Wednesday, June 3, 2020

Lightning Season

Thunderstorm activity has sprung to life in parts of Alaska in the past several days, but somewhat unusually the lightning has been focused in the southwestern interior and near (and over) the Norton Sound.  Historically only about a quarter of Alaska's lightning occurs west of Tanana.  Here are maps of lightning strikes from the past few days (click to enlarge):

Saturday May 30

Sunday May 31

 Monday June 1

Tuesday June 2

The onset of lightning season is right on schedule, as the historical data show a very rapid increase in Alaska's lightning activity at the beginning of June.  See this 2016 post for previous comments on this:

The outbreak of thunderstorms also signals that humidity has risen to a level high enough to support widespread deep convection (atmospheric thermal overturning); moisture is a key ingredient for thunderstorms in general.  To explore this idea in more detail, I looked at the historical data to see if there is a close link between rising humidity levels and the first widespread lightning activity in Alaska.

To measure humidity, I used the average of daily mean dewpoint at McGrath, Fairbanks, and Eagle, and I smoothed the daily values over 3 days to remove some of the daily noise.  Using data from 2000-2012, I then compared the peak year-to-date dewpoint with the peak year-to-date daily number of lightning strikes; the hypothesis is that there might be a threshold value for dewpoint that, once reached, allows thunderstorms to suddenly become widespread.

Results are shown below.  If the chart is a little confusing, consider the black line (the year 2000): the data show that there was essentially no lightning until the 3-station dewpoint index reached 40°F, but then 6600 lightning strikes were observed in one day when the dewpoint index first exceeded 42°F.  Of course there's no time dimension in the chart, and it doesn't deal with multi-day lightning totals, but I think it's quite illuminating nonetheless.

The results indicate that widespread lightning (more than 2000 strikes per day) tends to be uncommon until the dewpoint exceeds 40°F, but it's also rare to avoid widespread lightning when the dewpoint index reaches 45°F.  This suggests there is a fairly narrow range of humidity that produces the first outbreak of widespread thunderstorms in Alaska; but I'll admit this could be partly a case of correlation without causation, as it may be that humidity and thunderstorms increase simultaneously without one (solely) causing the other.

As for this year, we're right on track once again; here is the 3-station dewpoint index from the past several days.

May 29   38.6°F
May 30   37.9°F
May 31   39.1°F
June 1   42.0°F
June 2   44.4°F


  1. Excellent analysis Richard. May I suggest it also takes a pre-seeding of lower atmospheric humidity to support low to higher level lifting (the CAPE or whatever) to create these events. It might be a day or two before the chemistry evolves. Examine prior weather influx followed by solar heating.


  2. I might also add that as we approach the solar maximum in later June to July it takes a lack of daily cloud to allow the sun to create the thunderstorms. Solar angle with low albedo and a clear shot at dark heated humid earth early in the day is most productive.


  3. Thanks Gary. I agree. Traditionally the 3 ingredients for deep moist convection are (i) conditional instability, i.e. sufficiently large vertical temperature gradient (conditional refers to saturated vs non-saturated displacements), (ii) moisture, and (iii) lift to initiate, often solar heating.

    So yes, there's a pre-conditioning of the thermodynamics and then typically a daily cycle of initiation.

  4. When I lived in the Phoenix area a few decades ago, I noticed that thunderstorms would refuse to form until the ground temperature at my house was at least 68°F. Only then would any cumulus clouds over the hills north and south have any chance for explosive growth. I expect something similar is happening here in the Interior. Though Alaska is not a traditional rock and sand desert like Arizona, the physics are all the same.

    1. Interesting observation, Eric. There aren't too many similarities between Phoenix and Fairbanks, but perhaps this is one!