Saturday, August 9, 2014

Sky Brightness 2

In my previous post I began to analyze the data from the sky brightness meter at Ladd Observatory. Now we'll take a closer look at the broader trends. Here is a scatter plot showing the data from the summer and fall of 2013. The plot is a little busy but we're really only interested in the "bottom line" where the data points are at the lowest values. All of the nights are superimposed on one another with the x axis showing hours UTC. This graph summarizes how the sky brightness changes during the course of the night. The many values between 3.7 and 4.3 are due to nights that are more or less hazy. There moisture in the atmosphere scatters light from the city back down to us and causes the overall sky to look brighter.

Sky brightness scatter plot

If we follow the lowest readings there is a definite trend where the clearest nights start off at about 4.2 at the end of twilight and slowly, steadily, decrease to about 4.45 at 4 hours UTC. There is then a small but rather sudden drop to 4.55 after which the slow decrease continues until we are at about 4.6 in the early morning. I'm not sure what is causing the drop at 4 hours but it may be due to city lights that are on a timer. The takeaway here is that the sky is slightly, but significantly, brighter in the early evening. The best time to observe is after midnight local time through the early morning.

This next graph takes a look at only the "good nights" where the sky is darker. Any night when the nelm (naked eye limiting magnitude) is brighter than 4.0 is likely too cloudy or hazy to see many stars. Each downward spike in the graph represents one night and the lower the spike the darker it was that night. The most common values that we see are between about 4.0 - 4.3 on mostly clear nights. I've inserted a horizontal dashed line at 4.3 to divide the data between moderately dark and darkest. Look at the number of spikes that drop below that line.

Sky brightness graph for 2013-2014

One interesting thing to note is that there were quite a few dark nights during the latter half of 2013. There is then a period starting in mid December where the dark nights get fewer and farther between. By February they completely disappear. I'm not sure what is causing this but the absence of dark nights seems to coincide with the polar vortex weather pattern that occurred in New England the past winter. This summer we've had some dark nights around June but overall we're not seeing nearly as many as last year. This confirms that observing in Providence really has been worse this year than what we were seeing in the latter half of 2013.

1 comment:

  1. In regard to that feature at 04:00 UTC, perhaps you could separate the data based on whether standard time or daylight savings time is in effect. Comparing the two graphs might show whether the feature is correlated with UTC (astronomical) or local time (human made).