavy rain is falling near the radar, storms farther away from the radar at the same angle from the radar will appear weaker. The TDWR is also a single polarization radar, not a dual-polarization radar like the NWS’. The biggest benefit that comes from dual-polarization is for rainfall estimation. For the wavelength issues highlighted above, TDWR rainfall estimates are basically garbage. Distant dual-polarization radars can give rainfall estimates, but because these distant radars don’t sample as close to the ground as Wilmington’s radar, radar estimates from distant radars are not that great.

So the spring outage problem continues into August. One has to hope we avoid severe storms in the next two weeks and those who represent us can invest in brand new technology and stop putting Band-Aids on dated technology.


Posted in Uncategorized | Leave a comment