Sunday, May 31, 2015


So yesterday (May 30), and again on July 13 of this year (2015), the setting Sun will line up with Manhattan's roughly east-west street grid - an event dubbed Manhattanhenge (making the connection, of course, to Stonehenge).

Here's a picture I saw on my newsfeed this morning.

So why does this occur when it does?

As the year progresses, the rising and setting position of the Sun moves as well.  If you went out every morning at sunrise, and plotted the position of the Sun against the horizon each day, you would see the following.

Obviously, Manhattan's streets aren't aligned in a true east-west fashion or else Manhattanhenge would occur only on the equinoxes (March 20 and September 23, 2015) when the Sun is rising due east and setting due west.

The summer solstice this year is June 21.  The two Manhattanhenge dates (May 30 and July 13) occur 22 days before and 22 days after the solstice.  So what is the direction of the setting Sun on those dates?  These can be obtained from the U.S Naval Observatory Astronomical Data Services website (a great website for observational astronomy nerds like me, by the way).

For Manhattan, the azimuth (compass direction in 000°-360° where 000° is north, 090° is east, 180° is south, and 270° is west) direction of  sunset is 300° on both of these dates.

Let's take a typical cross-town street - the famous 42nd Street.  What is its western azimuth direction?  It's 300°, of course.

This works particularly well for Manhattan because of the clear view across the Hudson River toward New Jersey and the tall buildings framing the view.

It also brings up an interesting issue for archaeoastronomy - the study of astronomical alignments in archaeological structures (like Stonehenge). Manhattan's street grids were built according to the geographic alignment of the island and the Manhattanhenge effect is completely coincidental.  But what about ancient structures that appear to have alignments with the Sun, Moon, or other astronomical objects (prominent stars and constellations).

If you can find such alignments in ancient structures, were they intentional or coincidental?  That's not always an easy question to answer, I'll defer further discussion to the next post.

Thursday, May 28, 2015

Niagara Falls Erosion - Part II

In my previous post, Niagara Falls Erosion - Part I, I discussed the geologic setting of the Niagara Escarpment and how the average rate of erosion of the falls since the last Ice Age has been about 3 feet a year.  This is based on the total distance of movement of Niagara Falls from the Niagara Escarpment since the last Ice Age.

Today I'll look at this in a little more detail.  The first accurate survey of the exact position of the falls was done by the famous New York State geologist James Hall back in 1842.  In 1905, Grove Karl (G.K.) Gilbert, who worked for the United States Geological Survey (USGS), studied the rate of recession at the falls since Hall's measurements by resurveying the falls, and came up with an average rate of 5.3 feet/year (about 1.6 m/yr) - read his report here.

This rate, however, drastically slowed over the 20th century.  The building of huge hydroelectric power plants in both the U.S. and Canada means that anywhere from one-half (daytime) to two-thirds (nighttime) of the water in the river about Niagara Falls is diverted away for power generation.  Today, the average rate of erosion is estimated to be less than 1 foot per year (0.3 m/yr).

Here's a diagram showing erosion of the Canadian Horseshoe falls over the past few hundred years (both estimated and directly measured).  It's noticeably fast.

As you can see by looking at a map of the falls (or an aerial view), the reason there are two waterfalls currently is because the falls have intercepted Goat Island with the Canadian Horseshoe falls on one side and the American falls on the other.

Prior to 600-800 years ago (depending on the exact rates of erosion), there was only one waterfall on the Niagara River.  In a couple of thousand years, after erosion moves the waterfalls up past Goat Island, the two falls will again merge into one (really, what will happen is that the Horseshoe falls, which is eroding much faster will move past Goat Island leaving the slower American Falls behind, high and dry).

After about 15,000 years, the falls will have moved far enough upstream that it will lose the resistant caprock, only cutting through shale (Salina Shale).  This will turn it into a series of rapids and we'll lose the spectacular waterfall (well, we'll all be dead, but our descendants, assuming humanity is still around, will lose the falls).

Enjoy it while you can!

Tuesday, May 26, 2015

Niagara Falls Erosion - Part I

In my last post, I mentioned that the Army Corp of Engineers "turned off" the American side of Niagara Falls back in 1969 to deal with erosion issues.  So how fast do the falls erode?

Here's a cross-section of the geology at the falls.  There is one important (for our purposes) rock formation here - the Lockport Dolostone (also called Dolomite but I prefer Dolostone) which is underlain by a few different sandstone and shale units.

Dolostone is similar to limestone.  Limestone, a very common sedimentary rock, is made of the mineral calcite - CaCO3 (calcium carbonate).  Dolostone, on the other hand, is basically limestone with some magnesium substituting for the calcium in the formula (which creates the mineral dolomite) - CaMg(CO3)2.  This generally makes it a harder, more resistant rock than limestone.

Outcrop of Lockport Dolostone commonly seen in the area

The shales underneath are much softer and more easily eroded.  Here's a local example of similar differential erosion near Lockport, NY (Whirlpool Sandstone and Queenston Shale - see cross-section above).

Since the Lockport Dolostone is so hard and resistant, and is slightly tilted (dipping to the south), it doesn't weather as much as the surrounding shales and erodes to form a prominent cliff called the Niagara Escarpment.

The Niagara Escarpment runs over a large area from western New York, through Ontario, along the Upper Peninsula of Michigan, and then into Wisconsin (red line in map below).

We're only concerned with the escarpment where it crosses the Niagara River between Ontario, Canada and New York State (note the dipping Lockport layer).

The Great Lakes and Niagara River formed after the most recent Ice Age (the sequence of their formation is a bit complicated so I'll avoid that here).  Basically, Niagara Falls formed at the Niagara Escarpment around 12,300 years ago and has been eroding upstream since then carving the Niagara Gorge through the Lockport Dolostone.  The Dolostone ledge forms the resistant caprock to the falls (without this resistant rock unit, the Niagara River would descend from Lake Erie to Lake Ontario in a more gentle slope with no major waterfalls).

The American Falls pouring over the Niagara Escarpment

Since Niagara Falls is currently around 11.4 km (7.1 miles for those of us in the U.S.) from the Niagara Escarpment, and took 12,300 years to erode that far, how fast is the average rate of erosion here?

(7.1 mi / 12,300 yr) = 5.8 x 10-4 mi/yr x (5,280 ft / 1 mi) = 3 ft/yr

Three feet every year - that's pretty fast!  The problem, of course, is that this is an average rate of erosion over 12,300 years.  Why might this be a problem?  How about an analogy...

If I got into my car and drove to Niagara Falls right now (350 mi), it might take me 6 hours.  A reasonable assumption is that I took the New York State Thruway and traveled at 58 mph (350 mi / 6 hr).  But maybe instead I drove at 70 mph for 5 hours and spent an hour leisurely eating dinner at a rest area.  Or maybe I drove at 90 mph for 1 hour, got pulled over for a ticket for a half hour, and then drove 58 mph the rest of the way paranoid about getting another ticket.  Average rates don't always tell us specifically what was going on during each moment of time.

So, what was going on during the past 12,300 years?  Was Niagara Falls eroding at a constant rate the whole time or has the rate varied?  We'll examine that in my next post.

Monday, May 25, 2015

Turning off Niagara

Back after a bit of a hiatus now that the spring semester has ended.

A few days ago, my family stopped at Niagara Falls after returning from a geology conference in Madison, Wisconsin (we passed through Canada coming back).

They are impressive falls (two really - the American falls in the foreground and the horseshoe-shaped Canadian falls in the background - separated by Goat Island).

There's a lot of neat stuff I could write about concerning the geology of Niagara Falls but I'll just concentrate on one for today - the time they turned off the American Falls.

They actually slow down the falls each night.  Ontario Hydro and the New York Power Authority pump tremendous amounts of water each night from the Niagara River upstream from the falls to top off their reservoirs.  They do this at night because this action reduces the flow over the falls from 100,000 cubic feet per second (CFS) to 50,000 CFS (half).  During the day, the flow is sort of left alone so the millions of tourists have impressive waterfalls to admire.  By "sort of", I mean that water is still normally diverted for hydroelectric generation 24/7.  Without this diversion, the flow over the falls would be a more awe-inspiring 200,000+ CFS!

Flow over Niagara Falls has actually been stopped (at least partially) three times in recorded history.

The first event was March 29-31, 1848.  People living near the falls noticed an eerie quiet on the morning of the 30th.  When they went to look, they saw a dry riverbed with fish and turtles flopping around.  The falls had maybe 30-40 CFS - a mere trickle.  People were able to walk on the riverbed and collected artifacts from the War of 1812 (muskets, bayonets, tomahawks, and the like).  Special church services were held on both sides of the border for anxious people who didn't understand why the river suddenly vanished.

The cause of this?  Strong winds formed a massive ice jam at the mouth of the Niagara River at Lake Erie.  On the evening of the 31st, however, the jam broke, people heard a low rumble, and a wall of water swept down the dry stream bed refilling the river and restarting the falls.

The next event was man-made.  In 1953, some coffer dams were built exposing only a section of the Horseshoe Falls nearest the Canadian side (where the big tourist center is today). This allowed engineers to stabilize this section of the edge of the falls which were, of course, eroding as falls are wont to do.

The last time the falls were turned off was also a man-made event.  The Army Corp of Engineers built a 600 foot coffer dams to Goat Island which turned off the American Falls (the water was diverted over the Canadian falls) starting on June 12, 1969.  All part of the Corp of Engineers never-ending struggle to control nature and stop erosion (an ultimately impossible task - see John McPhee's The Control of Nature).

Tourist attendance at the falls topped all-time records to see this - they even allowed people to walk out onto the dry river bed.  The Corp of Engineers injected concrete into the bedrock fractures and inserted a bunch of rock bolts over a period of about 5 months.  Once "repaired", the falls were restarted on November 27.

In my next post, I'll talk a bit about erosion at Niagara Falls.