Understanding Pollen Data


Finding Good Information

My friends and acquaintances know that I am a bit of a pollen geek. Every spring, at least a few people will come to me with blood shot eyes and a runny nose and ask, “What is in the air right now? It’s killing me!” They ask because they are lucky to know me, but what if you weren’t lucky, and didn’t know me, what would you find if you did an internet search?

You can see in the screenshot below what I found:pollen data search screen grab

So let’s talk about the results and the kind of pollen data out there right now. There are basically two categories: Actual Counts and Modeled Counts.

Actual Counts

When someone asks me what is in the air, I usually know off the top of my head what’s probably in the air, but I like to be accurate, so I usually check the local National Allergy Bureau (NAB) station counts on my phone and give them the best information available. The closest station to me is 30 miles away and is run by Intermountain Allergy and Asthma.

So what is the NAB and how do they count pollen levels? The NAB is an organization that has reported 24 hour average pollen counts to the North, Central, and South American public since the 90’s. There are similar organizations in many European countries. These stations, run by allergists and researchers, are pretty intense by modern standards: every day, someone at each station collects a 24 hour sample, mounts it on a slide, stains it, and counts/identifies pollen one by one using a microscope. So a 24 hour average reported at 10 am (for example) is basically telling you what conditions were yesterday. It’s essentially yesterday’s count because pollen peaks in the middle of the day. This is great, this is actual information and it can help you, but it doesn’t necessarily tell you what is in the air right now. I’ll talk later about ways to get the most out of these counts.

Modeled Counts

All of the search results in the screenshot above are modeled counts except “Utah Pollen Count|Intermountain Allergy and Asthma”. So what is a modeled count? Modeled counts work like a weather forecast, or more accurately, they use weather and actual pollen count data to predict what the pollen count is and what it will be. Yes, I know it sounds weird to “predict what the pollen count is” but that’s exactly what they do. There aren’t any manual counters reporting very current pollen counts. The best real sensor data they can get is the average count (NAB) from the last 24 hours, which has little to do with right now.

If you want a prediction about the conditions right now, you have to use algorithms, fancy equations that use yesterday’s counts and what we know about the relationships between pollen levels and weather and seasonality to predict counts. Imagine if weather worked like that! Your weather person would say things like, “Yesterday the temperature was 35 degrees, based on the amount of cloud cover, the time of day, and the time of year, we think it is about 30 degrees.” Modeled data can’t do much to help you right now, but it can train you a bit about the relationship between pollen counts and weather conditions. Here are some simple concepts: If it’s raining, there’s probably no pollen (although there is such a thing as thunderstorm asthma). If it’s sunny, warm, and spring, the pollen count is probably high. The forecast is not worthless, but the “current conditions” specific to you are often less valuable. There are even studies that demonstrate that modeled pollen counts are not very accurate.

Automated Pollen Counts

Automated pollen counts are available now. They are created by our device using state of the art technology from Pollen Sense. This is the future, the Pollen Sense app will allow you to check pollen levels that are reported up to every two minutes or so. Pollen Wise, our app (currently in active development), will tell you if the conditions are favorable to go for a run, walk the dog, or even just open your window.

Get one for your school, neighborhood, HOA, agency, or medical office.