Tracking the Weather and Hurricane Sandy with Citizen Science - Part III

Hopefully this posts finds everyone safe.  Hurricane Sandy walloped us with a big punch in the DC area, but the storm has pretty much passed and the cleanup can begin.  So let's look back at what happened as told through the eyes of our backyard weather station.

First we have a nice chart of barometric pressure (green) as well as average wind speed (blue) and wind gusts (red).  Wwe can see the slow but steady drop in pressure as Sandy nears.  As the pressure drop accelerates we can also see how the wind speed start to ramp up.  Winds near their peak strength at the lowest pressure point (28.70 in/Hg) and then quickly drop as the pressure increases.  So once the storm actually hits and starts moving away we immediately see a drop in storm intensity.  The time markers weren't able to display in the actual image, but the sharp decline in pressure started around 9:00 AM with peak wind/lowest pressure at 9:40PM.  I can't claim an end time since the storm had not fully dissipated when the data was downloaded, but certainly things have abated a lot.

Relative Pressure, Wind Speed, and Wind Gusts During Hurricane Sandy
Image Courtesy: OpenScientist

Now let's look at rainfall.  For the length of the storm (until I downloaded the data) we received approximately 4.8 inches of rain.  I was actually expecting more but this is in line with predictions.  Most of the time the rain was not overly intense...averaging under 0.2 inches per hour.  But we do have a big surge of rain (1.1 inches per hour) around 9:00 AM on Monday.  That is the same time as the barometric pressure really starts falling and the storm is now being felt.  So it's interesting how the actual moisture falls the most at the beginning.  However, excluding that one big surge, we also see a clear overall trend where the hourly rainfall generally increases/decreases in reverse proportion to the barometric pressure (in other words, one goes up as the other goes down).

Relative Pressure and Hourly Rainfall During Hurricane Sandy
Image Courtesy: OpenScientist

Looking beyond the scientific part of the hurricane, the data also tell a very human story too.  Using the weather station's sensors inside the INDOOR control station we can measure temperature inside our house as well as the weather outside.  The first thing you'll notice is a drop from 80 degrees to 74 degrees...this is me taking the console out of the box and letting it start record.  It was previously in a box next to a heating register and had to cool down a bit before registering room temperature.  There are then a few blips while the heater flips on and off to maintain a constant temperature in the house.  Still quite normal.  Then we see a big spike...that's around 6:00 when my wife started a fire in the fireplace.  Perfect for staying warm on a stormy night.  But then a problem hits right at 10:10PM Monday night (indicated by the red line)...that's when our power went out and we lost heating.  Temperatures keep dropping unchecked until 3:00AM earlyTuesday morning...that's the green line when PEPCO restored our power.  Temperatures rise as we stay snug in our beds until it hits 72 degrees; the new level set by my wife after it almost felt too hot.

Taking this one step further, compare the times between when we lost and regained power on the other charts (signified by the yellow lines).  This coincides with the period of strongest wind gusts.  Right up until 3:00 when after the wind drops down sharply...only then could power crews get back to work and restore our power.

So not only do we get interesting scientific data, it also tells the story of electric power and how quickly it's loss can be felt inside the home.

Internal Home Temperature During Hurricane Sandy
Image Courtesy: OpenScientist

Finally, there is one other interesting thing from all this data.  Look at the wind speeds again in the first graph.  Notice how even during the biggest increase at the height of the storm, the average wind speed was never more than 5.4 mph and the wind gusts were never more than 9.2 mph.  Very low for a hurricane.  What's happening here?

My first guess was the siting of the weather station...but I kept it away from the house and nearby tress as it sat at the edge of my patio near the middle of backyard.  So that's not it.  My second guess was it was too low...but the patio is raised a few feet above the yard itself and the station is a foot above the actual stand.  It could be a broken gauge...but I watched it spin quickly during large gusts and slow down during low times.  None of those could be it.  One clue I have was sometimes a big gust would come along and greatly shake the large (60-80 foot tall) trees around my house.  At this same time, however, the shorter trees (10-20 feet tall) would not be shaking, their leaves would not be falling off, and the wind gauge would only be spinning at a moderate pace.  I also noticed that my neighbor had left many small pots and lightweight patio furniture outside during the storm, but none was picked up or thrown around by the wind.

This leads to two theories:

1. Wind gusts can only be felt much higher in the sky than the first 20 feet or so.  But this doesn't make sense when walking the streets of watching normal hurricane footage.  Clearly high winds are at ground level too.
2. Strong winds and strong wind gusts are greatly mitigated in a forest setting.  My house is in a relatively wooded area and is surrounded on most sides by large trees and many smaller trees.  Could the winds be hitting this foliage and being somewhat neutralized?  Are they creating a wind barrier for my house?

Out of sheer personal interest I'm really curious whether hypothesis number 2 has any real merit.  So it's time to hit the library and maybe set up some additional experiments to test it.  All that will be in a future set of blog posts, so stay tuned to this page!



FOR MORE ON THIS SERIES SEE:

• Part I - Choosing a Home Weather Station
• Part II - Installing a Home Weather Station
• Part III - Analyzing Hurricane Sandy Storm Data (today)