Estimado Sr. Guaraglia,

I am not aware of fog detection systems in Latin America but our Sentry
Visibility Sensor is being used in fog detection systems around the US, Asia,
Europe, and the Middle East. I will be glad to discuss your fog detection
requirements in Argentina. Please email me directly at
jcrosby@envirotechsensors.com.

Atentamente,

John Crosby

George made an excellent and informative e-mail reply and gave me permission to
copy it to the National Highway Visibility (NHV) Community of Practice (CoP)
at knowledge.fhwa.dot.gov/nhv:

-----Original Message-----
From: George Small [mailto:GSmall@sha.state.md.us]
Sent: Friday, August 29, 2008 5:42 PM
To: dguaragl@ing.unlp.edu.ar
Cc: Tom Hicks; Bob Fisher; Chandler, Mark; George & Nancy Small
Subject: RE: Argentina Fog Issues
Importance: High

Dardo. I hope you did not think I forgot about you or fell off the end of the
earth. Things here have been so busy and a lot of my staff has been either sick
or on vacation that we just could not get around to replying. I do apologize
for the delay.

I want to go down your list and answer some of you questions in order so it
keeps things straight. I will also give you some contact information that you
asked for as well as some additional comments based on the new information you
have shared with me on the roads in Argentina.

1) There are a lot of camera manufacturers around the world that have devices
that could meet your needs. You just have to be sure the do what they claim. We
used equipment from a company called - Quixote Transportation Technologies Inc.
They are part of a larger group called SSI (Surface Systems, Inc.) The web
sites are www.qttinc.com and www.ssiweather.com. SSI basically supplied the
entire sensor package since we tied into existing roadway weather sensor
stations and then up graded them to visibility sensors. Another contact might
be Enviro Tech Sensors, Inc. Mr. John Crosby. jcrosby@envirotechsensors.com.

Another company that makes very good visibility sensors is Belfort Instruments,
located in Baltimore Maryland. The contact there would be Dr.Mark Kukucka.
There web site is www.belfortinstrument.com. Their email address is
sales@belfortinstrument.com. If you mention my name Mark may remember meeting
me in Madison Wisconsin. He is very helpful as well as the people at SSI. I am
sure if you wrote either on they would be happy to give you all the information
you need and more. I am going to stay away from costs and model numbers as
there are many options and prices have definitely changed. We did include some
specific information in the reports I sent. I could also provide more specifics
on all our units if you would like.

2) The question you pose here can be addressed to the people noted above. I can
tell you that our cameras and visibility sensors work in the most extreme of
conditions; Sub-freezing to high heat; rain; blizzard conditions and fog, of
course. There are no heat elements in the system.

3) Our skip line markings are 10' lines and 30' gaps (3m and 9m). When I
mentioned only being able to see several skip lines this equated to a
visibility of 15.2-30.5m. You might think that you could see farther than that
but I can attest to the fact that there are some times you can not see past the
front of your vehicle. I have driven in those conditions. It is not nice.

4)In extreme fog conditions the rear lights are barely visible as the light
from them is disbursed by the water droplets. Some rear lights are brighter
then others assuming that they are even on and you could possible see the rear
of a car ahead farther then you could see the road markings ahead. If you have
your headlights on the condition is worsened and totally blinding if the high
beams are on. The light reflected back totally destroys your ability to see
ahead. While this may seem strange, a lot of people turn their headlights off,
especially at night in dense fog and only use their fog lights that are much
lower to the ground and do not create the same glare.

In the US, driving directions tell you that if your vehicle brakes down you
should move to the side of the road (shoulder) and put on you 4-way flashers.
Large trucks are told to use their flashers when they are moving much slower
then the speed limit. One problem that this causes in blinding fog is that when
you come up on a vehicle with flashing lights you don't know if it is in the
travel lane or shoulder.

Comments on driver behavior.

1) This is certainly a big factor in the whole situation and has been studied
by many Human Factors Engineers whose whole careers are dedicated to
understanding driver behavior. One thing I may have mentioned before, and is
the cause of the phenomenon of higher speeds in fog, is the lack of reference
points to a driver. Drivers over driving their visibility range is the main
cause of crashes, in my opinion. In normal day or night driving conditions we
can see reference points passing by like trees, poles, signs, delineators and
other vehicles.
From this we get a feel for how fast we are going, In fog you lose these visual
queues, thus you have a tendency to drive faster.

Another issue is the fear of being run over from behind. You are traveling at a
speed you feel comfortable with given you particular driving abilities, vision
and other factors. All of a sudden another car or truck speeds by and you know
he can not see any further ahead then you. That is when you begin to think, if
I don't go faster someone is going to run over me. So you increase your speed
even though you can not see vary far ahead.

2)When we experienced all of our fog crashes it was during holiday traffic flow
conditions where the report for our study was done under normal traffic flow
conditions. That said, during heavy traffic flow we might be running at a LOS D
or worse. Traffic is literally bumper to bumper and was when we had our big
crash. In these conditions there is little measurable headway. The only
constraint is the vehicle ahead or in the lane next to you. If you have spotty
fog conditions as we do then there is great visibility up to the fog bank then
almost zero visibility. Once up stream traffic begins to slow as they get into
the fog the down stream traffic sees no reason to slow and enters the fog at a
high rate of speed only to discover unseen traffic at a standstill...the result
is a multivehicle crash with approaching traffic continually piling into the
end of the line.

This is the reason we have gone to providing notice of fog conditions well in
advance of the actual fog area in hopes that drivers will adjust their speeds.

The last statement you made about not understanding why we drive at 65 mph when
you can only see ahead 150' is at the heart of the whole issue.
If we had the answers to this and could have an effective, positive way of
altering driver behavior we would all be rich.

Now that you have shared the fact that you are dealing with two lane roads in
Argentina (which we to have here in the mountains) let me share some other
traffic engineering measures that might be helpful and that we use frequently
here in similar road conditions but not specifically for fog.
1. Center line rumble strips. These are grooves made in the pavement on the
centerline that run perpendicular to traffic. These cuts are 1/2" to 5/8" deep
7" wide and spaced 12" c-c. I will try to send you a typical detail later if
you wish. These are very affective in preventing cross over crashes.
2. Edge line rumble strips. Same as above and are very effective in addressing
run off the road crashes.

Both of these are inexpensive to do and provide a tactile sensation and audible
sound when they are run over. It grabs your attention fast but also can give
you valuable information in the fog when you have strayed in either direction
from your travel lane. Vandal proof also. Our snow plow drivers have found them
useful in blizzard conditions to help locate the shoulder of the road.

3. Recessed Pavement Markers. These are reflective markers placed in a groove
cut parallel to traffic. For two way traffic we use a bi-directional yellow
marker set in the center of the groove. Vandal resistant. You could dig them
out if you wanted to.

4. Delineators. We have gone to a larger reflective surface running down the
entire flat surface of the plastic post. Normally the reflective area is about
6" but in our fog areas we have a strip that is about 1' long. These would not
be vandal proof.

All of these suggestions are much less expensive then a fog warning system and
might be something you could suggest trying in the interim.
You may find it very effective.

For more information on similar road conditions and other ideas you might go
back to posting a question on the FHWA web site where you first got my contact
information from.

I know I have given you a lot to go over. Hope I did not over do it. I will be
away for the next two weeks on vacation. By then you may have more questions.
Look forward to hearing from you as to how you are making out with your project.

George Small
Assistant District Engineer - Traffic
Maryland State Highway Administration District -6
1251 Vocke Road
La Vale, Maryland 21502
Office - 301-729-8440
Cell - 301-707-8004
Fax - 301-729-6968










-----Original Message-----
From: dguaragl@ing.unlp.edu.ar [mailto:dguaragl@ing.unlp.edu.ar]
Sent: Saturday, July 26, 2008 7:19 PM
To: George Small
Subject:


Hi George,

First of all I want to congratulate you by the job done. It is a good job!, you
have been able to build a detection and warning system at a very low cost. Not
an easy task!

I saw a document on the system built for the I-75 in South Georgia that cost
1.7 million dollar to be replicated. I wander what would you do with such
amount.

I read the documents entitled:
I-68 Fog Detection System Planning Report.
I-68 Reduced Visibility Fog Detection and Warning System: Evaluation Report.

I have some questions and comments about the above mentioned documents.
Let me begin with the questions:

1) What is the information you could give me (manufacturer, model,
cost, etc.) about the CCTV camera and the visibility sensors used in your
project?.
2) Is it possible to use the CCTV camera images as ?ground truth?
disregarding the weather conditions?. I mean, When the weather is very cool or
frozen, is not necessary to heat the camera lenses to avoid building up of
moisture on them?
3) When in the second of the above mentioned documents you say
?being
able to barely see one or two skip lines? I am not sure what distance is this.
I think that the distance between skip lines changes from one country to
another, then it will be useful to me to have this estimation in feet or
meters. Have you a raw idea how much is this distance measured in feet or
meters?
4) In the case mentioned in question 3 (barely see one or two skip

lines), have you a raw idea about what is the distance at which the back red
lights of a car can be observed? Intuitively, I think that due to the emission
of the incandescent bulb light, it should be a longer distance than that of two
skip lines visibility. This could explain, in part, the driver?s behavior.

Trying to understand the driver?s behavior I have some hypothesis that I would
like to consult with you.

1) Perhaps, even when the camera shows only two skip lines, the
back
red lights of the precedent car are observed from a longer distance.
Then, the driver feels safe. Perhaps, the visibility for a precedent
car is far beyond that of the camera. 2) From the photos I
understand
that the I-68 is two tracks (one upstream and other downstream) with two lanes
on each track. I observed that the maximum count for 15 minutes was about 150
vehicles, what gives 6 seconds of headway (time between
vehicles) for one lane. But because you have two lanes, we could say that ?in
average? you have 12 seconds of headway. Perhaps, due to the relative ?average
long time between vehicles?, the driver does not perceive the low visibility as
dangerous (?).

Well, these are all my hypothesis. I cannot understand that if the driver do
not see more than, perhaps, 150 feet it could keep running at
65 mph!!!.

In most of the roads of Argentina we have only one track with two lanes (one
lane upstream and the other downstream without physical separation, only skip
lines). Because our train system has been dismantled, we have many trucks on
the roads. Do you know a similar dangerous situation in USA with frequent (let
say six a months) fog events? I am very interested in knowing how to design a
fog warning system for this difficult case. Any suggestion, contact,
bibliography, will be welcome.

I hope that you will have time to answer this letter and let me keep in contact
with you for future consults.

Best Regards,

Dardo.