Jun 252013
People observing the nearby Small Boy test in the NV desert, 1962.

People observing the nearby Small Boy test in the NV desert, 1962.

This is the first part of a three-part article that looks at how to calculate the potential impact on your retreat location of the release of radioactive material somewhere else.  In other words, if a nuclear explosion goes off, or a nuclear power plant has an accident, some distance from you, how will that affect you at your retreat?

After you’ve read this first part, please follow the links on to parts two and three, including some worked examples and links to the resource materials you need to do your own calculations and planning.

Once you have started to zero in on an area for your retreat, you then start considering reasonably local issues such as the potential risk of radiation/fallout from nearby nuclear power plants, in the unlikely event something might go wrong with them, and similarly from any probable nearby targets in the event of nuclear war.

Sometimes you can just simply look at a map showing potential risks and guess as to if you’re safe or not.  You don’t need to do too many calculations to understand that downtown DC is probably at great risk of some type of nuclear event, if/when TSHTF.

Unfortunate, there’s no opposite situation where you can instantly tell by a quick glance at a map that any given area is clearly completely safe.  Plus or minus a hundred miles or so, nowhere in the US is more than 250 miles from either a nuclear power station and/or a possible nuclear target in a high intensity conflict.

Although 50 miles is more than enough distance to survive the immediate heat, blast and radiation effects from a nuclear explosion, the fallout radiation implications can be very different indeed.  As we mentioned in our article about the safety of nuclear power stations. dangerous radiation levels as a result of the Chernobyl disaster occurred in what seemed to be a semi-random manner as far as 300 miles away, and radiation from the Fukushima disaster made it all the way across the Pacific to the US.

Note our italics for the phrase ‘what seemed to be’ – because, of course, there is underlying scientific sense that applies to the observed fallout pattern from Chernobyl.  This article series talks about these issues.

Radiation – A Quick Refresher

We discuss the five most common types of radiation in detail in our article on Radiation and Fallout Risks.  Radiation itself is generally both extremely short-lived and also usually only capable of traveling short distances and easily blocked by most things (quite the opposite of public perception).  There is one notable exception to the easily blocked issue – neutron radiation, which while limited in distance to several miles, can penetrate quite a lot of shielding during the course of its travel.

The initial release of radiation from a nuclear explosion is very intense, but only harmful to people within a few miles of the explosion.

The bigger problem is that as part of the nuclear explosion, some of the radiation released will react with non-radioactive materials, and make them become radioactive in the process – that is, they will then start steadily emitting radiation for some time into the future – maybe only hours, but possibly also days, weeks, years, or even centuries.

The second part of this problem is that the heat and blast of the nuclear explosion will turn whatever these things are into finely powdered dust/sand/dirt type particles and then, via the blast wave, propel them long distances.

These tiny particles of now radioactive dust are what is known as fallout and represent the biggest problem that most people will encounter as a result of a nuclear explosion.

So for us, hopefully having located ourselves safely away from the immediate effects of the initial heat, blast, and radiation burst from the bomb, our concern is to do with the fallout.  As you know, normal dust can go anywhere and get anywhere, and if there is radioactive dust falling in your area, it will have the same ability to spread out and cover surfaces, to hang suspended in the air, to be kicked up into dust clouds when the wind blows, and so on.

Furthermore, while the radiation it releases might be weak, there’s one huge danger, and one more insidious danger.

The big danger – if you ingest the dust – by breathing, or if the dust gets onto food or into water – you then have radioactive substances inside you, where there is no distance and no barrier protecting you and your organs from the full effects of the radioactivity.  Even the weak radiation that can only travel a few inches and be blocked by a sheet of paper is now hitting and harming your internal tissues and organs.

The more insidious danger is that low levels of radiation being emitted from fallout may also be fairly long-lived, and so over a period of months (or years) their effects will start to accumulate and cause problems.

So you really want to avoid being in an area with radioactive fallout.

Three Types of Nuclear Explosion

When evaluating your fallout risk, you need to know whether the initial ‘event’ that creates the radioactive fallout will create a little or a lot of fallout, and whether this fallout will be propelled way up into the jetstream or if it will stay closer to the surface.  The ‘jetstream’ is (are) fast-moving bands of air, situated somewhere between about 22,000 ft and 52,000 ft above the ground, and traveling at speeds of anywhere from about 60 mph up to sometimes in excess of 250 mph.

How do you determine the answers to these questions?  You need to decide if the nuclear blast will most likely be an airblast, a surface blast, or a sub-surface blast.


Airblasts are the most common sort, and happily create the least amount of fallout, because the fireball from the blast is mainly above the surface of the ground, and doesn’t impact on as much ‘stuff’ to vaporize and transform into radioactive fallout.  Even more happily, the fallout they do create tends to go up into the upper atmosphere, where it might get caught in the jetstream.

If the fallout gets into the jetstream, this will quickly whisk it away from the site of the blast, and by the time the particles start to drop out of the jetstream again, they may have traveled anywhere from a few hundred miles to many thousands of miles.  The good news is this means the fallout gets dispersed over a very wide area.  The bad news is that while you might not be at risk from a nearby blast, you may get a small amount of fallout from far away blasts (for example the jetstream took fallout all the way from Japan to the US after the Fukushima power plant problem).

Even if the fallout particles from an airblast do not mix into a jetstream, the simple fact of throwing them up a long way means they’ll spread out over a greater area than if they were not thrown up as far to start with.  You can see this for yourself with a simple thought or real experiment.  Take a handful of rice, hold it a couple of inches above the floor, and then drop it.  See where it lands and how it spreads out.  Now, after cleaning up your mess, take a second handful of rice, toss it as far into the air as you can, and see where it lands.  It will be much more spread out, won’t it (good luck with the cleanup!).

Surface Blasts

These are the nastiest of the three types of nuclear explosion.  The initial fireball from the blast is, by definition, right on the ground (or within 100 ft or so of the ground), and will be vaporizing literally tons of material – buildings, dirt, people, anything and everything.  A massive amount of fallout is created.

Much of this fallout does not get thrown as far up into the atmosphere, and will fall back down to earth without reaching the jetstream.

So a surface blast has two severe consequences (compared to an air burst).  The first is the creation of massively more fallout material; the second is that this material is not distributed thinly (and almost safely) over a wide area, but is concentrated in a dense and potentially lethal area, within a few hundred miles of the explosion.

Subsurface Blasts

If the enemy is trying to destroy hardened bunkers, and possibly missile silos, they may use bombs that are designed to penetrate a distance into the ground before exploding.

The best case scenario would be a bomb that traveled so far into the ground before detonating that the earth above it contained the full effects of the blast.  This is what happened most of the time with underground testing of bombs – sure, you’d see the surface rise and blister some, and the shockwave would cause dust to rise from the surface, but it would be non-radioactive dust, and if the calculations were correct, the covering of earth above the bomb would contain the force of the explosion.

However, it is unlikely that missile penetrators would go that deep.  A shallow subsurface blast might actually be even worse than a surface blast, because more of the fireball will interact with material rather than the top half of it less harmfully going upwards into the sky.  A bit deeper and the effects would be similar to a surface explosion, and deeper again and the amount of fallout sent up into the air would start to reduce, as would the height it reaches and therefore the distance it travels.

Nuclear Power Plant Failures

In the case of a nuclear power plant failure, it is likely that the failure will not be akin to a nuclear explosion, but rather it will be some sort of secondary effect caused by runaway heat causing fires and steam buildup (which might possibly then burst through a containment external barrier with explosive force) and the effects of fire and heat and steam releasing radioactive material.

This is most likely to mirror the effects of a surface or subsurface blast, although we again look at the Fukushima event and learn from it – the explosions it experienced sent some radioactive material all the way up into the jetstream.

At the risk of massively oversimplifying, we’re going to say ‘forget about radioactive material that makes it up into the jetstream’; because most of that material goes anywhere and everywhere.  Having happily forgotten about that, we still have to face the remaining reality – ie, that nuclear power plant failures risk releasing potentially large amounts of radioactive fallout type material that will be deposited in the several hundred miles around the plant.

What Type of Blast to Expect

As a quick rule of thumb, attacks on civilian structures will be air bursts because they create the most blast damage over the greatest amount of surface area.

Attacks on hardened (ie military) structures are more likely to be ground or subsurface blasts, because an airburst may not be sufficiently strong to destroy them.  If it is simply an attack on a place with massed troops or equipment, it will probably be an airburst.  Only if the target has unusually strongly constructed buildings will an enemy feel the need to transition from an air burst to a ground burst.  Unusually strongly constructed buildings might include ammunition bunkers (especially if likely to contain nuclear munitions) and some types of airplane protective hangar.

Attacks on bunkers and other underground facilities will definitely be sub-surface.

Continued in Parts Two and Three

Please now click on to the second part of this three-part article series.  Part two explains how the two different types of winds will interact with the fallout and how we can calculate the effect they will have on our retreat location.  The third part then looks at sources and types of wind data you can obtain to work out the impact on your retreat from radiation releases in other places.

Jun 252013
Part of a diagram explaining how jetstream winds are formed (full size here).

Part of a diagram explaining how jetstream winds are formed (full size here).

This is the second part of a three-part article series on using wind data to evaluate the risk of nuclear fallout at your retreat location.

If you arrived here directly from a search engine or link, you should probably go to the first part of the series that discusses the different types of fallout patterns and read that first before reading this second part (and of course, then following the link at the bottom on to the third part of the series).

Jetstream Considerations

If your nearby potential nuclear target is one which you deem likely to receive an airburst, then your calculation becomes fairly simple.  Sure, there will also be some local fallout too, but most of the fallout will be swept up into the jetstream and then distributed pretty much everywhere the jetstream travels.

There are two main jetstreams to consider in the United States.  In the northern part of the country, there may be a polar jetstream, and in the lower states, there may be a subtropical jetstream.  Both of them have their wind blowing generally from the west to the east, but they can also rise northwards or dip southwards, with such movements varying from day-to-day.

In general it is possible to say that while the winds are predominantly flowing from west to east, they start off, on the west coast, with a slightly southerly direction added to their movement, and by the time they reach the east coast they are flowing in a slightly northerly direction.  You can see the current form of the jetstream here.

Here’s an interesting series of four charts showing how much the jetstream can move, with its location plotted over the course of four days.


What this means for us as preppers is that, in general terms, we want to be mildly concerned about airblasts that might occur to the west of our location – whether they are directly west, to the northwest, or the southwest.

But in reality, jetstream carried fallout is not a great concern, due to it being distributed so widely.

An exception to that is if you’re downwind of the potential jetstream plume from many nuclear blasts – this is of course the case in much of the eastern United States.  Due to the potential distance between the source of radioactive fallout and where it lands, this is about as accurate a statement as one can make – the further east you go, the greater the fallout from the jetstream you might get.

Closer to the Surface Wind Patterns

If you need to evaluate the likely effect of closer to the surface fallout, ie from a surface or subsurface blast, or from a nuclear power plant mishap, then you have a much more complex – and much more important – calculation.

First of all, to show the degree of variability of fallout from a ground burst, have a look at this map of remaining contamination from Chernobyl (as of 1996).  As you can see, some areas close to Chernobyl are looking fairly safe, while other areas, a much greater distance away, are looking still very dangerous.

A much more dramatic view of the fallout from Chernobyl can be seen in this animated graphic which shows how the fallout traveled over time.  It isn’t clear exactly what the colors represent or what the time frame is, and an email to the person who posted the animation on Youtube went unreplied.

We guess this is showing lower atmosphere winds, and in total, probably over a ten to twenty day period.  Another way of ‘calibrating’ this animation is to understand that it took two days for fallout to reach a point in Sweden just out of Stockholm, 700 miles away (in other words, the fallout cloud was traveling at a rate of about 15 mph).

Another example of fallout distribution is shown in this chart, reporting fallout from the largest ever US nuclear explosion – the ground burst of the Castle Bravo device in 1954.  Fairly consistent winds, and less ‘friction’ caused by the sea, rather than the interference and disruption to smooth wind flows on shore (buildings, trees, hills, etc) saw high levels of fallout as much as 280 miles to the east of the explosion, but barely ten miles to the west.

The blast scooped out a crater almost 1¼ miles in diameter, and up to 250 ft deep.

The 15 MT Castle Bravo outcome can be considered a ‘worst case’ scenario.  Currently Russia has some 20 MT warheads on its SS-18 missiles, but most of its missiles are armed with warheads ranging from about 100 kT up to 1 MT.

Learning from Volcano Eruptions Too

More information about how fallout would travel can be obtained, in general, by looking at ash distribution from volcano eruptions.  Volcano ash distribution is interesting, and needs to be considered in light of the height the ash rises to in the atmosphere – sometimes it reaches well up into the jetstream, and sometimes it doesn’t.

The Eyjafjallajökull volcano eruption in Iceland in April 2010 disrupted air traffic over much of Europe for a couple of weeks, due to the enormous amount of ash thrown up by the volcano (about 130 million cubic yards), and its wide spread around much of Northern Europe.  Some of the ash made it up into the jetstream.

Here is an interesting series of maps showing the fallout spread as it evolved during the period of major activity by this volcano, and this page has an interesting discussion of how clouds of volcanic ash (or fallout) are swept along by winds).

Understanding the Relevant Wind Patterns in General

To appreciate your degree of risk from lower level fallout (ie fallout that only rises a limited distance into the atmosphere and then drops back down to ground, missing the jet stream) you need to ‘join the dots’ between wherever the radioactive fallout might be released and where you are.

By this we mean you draw up a series of charts showing the prevailing winds at your location, at the point where the release would occur, and at relevant points in-between.

The first point to focus on are the prevailing winds at your planned retreat.  If your retreat generally has winds blowing towards the possible release location, that’s a very good thing.

But there’s a bit of a trap to consider, which requires you to look at more data to better understand things.  For example, let’s assume you are at Point R and the nuclear fallout comes from Point X, and you ascertain that the wind at your retreat (Point R) is blowing towards Point X, where the radiation release occurred.  So that means you’ve nothing to worry about, right?

Not necessarily.  Let’s also see where the wind at Point X is blowing.  You do that and you believe you’ve found more good news – the wind isn’t blowing directly towards you, instead, it is blowing at right angles to you.  Here’s a sample picture showing this – we use red arrows for winds carrying contamination from the release point, and green arrows for ‘safe’ winds.


So that looks close to optimum, don’t you think?  We’ll agree that you’re a prudent person, and so you obtain some more wind readings, and now you see a more detailed view.


This looks even better, yes?  But let’s not stop, let’s be obsessive and get as much wind data as we can.  You do some more research, and now you see this as a wind flow model.


This would seem to confirm all you need to know, right?  However, with so much information on the internet these days, and remembering just how far fallout can be carried by even mild winds, let’s grab some more data.


What do you think now?  Could it be…?  Let’s get a few more data points, and all of a sudden :


This now shows a very different scenario, doesn’t it.

As you’ve already seen in the real world examples linked to above, winds – and whatever they carry can travel on irregular paths, like a meandering river delta, with lots of crossing and inter-twined threads.  The winds that start blowing down the image in the first example then looped around and started blowing up again, taking the contamination and dumping it directly on you, even though your first data points seemed to suggest completely the opposite.

Clearly, this was a staged example, not a real world one.  But it teaches us two things.  First, we need to trace not just where the wind blows to, from our retreat, but also where winds are coming from prior to getting there.  Second, we need to follow the wind flow from the contamination release point for at least 200 miles to get a feeling for what twists and turns they might take.

Okay, so that’s an easy enough concept to be aware of.  But there’s a complicating factor that you’re probably already thinking of.  Winds don’t blow steadily, all day every day (and all night) and always from and to the same directions.  How do we factor in the variable aspects of wind flows/currents?

Read More in Part Three

To answer the question about understanding and applying the wind data information you have collected, please now click on to the third part of this three-part series, Available Wind Data Sources and How to Use Them.  Also, if you’ve not yet read it, the first part of the series, Using Wind Data to Estimate Fallout Risk, is also, of course, helpful and relevant.

Jun 252013
A NOAA map showing average wind speeds for the month of May.

An NOAA map showing average wind speeds for the month of May.

This is the third part of a three-part article series on using wind data to evaluate the risk of nuclear fallout at your retreat location.

If you arrived here directly from a search engine or link, you should probably go to the first part of the series – Using Wind Data to Estimate Fallout Risk – and read that first, then continue on to the second part – The Two Types of Wind Effects to Consider – before returning to read this final part.

We are fortunate to have a great deal of information available to us which can be used to help us understand which way the wind blows.

But before we look at the data sources, we need to consider four additional issues when interpreting the data.

1.  Use Data Averaged Over Many Years

We know that winds vary by the season, month, day, hour and even sometimes by the minute.  We know that some days are unusually windy, and others are unusually calm.

Clearly we can’t base our expectations for future wind patterns on only one day of observations and history.  Instead, we need to average the data – but not by taking the average of many days together, but rather by taking an average of the wind on the same day of many different years.

You want the wind data for, eg, 15 July, to be the average of the winds on the 15th of July over each of many previous years – the more previous years, the better, although there comes a point where adding more years no longer really adds much extra value to the data.  An average of ideally ten or more years will give you a better feeling for what the general probability will be for the next 15 July.

It would help also to understand the ‘SD’ or standard deviation of the averaged number.  This is a statistical term (click the link for a detailed explanation if you’d like) that simply measures whether most of the data points that you are averaging are close to the average number or not.

For example, if you had seven numbers  1 5 9 13 17 21 25, and another seven numbers, 10 11 12 13 14 15 16, in both cases, the average of the numbers is 13.  But the spread of values is much wider in the first series than the second series.  In the second series, the average is 13, and the maximum is only a bit more, and the minimum only a bit less, but in the first series, the maximum is almost twice the average, and the minimum is 13 times less.  All values are close to the average in the second series, but only a very few values are close to the average in the first series.

In the case of the first series, it has a standard deviation of 8.6, in the second series, a standard deviation of 2.2.  The smaller the standard deviation, the closer the individual data values are to the average and so the more accurately the average predicts the reality of individual days.

2.  Use Monthly Wind Values, Not Annual Wind Values

Sorry, but you’re going to have to do these calculations twelve times (or use data already calculated for you).  As the seasons change, so too can wind patterns – not only wind speeds, but also prevailing wind directions, too.

While ideally you’d do the calculation for every day of the year, you’ll end up drowning in data, and with wind being variable and unpredictable at the best of times, probably it is good enough to simply look at monthly data.

We suggest you create 12 maps, one for each month, and then plot wind directions at your retreat location and other close weather stations to it for each month, and also, as per the examples in part two, also ‘walk back’ the wind to see where it is coming from.

If there are nuclear targets or power plants within 200-300 miles or so of your retreat location, you want to add them to your maps and add wind directions to their locations too, plus then move forward to follow those winds from the possible fallout release points.

3 & 4.  Consider All the Varying Speeds and Directions that Wind Blows Each Month

Okay, now it is time to start to use the understanding you’ve developed and to look at the specifics for what to expect.

There are a couple of things to keep in mind now that you are looking at actual data.

Understanding wind direction descriptions

The first point is very important.  When you are looking at wind direction information, is it telling you the direction the wind is coming from, or the direction that the wind is going to?  Some data will provide information one way, other data will provide data the other way.

To explain, say there is a wind that is moving its air in a direction such that air to the south of you blows past you and then continues traveling north.  This is what is traditionally called a southerly wind – wind is coming from the south, from a 180° direction on a compass heading.  If you had a traditional weather vane, it would swing and point to the south.

But if you were flying a flag, the flag would extend out to the north of the flagpole – the opposite direction to the weather vane.  The wind is blowing to the north (ie 0° or 360°) (as well as from the south).  If you were talking about a car instead of wind, you would say the car was going in a northerly direction.

How should this wind be described.  Is it a wind from the south, or to the north?  It could be either, and it is both!  So be very certain to understand which descriptional method is being used.

Wind height measurements

Wind speeds are slower, the closer to the ground they get.  At 1/10″ above the ground, they are close to zero, and then they increase up from that until reaching a point of maximum speed, then they start decreasing again above that.  It is even possible to have winds traveling in different directions at different heights above the ground (balloonists use this to help them steer their balloons).

When you are comparing wind speed measurements, you need to know at what height the measurement was taken.  Clearly if you are comparing a measurement at one location that was taken at 3 feet above the ground with a measurement somewhere else that was taken at 30 ft or 300 ft, you’re not comparing like with like.

These days the standard measurement height for wind speeds is 10 meters (about 33 ft), and if you are reading official wind data from a weather service, that is probably what is being provided.  Often the actual measuring device is not at exactly 10 m above the ground, and so the speed it displays is then converted to display the theoretical/probable speed at 10 m.

But if you are using data prepared for siting a wind turbine, that will have data on wind flows at much greater heights.  This is useful information too, but be careful not to mix together data from different sources at different heights.

Data Sources

To get you started, here’s a nice simple map showing annual average wind directions and mean speeds.

But this map – while giving you a nice overview, is potentially very misleading.  The national averages can mask huge differences from season to season.  You need to drill down from the annual information to monthly information, and ideally, you want to get information from more locations, too.

First, however, to compare like with like, here’s a series of maps drawn from the same source for each of the twelve months.

January  February  March  April  May  June  July  August  September  October  November  December

(If the links no longer work, you can hopefully recreate them from this page.)

Here is an alternate series of monthly averages for wind speeds and directions, calculated for the period from 1930 (or later) until 1996 here.  That gives you another perspective for monthly variations, and has been provided for a much larger number of locations.  Hopefully the data you see here is sort of consistent with the data in the previous links.

The National Climatic Data Center and its US Climate Reference Network has a lot of data, but only from a few reporting stations, and while it gives wind speeds – typically measured at a nonstandard height of 5 – 15 ft above ground, it does not report wind direction.

If you’d like to see the wind information in a more visual manner, albeit slightly abstracted, here’s a fascinating series of maps.  It seems they only provide monthly data, without averaging the same month over several years, but if you click the link for any given month/year, it then provides both information on the specific month/year and also shows you the 30 year average for the period 1971 – 2000 and thirdly shows you how the specific year’s monthly average varies up or down from the typical month.

But wait, there’s more.  These first maps just show wind speed, not wind direction.  There are two more sets of maps, which between them show direction.  One shows what the call the U component, which is the east-west direction of the wind, and the other shows the V component, the north-south direction.  This is very interesting to see split out.

Wow.  Are you starting to get overloaded with data, yet?  Well, there’s one more wonderful source of summary data for many different locations.  That starts from this page here, and you can then choose the locations you want data about and the data you wish, either from their map interface (perhaps the easiest) or from data searching and other options.

One more resource.  If the previous resources have not had sufficient information for you, go to the appropriate regional climate center (choose from this main menu/map) and you’ll find an enormous amount of extra data for very many different reporting stations in the region.

But, wait – we’ve still not finished.  There are two more things to show you.

Leaving the best until last, remember when we spoke in the previous part of this series about wanting to know not only the average speed and direction of wind each month, but also the range of actual values?  Here’s a totally marvelous site which does exactly that.

Zoom in to the region of the map you are interested in, and you’ll see little wind roses appear for each reporting site.  Click on one of the wind roses, and that brings up a page showing you detailed information.  You can then click on the months in the circle as you wish.  Note if you keep clicking months ‘on’ you get averages of all the months clicked ‘on’; you need to click months off again so as to see the single month patterns.

And, lastly, some eye candy which gives you a fascinating feel for how it all ties together.

You can click to zoom in on any part of the map you wish, and when you start clicking to zoom in, an ‘unzoom’ button appears on the left.

The ‘problem’ with this map is that it is limited to showing you only the wind patterns right now.  It doesn’t give you a feeling for average flows.  But just seeing the ‘right now’ flows is still fascinating and helps you to understand how winds don’t flow consistently in straight lines for long distances.


Nuclear fallout may be transported from the location it was generated to your retreat either by high atmosphere jetstream winds, or lower down ‘regular’ winds.  The jetstream winds move swiftly and in doing so, tend to thinly disperse fallout over a wide area, while the lower down winds, moving much more slowly, tend to drop more concentrated fallout patterns within 200 – 400 miles of the nuclear release.

If you are within several hundred miles of a nuclear power station or a nuclear target, you should understand the degree of risk associated with fallout traveling to your retreat.

Winds are unpredictable on an hourly basis, but they follow certain trends and are more likely to be blowing in some directions than others on a daily and monthly basis.  There are lots of online resources to help you understand exactly what type of wind flows to expect at your retreat, and at potential fallout release sites too.

This was the third part of a three-part series about using wind data to estimate your retreat’s potential risk of being affected by nuclear fallout.  If you’ve not yet done so, we recommend you also read parts one and two of the series.

We have additional information on nuclear power related vulnerabilities in its own section too, and lots more information on weather related topics also.

Jun 242013
This map shows the maximum posted daytime speed limits on rural interstates.  TX is fastest at 85, the grey states are all 75.

This map shows the maximum posted daytime speed limits on rural interstates. TX is fastest at 85, the grey states are all 75.

What would life be like without cars and other forms of motorized transportation?  That’s a question we’ll almost surely find the answer to in a future Level 2/3 situation, but until such time, having convenient transportation is an essential part of our lifestyle.

Actually, convenient transportation will become even more vital in a Level 2/3 situation in the future, in a scenario where you might be more reliant on horse or other animal power rather than gas/diesel power.  These new constraints will completely redefine what constitutes acceptable and unacceptable transportation issues/constraints, and some of the present day issues (eg congestion) will probably disappear entirely.  We’ll also see the gradual decay and diminishing of our amazing current national roading system, bridges will fail, and so on.

But the future issues and challenges are a matter for other articles.  In this article, we mainly look at many of the issues associated with present transportation.  These issues impact on the desirability of potential locations as retreats, because hopefully for the indefinite future, life will continue as normal, and our experiences will be shaped by present day issues rather than by the challenges of TEOTWAWKI.

There’s another reason for looking at such issues as well.  How a state legislates for traffic matters gives you an oblique perspective of how intrusive and controlling the state wishes to be in the lives of its citizens.  The more traffic laws, and the higher the penalties, the more likely there are to be too many laws on too many other things too, and draconian penalties for all sorts of other minor offenses too.

Here are a number of criteria to consider when choosing retreat locations.  Our map graphic at the start of this article touches on one consideration – the freeway speed limits each state allows.  You can see a larger size map here, and this page has a more detailed table of data for each state.

If you are like us, you’ll probably equate being able to drive faster with a better state in general to live in.  🙂

Driving Safety

Of course, the justification for lower speed limits is usually safety.  Dubious data suggests correlations between traffic speeds and traffic safety.  We’re not going to argue the point about how fast is too fast, but we will definitely agree that there are very large differences between states in terms of vehicle accident rates.

The most relevant measure of the safety (or danger, if you prefer) of driving in each state is to look at the deaths per 100 million miles traveled.  This is more relevant than the deaths per 100,000 of population, because some states have people driving much greater distances than others.  Here’s a table that shows this data both ways.  The safest state was MA, while the most dangerous state was MT, with nearly three times the rate of fatalities (1.79 per 100 million vehicle miles in MT, 0.62 in MA).

One word about the Insurance Institute for Highway Safety.  We’ve used their data for many of the elements we look at in this article, but we also understand them to be funded by a group with a massive vested interest in the matter – insurance companies.  What is the vested interest that insurance companies have about road safety?  That’s a good question, and there are two possible answers.

The first answer is that by making the roads safer, insurance companies can lower their premiums and also make more profit from lower premiums, because they don’t need to pay out on accidents as often.  That’s obviously the positive view.  But there’s a second answer, too – by encouraging states to penalize more and more types of driving, the insurance companies create opportunities to raise insurance premiums based on a driver’s ‘safety record’.  Some cynics feel that this may be the stronger motivation.  We make no statement, but we do point out that there are both these issues driving the apparently laudable promotion of safety issues by the IIHS.

What about the role of alcohol in fatal accidents?  Less is known about this than you might think, because not all drivers involved in fatal accidents have their blood alcohol tested.  Furthermore, the total numbers of cases by state are surprisingly low, so statistically, the answers are not always very significant.  You can see a table here, however, and most of the states score very similarly to each other.

These days all states have a limit of 0.08g of alcohol/100ml of blood, but penalties vary.  This table shows how severely different states treat DUI/DWI.

Driving Costs

The cost of driving varies appreciably from state to state.  The main variations in cost are insurance, gas prices, and registration costs.

This table lists typical insurance costs by vehicle, ranging from the most expensive states (LA, MI and GA – $2699, $2520 and $2155) to the least expensive states (NC, IA, ME – $1085, $1028, $934).

This table shows the cost for a vehicle title and annual registration by state, although it seems to us that some states have additional fees imposed by city and county authorities in addition to the state fees shown in the table.

Fuel taxes hit you every time you go to the pump.  This table has 2010 data by state, including not just a simple statement of how much is taken in state and local gas taxes out of every gallon, but some additional data too.  Page 8 probably has the best table, highlighting the huge range in tax levels, from a high of 58.1c/gallon in IL to a low of 8.0c in AK (or 14.0c in the lower 48 states, in WY).

Depending on where your retreat would be located, and where you might regularly drive, you might find yourself up for turnpike fees too.  Here’s a list of toll roads in the US and here’s some more data on the fees they charge.

Seat Belts, Helmets, and Phones

A difficult compromise that all states, counties and cities have to wrestle with is where to draw the line between allowing their citizens the freedom to make wrong/foolish decisions on the one hand, and insisting on proper/best behavior on the other hand.

We make no value judgments about these issues, but you might find the different ways that different states respond to some of these bellwether issues to be illuminating.

The first of the big three issues is requiring people to wear seat belts.  Although all states except NH now require front seat passengers to wear seat belts, there are different approaches to enforcing the law, and a wide variation in terms of special child restraint laws.

This map distinguishes between states that have seat belt laws as a primary enforcement item, and those with it as a lesser secondary enforcement item.  This map shows the age below which children have to be in an appropriate restraint system, and this table has detailed information on state seat belt and child restraint laws.

A related topic is requiring riders on motorbikes and bicycles to wear safety helmets.  Only 19 states require all motorbikers to have helmets, and 28 more require helmets of some riders (eg younger riders).  As for bicycles, 21 states have bicycle helmet laws, although none apply to all riders, state-wide (but there may be county or city laws applying to all riders).

This map shows motorbike helmet laws by state, and this map shows bicycle helmet laws by state.  Here is a table with information on the applicability of such laws.

The third of the ‘big three’ things is the use of cell phones while driving.  Hand-held cell phone use while driving is banned in 11 states, and text messaging is banned in 41 states.  This map shows state laws on hand-held cell phone use, and this map shows state laws on texting while driving.  Here is a table of information about these two issues.

Traffic Enforcement Issues

Depending on your perspective, states that are less aggressive at traffic enforcement either show a wanton disregard for the importance of human life, or perhaps, alternatively, are less intrusive and obsessive at controlling every last detail of our lives.

In particular, we have a strong dislike of states that aggressively use photo-radar and red-light cameras.  Again, opinions differ, but there are credible concerns widely expressed that suggest such devices primarily exist to make money for the local authorities (and for the companies that operate them under contract).  Too often we’ve read about cases where traffic lights have their timings changed (ie shorter orange light times) when red-light cameras are installed, and speed cameras are as likely to be located where normal average speeds are high as they are in areas where accident rates are significant.

This table on the Insurance Institute for Highway Safety’s website lists state and local policies on the use of such devices.

The National Highway Traffic Safety Administration has an interesting summary table of state policies and penalties for speeding and ‘reckless driving’ (a concept which is very subjective) and more detailed information on each state from this menu.

We really don’t like states which potentially can jail first time speeding offenders.  Of course, that almost never happens, and if you’re speeding truly fast, then even in a non-jailable state, you can find yourself locked up, because the officer who stops you will simply upgrade your ticket to reckless/dangerous driving or some other more serious charge.

Traffic Congestion

No-one likes getting stuck in traffic, but it seems to be an unavoidable part of living in any moderate to large-sized city.  For many reasons, all ultimately being, of course, based on money, few if any roads are built to a traffic handling capacity such that they can conveniently handle not only average volumes of traffic but also peak surge volumes.

However, your retreat is unlikely to be anywhere near a big city, so we’ll ignore those issues (but here’s a good starting point if this is relevant to you).

Instead, let’s look at more rural parts of the country, and traffic flows there.  Here’s a map showing freight traffic movements across the country (we think it dates back to 2010 or earlier).  It provides an interesting perspective on where commercial traffic flows across the country.

Looking ahead, here’s a second map that shows only the extra amounts of freight traffic expected to be added in addition to the freight traffic already shown in the first map, above.  That gives you a good impression of where future traffic will be appearing.

Both these two maps were taken from this report.

Here’s a more forward-looking map, showing projected truck traffic in 2035.

In addition to simple traffic, how about congestion?  This map shows the predicted level of congestion on freeways and other major roads in 2020, and this map adds more secondary routes to its 2020 congestion display.  Both are taken from this report.

Other Transportation Issues

There are many other considerations that you might want to also evaluate.  For example, here’s a map that ranks states by the quality of their bridges and what percent are deficient and in need of priority repair/replacement.  PA is the worst state, FL the best.

This map is part of a fascinating website that gives you detailed information about all the road bridges in your area.  That’s a relevant issue to understand, because it gives you a clue to what may happen in the future WTSHTF and road maintenance stops – how long before the essential bridges in your area start collapsing?

A related, but more difficult to get hard data on, issue is that to do with road maintenance needs in general.  For example, do you have roads along hill-sides that are subject to landslides falling onto the road, or slips/floods washing the road away?  Do you have roads lined by large trees that could fall over and block the road?

Another issue to consider is snow removal in winter.  If you’re in an area with appreciable winter-time snow, what happens to the major and minor roads in your area?  Will you get snowed in, and if so, would it be for a few days or might it be for many months?  As for WTSHTF, there’ll of course be no snow removal in that type of scenario.  What will you do in that situation?

A related part of these questions is to consider what the potential seasonal problems could be if/when you need to bug-out to your retreat.  How much of the year might the roads be impassable?  Are there any major risks on the routes you would have to take that may interfere with your bug-out plans?


The quality of our roading system, its reliability, and the associated costs of traveling by private vehicle are essential aspects of our present normal life.  At the present, they are factors to consider in choosing a retreat location.

In the future, if a Level 2/3 situation does eventuate, some issues will become irrelevant, but other ones will become vitally important.  You need to consider both present and future issues when weighing transportation considerations as part of your retreat selection process.

Jun 172013
The cost of living is a relevant factor to consider when choosing a retreat location.

The cost of living is a relevant factor to consider when choosing a retreat location.

Whether you’re looking at a retreat in a nearby state, or a far-away country, one of the issues to consider is how much the cost of living will be.

For sure, WTSHTF, there will be a massive rewrite of the cost of living equation, with unpredictable and uncertain results into a Level 2 and/or 3 situation.  But if you’re considering spending time at your retreat prior to a Level 2 or 3 situation, then the local cost of living is helpful to understand.

In particular, an appreciable number of preppers seek to find a dual-purpose location which is suitable to live at, permanently, and also to stay at as a retreat if things go bad.  There might also be a (very weak) correlation between cost of living and the cost of creating a retreat.

Here’s an interesting list of the 30 cheapest cities in the entire world to live in, as of April 2013 (Note that these indexes are regularly updated, so depending on when you’re visiting, you might want to search out the latest list from the site’s home page).

And, for its twin, here’s a list of the 30 most expensive cities to live in.

The website uses a detailed methodology to create these lists, but there’s one important thing to appreciate.  It calculates costs to create an equivalent cost to live at a reasonable western standard, rather than the costs to live like a local, in whatever sort of average lifestyle locals have.  That is presumably why Luanda, Angola scores so high, and probably also why six US cities appear in the list of the 30 cheapest cities.

Of course, if you are considering a move anywhere outside of the US, you’ll want to be able to accept a lifestyle that is perhaps somewhere in the middle, a compromise between how the locals live and how you’d prefer to live in a perfect world, and for sure, in an emergency situation, you’ll need to be able to live completely like a local, because all your favorite imported luxuries will no longer be available.

In addition to the 60 cities listed as either most expensive or least expensive, you can see ratings for their entire database of 780 locations, all around the US and all around the world, here.

Here’s their list of all 50 US states, ranked from lowest cost of living (a bit confusing, the higher the score, the lower the cost) to the highest.  We also show the US average (443) cost of living to give a relative measure of better and worse states than average.

TN  702
KY  688
AR  683
OK  670
KS  663
MO  657
TX  650
NE  640
AL  632
MS  621
GA  619
LA  611
ND  609
IA  607
ID  598
WV  593
UT  590
OH  585
IN  582
SD  569
NC  566
MI  560
IL  554
WY  552
WI  551
SC  545
NM  541
VA  535
FL  513
MN  503
CO  497
MT  483
NV  470
DE  465
AZ  464
PA  452
US  443 National Average
WA  441
OR  408
NH  402
ME  381
VT  359
RI  347
MA  342
MD  307
NJ  296
CT  265
NY  257
CA  253
AK  171
HI  149

Interestingly (perhaps) using this company’s ranking methodology, the first of the American redoubt states doesn’t appear until you reach the fifteenth place (ID), and the two partial states in the redoubt (OR and WA) are 11th and 12th worst states, no doubt due to the big city and liberal influences in the western part of these two states.

Here’s a map showing the top five states in strong green, the next five states in pale green, the worst five states in strong red and the following five states in light red (we ignore HI and AK).




There’s no real surprise to see where the red ends up, but the green might be more of a surprise.

The local cost of living is of course only one factor to consider, and also bear in mind that these are state-wide averages, so can vary greatly across a state.  As an example of variations within a state, Texas scored very positively with a state-wide rating of 650, but there is quite a spread between individual cities, ranging from, eg, Houston at 661, El Paso at 634, Dallas at 610, and Austin at 583, but Amarillo gets a very different score, at 300.

Other Criteria

The results we’ve been discussing here are based on one set of assumptions, tailored to help companies adjust compensation packages based on where they relocate an employee.  There are many other ratings for cost of living, using a mix of the same and perhaps different criteria, and with different weightings for different factors, all giving slightly different sets of results.

For example, here’s another excellent set of ratings, also from April 2013.  There are both similarities and also some differences between this other rating system (MERIC) and the first (Xpatulator).

We summarize the best and worst states as per Xpatulator and show their comparable ratings with MERIC in this following table.  We rate 1 as the cheapest state and 48 as the most expensive (we again ignore HI and AK).

State  Xpatulator  MERIC 
TN 1 2
KY 2 3
AR 3 7
OK 4 1
KS 5 8
MO 6 12
TX 7 9
NE 8 5
AL 9 13
MS 10 10
ID 15 4
NH 39 40
ME 40 39
VT 41 41
RI 42 44
MA 43 43
MD 44 42
NJ 45 46
CT 46 48
NY 47 47
CA 48 45


The results are similar, but not identical.  Both agree on the same ten ‘worst’ states, in slightly different order, but there are slightly greater differences in the best ten list, with only seven states appearing in both top ten lists.

One notable difference is Idaho comes in at number 4 in the MERIC ranking, but only made 15 in the Xpatulator ranking series we first looked at.

The best state for the MERIC ranking series is Oklahoma (scores as 4 with Xpatulator) and Tennessee for Xpatulator (scores 2 with MERIC).

There are plenty of other rating series too, but even ‘just’ these two sets of data give a reasonable consistent picture already – maybe more rating series start to add confusion rather than clarity!


There are major differences in the cost of living between the various states in the US, and of course, even greater differences when you start to look internationally, too.

While this is only one of the very many factors you need to consider in choosing a location for your retreat, it is a valid consideration to bear in mind.  The Xpatulator website has information on all US states, many US cities, plus a large number of other cities, regions and countries around the world that gives you one perspective on how the various costs are made up to live in different locations.  Another good rating series is the MERIC series.

If you are considering off-shore locations, the calculated cost of living is an even smaller part of your total evaluation, and you also need to realize that many of the apparently more desirable countries and locations have what can be politely termed ‘developing’ economies, meaning that these numbers are subject to potentially great change with little warning.

Jun 122013
A spectacular fire at the Fukushima Daiichi Power Plant in March 2011.

A spectacular fire at the Fukushima Daiichi Power Plant in March 2011.

Even the most stubborn of non-prepper types might be worried about living too close to a nuclear power plant.

Indeed, the concerns about safety are not just at a personal level (some people attempt to keep their distance from nuclear power plants) but also at a national level – many countries forbid their construction, and more recently, some countries which have been reliant on nuclear power for a substantial part of their total electricity needs are abandoning nuclear power and closing down their plants, again due to concerns about the safety of the reactors.

There’s probably not a single person in the US who doesn’t know about the Three Mile Island nuclear power plant’s near total meltdown, even though it occurred 34 years ago in 1979, and there’s even fewer people in the entire world who are not acquainted with Chernobyl (a 1986 event).  Most recently, even if people don’t remember the name of the location, the Japanese nuclear power plant crisis at Fukushima had so many people stocking up on Iodine tablets that they became as scarce as ammunition is at present.

So it is unsurprisingly common to see on typical sorts of hazard and ‘do not locate here’ prepper retreat selection maps, that prominent marking is given to not just the location of currently operating nuclear power plants, but also to the location of decommissioned plants as well.

All this focus on nuclear power plants begs the question – just how much of a threat do they pose?  Let’s have a look, first of all, at the potential dangers of what has gone wrong in past nuclear power plant accidents, and then secondly, evaluate the risk of such events in the future.  The results are surprising.

The Implications of a Nuclear Power Plant Mishap

Problems with a nuclear power plant can vary from trivial to catastrophic.  Trivial mishaps occur from time to time, but they never escalate to an emergency because backups and safety systems kick in and resolve the issue before it escalates.

If backups fail, there are other design features that are intended to minimize the extent of any worst case scenario events, chief among which is the containment structure around the power plant itself.

But, of course, Murphy’s Law applies to nuclear power plants and even to triple redundant safety systems, just as much as it does to anything and everything else.  So we have, on occasion in the past, seen events that have resulted in the release of nuclear/radioactive materials.  A chilling consideration is that even the ‘worst case’ scenarios that have occurred to date (especially Fukushima) weren’t truly worst case – as bad as they were, they could have been much worse than they ended up as being.

So what is prudent to consider in terms of possible worst case events?

According to this document, local government authorities plan for two rings of danger around a nuclear power plant.  The first has a 10 mile radius, within which there’s a danger of direct radiation exposure, the second is a much larger zone with a 50 mile radius where radiation and fallout might contaminate food and water.

These are fairly nonsense measures with little science underlying them.  Let’s compare this plan with the reality of what we’ve seen in past events.

The problems at the Japanese nuclear reactor complex at Fukushima are interesting to consider.  According to this page, the Japanese created a 20 km radius exclusion zone (ie about 13 miles), but considerable radiation was received at much greater distances and small levels of fallout made it all the way across the Pacific and landed in the US.  Furthermore, some commentators have suggested maybe the Japanese understated the amount of radioactivity released and its impacts.

Going back to the Chernobyl disaster of 1986, for which there is still a 30 km radiused exclusion zone around the power plant (18.5 miles) there are also remaining radiation hot spots in places out as far as almost 200 miles from the reactor – see the map on this page.  In the early period shortly after the release of radiation, some hot spots existed as much as 300 miles from the reactor.

On the other hand, our own Three Mile Island event in 1979 involved the partial meltdown of a reactor, but only a very limited release of radiation, and according to this page, no-one has subsequently died from the radiation released.  At the time of the event, initially a 5 mile radius evacuation zone was suggested, and then that was briefly extended to recommending voluntary evacuation within 20 miles of the plant, but many people quite sensibly ignored this.

Our point is simply that there’s little way of predicting what the impacts of a problem at a nuclear power plant may be, but worst case scenarios can see seriously elevated levels of radiation from fallout, even over 200 miles from the plant, depending of course not just on the nature of the event but the prevailing winds.

The Scope and Scale of Nuclear Accidents

Clearly, there is a huge difference between trivial and massive nuclear accidents.  This page from the International Atomic Energy Agency has  an interesting explanation of the INES – International Nuclear Event Scale.  The scale is logarithmic, like the Richter scale for earthquakes, which means each level is ten times more serious than the level before.

Events that occur at levels 0 – 3 on this scale are trivial and probably of no impact to people outside the power plant.  Level 4 events are ones which there might be some local release, but of a minor level – you could leisurely walk away from the contaminated area and be safe.

Level 5 is where we start to become at more widespread risk, although as this Wikipedia list of INES events points out (somewhat obliquely) ordinary members of the public have apparently never died as a result of a Level 5 event at a power plant.

Level 6 definitely involves casualties, but there has only ever, in the entire history of nuclear power, been one such event, in the Soviet Union in 1957.

Level 7 events – the maximum level on the scale – have only occurred twice, at Chernobyl, also in the Soviet Union (now Ukraine) and Fukushima.

Part of the reason for the high number of resulting deaths from Chernobyl was the slowness of the Soviet authorities to acknowledge the events at the reactor, and the slowness of them subsequently evacuating civilians from the affected area.

In contrast, the Fukushima event was reasonably responsibly handled, with apparently – so far – only two directly related deaths, and neither from radiation (two employees who drowned).

Here’s an interesting table comparing the Fukushima and Chernobyl events.

Here’s a helpful Wikipedia page with a reasonably complete listing of accidents at all civilian nuclear facilities (not just power plants) and – for the sake of completeness, although not part of this discussion of nuclear power plant risks, here is a similar page on military nuclear accidents.

The Degree of Risk

Now it is time to put these past events and present risks in perspective.

Sure, we all are very aware of Fukushima, Chernobyl and Three Mile Island.  But how many others have also occurred?

The answer might surprise you.  As you can see from the preceding section, very few other problems have occurred at a level that resulted in the release of any radiation outside the power plant itself.  Furthermore, although it is hard to appreciate this from the hysterical coverage given at the time and for ever after, the Three Mile Island accident was trivial in nature and scope.

So, let’s now look at the two – only two – accidents involving appreciable releases of radiation.

The first of these – Chernobyl – involved a type of reactor design that is not used in the west.  It has a terrible weakness inherent in its design – if temperatures increase in the reactor, this creates a positive feedback loop that causes the reactor to increase its level of activity, which increases the temperatures further, which speeds up the reactor more, and so on until, all of a sudden, you have a Chernobyl disaster.  All western style reactors have a negative feedback loop design – if the temperature increases, the nuclear reaction decreases, and so the temperature drops back down again.

In other words, a Chernobyl type accident could not happen in the west.

As for the Fukushima event, this was precipitated not by any problems originating from the reactor itself, but by the earthquake and subsequent tsunami.  The reactors shut down because of the earthquake, and so stopped generating power, which was needed to keep the cooling pumps operating.

The earthquake also caused a loss of outside power coming into the facility, so the on-site emergency generators started up.  All was good until the tsunami arrived, flooding the generator rooms, floating away the fuel tanks, and causing the generators to fail.  With loss of cooling, the power rods heated up beyond safe limits and melted down.

The Fukushima event was the result of poor planning and lack of consideration of external vulnerabilities.  Everything actually worked perfectly and as it should until the tsunami swamped the generators.

Our point is simply that the two disasters that occurred were each unique and not symptomatic of systemic weaknesses that apply to other or all nuclear power plants in the US.

So, let’s think about this.  In some 60 years of civilian nuclear power plants being in operation, there have been only two accidents that involved serious amounts of radiation being released.

Now, remember that one of those two accidents could not occur in the US due to different reactor design, and the other is very unlikely.  But let’s still consider the Fukushima event as relevant, because maybe there are other unexpected vulnerabilities hidden within US reactor designs.  So, in approximately 15,000 reactor years of operation, there has been only one significant and arguably relevant accident.

People feel perfectly safe living next to a volcano that erupts maybe once every 15,000 or so years.  You should feel similarly safe having a nuclear power plant in your back yard.

Oh yes – the number of radiation deaths from the accident at Fukushima?  Zero.

Extending the Analysis

As of May 2013, there are 436 nuclear reactors operating within 127 power plants around the world, and another 70 reactors under construction.  Within the US, there are 104 reactors operating within 65 power plants.

There are also tens of thousands of nuclear weapons, none of which have ever accidentally exploded all by itself.

There are also many hundreds of other nuclear reactors.  You might live close to one without even realizing it – some universities have nuclear reactors on campus for teaching and research.  Military and research facilities also have nuclear reactors.

Then there are nuclear powered submarines and surface vessels – perhaps 200 or more in total.  And – oh yes – don’t forget other nuclear power sources, even if not involving critical reactors – that are located in many other places, below the earth, on the earth, and above the earth in satellites.

Consider all these things, and then contrast the two significant nuclear accidents we’ve seen, neither of which directly exposes any similar vulnerability in US reactors.

Bottom line – there are more nuclear reactors ‘out there’ than you’d think, and fewer nuclear accidents than ‘conventional wisdom’ seems to perceive.  Nuclear power plants are extremely safe.  Many more people die from the pollution released by coal and oil-fired power plants than by radiation that is (not!) released by nuclear power plants.

Based on this analysis, you should have no concerns whatsoever in locating yourself close to a nuclear power plant.  Unfortunately, we’ve not yet finished our analysis.

Two More Vulnerabilities?

There are however two types of vulnerability not yet considered.

The first of these is the potential for a deliberate radioactive release as the result of some form of sabotage or terrorist attack.

For sure, all nuclear power plants have security forces protecting them, but alas, equally for sure, we have to fear that a determined and clever terrorist attack may have an appreciable chance of gaining at least temporary access to the control circuitry and the physical reactor and related parts of the power plant.

We also have to concede that a well researched terrorist exploit of power plant vulnerabilities could result in grave harm being inflicted on the reactor, possibly to the point of a melt-down and an uncontained release of radioactive materials.

It is necessary to also assume that terrorists could infiltrate the actual power plant staff – if, as we’ve seen, an organization as security obsessed as the NSA can allow a contractor (Edward Snowden) with some obviously non-conformist and pro-privacy views to not only work for them but to walk off with large amounts of ultra-highly classified materials, and if we see muslims infiltrating the army before massacring soldiers (13 killed and 30 wounded by Nidal Hassan at Fort Hood), we have to accept that nuclear power plants are every bit as much at risk of infiltration and subtle sabotage, and maybe more so.

So it seems we have to accept that enemy terrorists can attack a power plant equipped with inside knowledge on security measures/capabilities and reactor vulnerabilities and how to exploit them.  That’s a little bit scary, isn’t it.

There’s one more way of terrorists attacking nuclear power stations, too.  By computer.  We can only guess at what possible vulnerabilities lie within all the computerized control systems at a nuclear power station.

And – wait.  We said there are two extra vulnerabilities.  That was only the first, even though there were three parts to it.  The second category of additional vulnerability is to wonder what would happen if an EMP attack disabled the control circuitry in a nuclear reactor.

We expect that nuclear power plants would have some degree of ‘hardening’ of their control circuits to make them semi-resistant to some types of EMP effects, but probably they would not be 100% invulnerable to all levels of EMP intensity.

We also understand that there are physical ‘fail safe’ design features, such as enabling the fuel rods to literally fall out of the reactor, causing the reactor to lose its criticality and stop functioning.

But we also then think about what happened at Fukushima – the fuel rods correctly fell out of the reactor in a SCRAM type action when the earthquake hit.  The problem there was when the power to drive the cooling pumps to keep the fuel rods cool failed.  That’s not a mechanical failsafe device – it requires electricity and electronics, which might be vulnerable to an EMP attack, and if this ended up with a similar type of Fukushima failure event, the result could be massively disastrous.

These are all hypothetical scenarios, and happily improbable.  But are they impossible?  No, alas, they are not.

You’ll have to decide how much importance to ascribe to such risks when making your own decisions about the risks associated with being close to nuclear power stations.


The popular perception of elevated risks associated with being located close to a nuclear power plant is generally wrong and hysterically overstated.

The well-known Three Mile Island accident was trivial in nature, no-one has died from radiation released at Fukushima (with the whole series of events at Fukushima not being due to anything going wrong with the reactors themselves), and the Chernobyl accident was directly related to a bad unsafe reactor design that is not used in the US.

As for other nuclear accidents, none have involved significant releases of radioactivity.

You’re more at risk of a nearby dormant volcano erupting than you are from dangerous levels of radioactivity being released from a nearby nuclear power plant.

But.  There is one unknown vulnerability that can’t really be quantified.  That is the risk of deliberate failure and intentional release of radioactive materials, caused by terrorist actions.  We don’t know how likely that might be, but we have to accept it is not totally impossible, and while we also don’t know what extent of release could be caused by deliberate action, we also have to accept that it could be very high.

So – nuclear power plants.  The good news – they have proven themselves to be very safe in terms of ‘normal’ risks.  But, the bad news?  They are potentially vulnerable to unusual risks and deliberate attack/sabotage.

Depending on other factors, it is probably wise to keep some distance from power stations accordingly.  Depending on the wind patterns between you and the power station, ‘some distance’ could mean as much as 300 miles.

Jun 102013
A map showing population changes, 2000 - 2010, across the country by county.

A map showing population changes, 2000 – 2010, across the country by county.

We’ve written regularly about the importance of population density when choosing a retreat location.

But there’s more to population density than just looking at the density of an area at a given instant in time.  Is the population increasing or decreasing?  That may impact on the future population density that you can anticipate to evolve.  And there is a very different dynamic in a region with rapidly growing population than a region which is steadily losing people.  Consider Florida, with new developments, new roads, new shopping, and formerly empty areas now becoming new towns and cities.  And then think about some of the rust-belt states, with decaying infrastructure, empty city centers, boarded up shops, etc.

So, do you even know if the retreat you may be considering is in an area with a growing population, or a retreat in an area with a shrinking population?  And, although the extremes of too rapid growth and too substantial decline are both obviously bad, what about moderate growth and moderate decline – which is better?

First, we should define what is normal, such as it ever is in a nation as varied and diverse as ours.  The country as a whole grew by 9.64% in the ten years between the 2000 census and the 2010 census.  So, in a sense, any area that did not grow by right around 10% is growing at a less than average rate and so is – in relative terms – sort of shrinking.

The 9.64% overall growth in the nation did not occur evenly over all 50 states.  According to the US Census bureau, the south and west each grew by over this amount, while the northeast and midwest grew by less.

The states listed in order of percentage growth are

Rank  State Growth (%) 
1 Nevada 35.1
2 Arizona 24.6
3 Utah 23.8
4 Idaho 21.1
5 Texas 20.6
6 North Carolina 18.5
7 Georgia 18.3
8 Florida 17.6
9 Colorado 16.9
10 South Carolina 15.3
11 Delaware 14.6
12 Washington 14.1
13 Wyoming 14.1
14 Alaska 13.3
15 New Mexico 13.2
16 Virginia 13.0
17 Hawaii 12.3
18 Oregon 12.0
19 Tennessee 11.5
20 California 10.0
21 Montana 9.7
AVG National Average   9.64 
22 Arkansas 9.1
23 Maryland 9.0
24 Oklahoma 8.7
25 South Dakota 7.9
26 Minnesota 7.8
27 Alabama 7.5
28 Kentucky 7.4
29 Missouri 7.0
30 Nebraska 6.7
31 Indiana 6.6
32 New Hampshire 6.5
33 Kansas 6.1
34 Wisconsin 6.0
35 District of Colombia 5.2
36 Connecticut 4.9
37 North Dakota 4.7
38 New Jersey 4.5
39 Mississippi 4.3
40 Maine 4.2
41 Iowa 4.1
42 Pennsylvania 3.4
43 Illinois 3.3
44 Massachusetts 3.1
45 Vermont 2.8
46 West Virginia 2.5
47 New York 2.1
48 Ohio 1.6
49 Louisiana 1.4
50 Rhode Island 0.4
51 Michigan – 0.6

Twenty one states grew by more than the national average, 28 states (plus DC) grew by less, and one state (Michigan) actually shrunk.

How Much Growth is Good?

We would suggest that you don’t want to relocate to an area that is undergoing significantly greater than normal growth.  A region and a state with ‘super-growth’ will see a changing demographic, will have an unpredictable future, will have greater pressure on land prices and availability, and inevitably spreading urban sprawl, traffic congestion, and other challenges.  After spending some decades in the Pacific Northwest, where the freeway capacity has never been able to catch up with population pressures, we speak from bitter experience.  🙂

If we arbitrarily define super-growth as being 10% above the national average, that would make five states undesirable.

We also suggest you don’t want to relocate to a moribund state with little or no growth.  There are seldom good reasons for a state suffering curtailed growth, and it may bring with it a feeling of resentment and greedy entitlement among those people remaining.

If we perhaps choose the bottom five states as being also undesirable (although we could as easily make this four, or six, or any other number) that provides a further point of focus.

So if we’ve eliminated ten states as growing either too fast or too slow, what about the other 40 states?  Are they all equally desirable/neutral from a population growth dynamic?

Clearly, some of these have to be better than others, but we’re also starting to shift from where it is appropriate to consider states as a whole to where it becomes more appropriate to think about counties.  However, and without thinking about the identities of the specific states, if we were to select say a matching ten ‘better’ states to highlight, we’d probably choose the ten ranging down from the average growth rate of 9.64%.  In order, these would be AR MD OK SD MN AL KY MO NE and IN.

To show all this on a map, we have shaded the ‘too fast growing’ states with red, the ‘too slow growing’ states with black, and the ‘slightly less than average’ states in green.


It is interesting to note that one of the fastest growing states (TX) is right next to one of the slowest growing states (LA).  However, LA is a special case state, due to the impacts of Hurricane Katrina in 2005, giving perhaps an anomalous result for this ten year period.  We also note that one of the redoubt states (ID) is also the fourth fastest growing state.

More Meaningful Data at the County Level

The state level information is interesting to see, but potentially misleading.  As we said above, we’re now moving to a situation where county by county considerations become more important, and so here’s a wonderful map showing exactly this.

This gives us a very different view.  We would suggest that the ideal color to look for would be the lightest blue color, showing a 0% – 10% population increase.  Beyond that, we would be probably comfortable with the lightest yellow color too (ie up to a 10% decrease) and then our third choice would be the mid color blue (10% – 20% increase).

Other colors could be considered too – remember, that even now we are at county level data, a single county can conceal conflicting trends in different parts, and need to be considered in the context of other counties around it, and also whether the county’s population is large or small, numerically.  For example, in a county with a very small population but a large land area, adding or subtracting just a couple of families could make a large percentage difference.

You can now see, for example, that even the most rapidly growing state (Nevada, with a statewide growth of 35.1%) actually shows most of the state with low growth, some parts with decreases, and almost the entire growth (in terms of actual extra people) occurring in one place only (Clark County – ie the Las Vegas area).

This is a great example of how average data for a larger area obscures major differences when you start to look in more detail.


The US population grew, nation-wide, by 9.64% over the period 2000 – 2010.  This growth was not evenly distributed over the entire country.  One state actually got smaller (Michigan), while the most rapidly growing state (Nevada) grew by 35%.

But within individual states, there is also a huge range of increases (and decreases) when you track the changes in each separate county.

Our general recommendation is to locate in an area with average to moderately below average growth, ideally between a maximum of perhaps 10% and a minimum of slightly less than 0% growth – a mild decrease, in other words.

But, just as how we saw a huge change when we went from state level data to county level data, even a single county can mix some areas of growth (perhaps a city) with other areas of no growth or population decrease.

As is the case with all data, our search does not end once we get county level statistics.  If anything, it is only just starting when you’ve created a short list of counties to consider in detail.

Jun 092013
A classic map of population density, but much too simplistic to be used to help you determine the rural nature of a retreat location.

A classic map of population density, but much too simplistic to be used to help you determine the rural nature of a retreat location.

One of the most essential tenets of prepping is that you need a retreat that is in a rural area, away from population concentrations and big cities.

It is fairly obvious when a region has too dense a population; when it is too ‘citified’.  But how can one measure the degree of rurality (is that actually a word?) or ruralness (another word the spell checker doesn’t like) of the region you are considering for your retreat?

The simple approach has been merely to look at population density per square mile, on the basis of ‘the fewer the better’.  Here’s a typical population density map.  But this is indeed a simplistic approach, with several limitations.  It can be appropriate as a very quick first filter of the nation, with some regions clearly being eliminated for having way too many people crowded into them, but beyond that, as you start to become more exact in your evaluation, it becomes increasingly limited.

For example, having no-one else living around you for many miles is probably too much of a good thing – or, in this case, too little of a good thing.  As we’ve remarked in other articles in our series about choosing a location for your retreat, you should compromise between too few and too many people, either too distant or too close.  So you’re not seeking the lowest population density.  Yes, you’re seeking a low population density, but the key issue is the type of people around you, not just the number of people around you.

The people who are within half a day or so of your retreat should be reasonably self-sufficient, so that in the future, they’re likely to survive and contribute to the viability of the region as a whole.

There’s also a difference between a region with low population density because no-one can comfortably live there (ie the middle of the desert or the top of the Rockies) and a place with a low population density because the people who live there are all on 80+ acre farms.  A simple population density map gives you no clue as to why some areas are empty or very sparsely populated.

Clearly you want to locate yourself somewhere such that your neighbors are growing all the food they need, and some more besides.  That not only implies you probably could also grow more than enough food for you and the people with you, too, but also means that your neighbors may have surplus food to trade for other goods, or, in an emergency based on unusual weather or crop infestation or some other disaster interfering with your own crop yield, to share with you.

This distinguishes such areas from other places which might be tourist resorts, or ‘service towns’ that happen to be somewhere otherwise in the middle of nowhere, or possibly places with a seasonal retiree population.  These types of groupings of people are not self-supporting, and rather than adding value and survivability to a region, they detract from it.

Maybe you’re not really cut out to be a farmer, but maybe you have some other suitable skill to offer in exchange for food.  You’ll get more food in return for whatever you do if you are trading with people who have a surplus of food than you’ll get from people who are food poor.  Or maybe you’re going to raise chickens, and you will swap chickens and eggs for fruit and vegetables and meat.

Whatever your plan, it is best done in a region with already viable other families living there.

A More Exact Statement of Population Criteria

If you don’t know what you’re looking for, you’ll never find it.  So let’s first try to make a formal statement of what you’re looking for when considering population density and rural character.

You want to find an area with a low population density – perhaps less than one person per 5 – 10 acres (whether right or wrong, a convenient rule of thumb is you need at least an acre of reasonably arable land per person for food, then add extra land to adjust for unproductive land, roading, buildings, livestock that require an acre or more each as well, and so on).  With 640 acres in a square mile, this translates to a maximum of 64 – 128 people per square mile.

Depending on where the people are located within the county, you also might prefer not to be in an area with a density of 0 – 1 people per square mile (ie one person for every 640+ acres).

Now, to move on from simple population density, you want the region your retreat is located in to have a primarily rural and sustainable economy with productive farming.

Let’s look at some ways we can better understand the country and its different regions.

Distinguishing Rural from Urban/City Areas

So –  when is an area considered rural rather than urban?  It seems there are several different measures to consider.  Population density is one, but as discussed above, it has limitations and flaws.

Another simplistic measurement is distance from major urban areas.  There are others.

For perhaps the most outstanding and definitive review of issues to do with classifying areas as rural or urban, this paper published by Dr Gary Hart of the University of ND analyses over thirty different classification methodologies and, in a dry and academic way, has a lot of interesting commentary.

If you don’t want to read its 50+ pages, the key finding is that most methodologies are flawed.  Let’s look at the common weakness that applies to most measurement criteria.

The Problem of Measuring Data at County Level

One of the problems of most approaches to analyzing regions and their uses is that they are typically based on county level data.  The problem with this is that counties vary in size, and sometimes obscure a mix of wildly different social and geographical areas within the overall county.

For example, in the Pacific Northwest, Washington state’s King County (map here) extends from Seattle and its adjoining high density cities to national forest, mountains, and empty areas of minimal population density; in total comprising 2,307 sq miles.

In terms of sheer size alone, King County is bigger than 60 of the world’s entire countries, but as big as it is, King County is not even one of the top 100 sized counties in the US, and is only the third largest in WA.  In terms of population (2.008 million) King County is bigger than 95 different independent countries, but is only the 13th largest county in the US by this measure.

Clearly, a county the size (either physical or population) of this and many other counties is way too large for accurate detailed information.  The collected/average data for King County ends up reflecting nothing that actually exists in reality (just like the average American family has 2.5 children, but have you ever seen a half child?).  Its overall scores obscure a mix of regions, some of which are intensely industrial and some of which are intensely rural, some of which have high population densities and some of which have low densities, some of which are skewed strongly Democrat and some of which are skewed strongly Republican.

Sure, not all counties in all states are the same size as King County, and not all have such a blended mix of city and country components.  But the point remains that country level data is often too broad, obscuring substantial variations within the counties.

How do we solve this?  We can really only start to address this issue once we’ve started to narrow down our area of searching; there’s no easy/convenient way to do it on a national level for the first few rounds of evaluation and regional scoring.  But when we start to zero in on regions, we then need to modify the county level data and subdivide it into smaller geographical sizings.

As we go through the first few exercises, we need to simply keep our mind open to two essential and opposite facts :

  • There may be excellent areas obscured within an apparently unsuitable area
  • An apparently excellent area may have, within it, bad areas to be avoided

Truly there are equal parts art and science to identifying ideal retreat locations.

Read More in Part Two

Let’s continue with a look at three scientific approaches, and see how useful they may be for our purposes.  Please now click on to Three Examples of Identifying Rural Regions, the second part of this two-part article.

Please also see our more general series of articles on choosing and evaluating the best location for your retreat.

Jun 092013
A yes/no categorization of regions by zip code into frontier or non-frontier regions, proposed by the National Center for Frontier Communities.

A yes/no categorization of regions by zip code into frontier or non-frontier regions, proposed by the National Center for Frontier Communities.

This is the second part of a two-part article about determining the rural character of any given region and therefore, its derivative suitability as a retreat location for prepper purposes.

If you arrived here direct from another site’s link or search engine, we’d recommend you first read the first part – How Rural is Your Retreat’s Region – and then come back here to read this second part.

The Index of Relative Rurality

In 2006, a researcher from Purdue University, IN, (Brigitte Waldorf) published a paper in which she detailed a measure she called the Index of Relative Rurality (abbreviated IRR) – a scale from 0.0 to 1.0 for measuring areas, with 0.0 being most urban to 1.0 being most rural.  At the time, this was an innovative new measure, and used four factors for giving a rating – population size, population density, percentage of urban residents (ie living in built up areas), and distance to the closest metropolitan area.

But it also has some major limitations.  Although its author opens her introduction by saying (and we agree with her)

Low population density, abundance of farmland, and remoteness from urban agglomerations are characteristics that people typically associate with rural places.

the problem is that her IRR scale, while considering the first and the third of the three factors, completely overlooks the second – ie, the land use in the region being measured.  We think this is an important additional factor.  She disagrees, saying on page 9

Are there additional dimensions or rurality? In the past, it may have been defendable to include the reliance on agriculture as a key dimension. However, today agriculture accounts for such a small share of economic activities overall as well as in rural areas, that it no longer qualifies as a key dimension. Similarly, many social characteristics (e.g., traditional) often associated with rural areas are —at best—outcomes but not defining dimensions of rurality.

That’s a very sad statement to make, isn’t it, and vividly and unintentionally ‘proves’ one of our key contentions – that our society is becoming unsustainably unbalanced in favor of cities.  Maybe, for her purposes (and she goes on to confess she chose simple measures with readily available data) it is appropriate to ignore agricultural issues, but for our purposes, we feel it is an essential consideration.  And, indeed, we’d have hoped that a woman based in Indiana would be more in tune with rural issues – apparently not.

Nonetheless, in the seven years since the IRR was published, there has been little work published to enhance it or to incorporate considerations about land use/lifestyle.

Another weakness of her scoring is that the author attributes equal importance to all four factors, apparently again for the sake of mathematical simplicity rather than due to any reasoned decision that they all apply equally.  If she adjusted the weighting for the four factors, or the scoring within each factor, the aggregate ratings would then change appreciably.

But, while recognizing the limitations of the Index of Relative Rurality, let’s at least look at what it shows.

Here’s a map, based on the 2000 census, showing, county by county across the nation, its IRR score.

In a follow-up article in 2007, the researcher points out an interesting additional measurement – a seven step series of levels of rural-metropolitan characterization of counties.  A map, again using 2000 census data, is here.  For our interests, we’d most prefer counties in the E, F, or G categories (defined here and detailed here.).

Here’s a link to her article where she presents this additional material.

Agricultural and Farming Issues

Although ignored by Professor Waldorf who claimed it to be irrelevant and who also implied it to be inconvenient to measure, we consider that the land use within a region is a vital measure of its suitability for a ‘grid down’ type retreat.

Unproductive land may be fine for recreational purposes, and may offer a lovely lifestyle when food, energy, and everything else is readily available and inexpensive.  But if (when?) our elaborate social support systems should collapse, it is essential that we can become independent and self-contained, able to produce our own food, our own energy, and survive with a minimum of external inputs.

So how can we now evaluate agricultural/productive land use?  Happily, and notwithstanding Professor Waldorf’s claim that such data was difficult to obtain, the USDA has some excellent information, in county by county form, and displayed helpfully on maps, dating from its 2007 rural census.

The USDA does these censuses on a five yearly cycle, so we hope the published 2007 data will soon be replaced by 2012 data.

Here’s an excellent map that shows the percentage of area, by county, that is used for farming purposes.

This encapsulates much of what we’d want to know, but it does have a few limitations.  For example, it considers farmland as a percentage of all land, and fails to distinguish between land that could theoretically be used for farming and land that could not be so used.  So if a county has a large amount of national forest/park on it, then it will be downgraded for farm area utilization, even if the remaining balance of the county is 100% farmland.

A slightly different representation of this data is on this map, showing acres of land in farms.  The more intense the blue dotting, the greater the abundance of farming.

You might also be interested to know the average size of each farm, and whether it is privately or corporately owned.  For our purposes, we might assume that we’d be more comfortable in an area of farms of similar size to whatever we would hope to own, ourselves, and we’d prefer to be among other privately owned farms than among those owned by abstract/remote corporations.  We also have a perception that smaller farms may be less reliant on high levels of automation and, as such, better able to transition to less energy intensive farming methods.

An interesting consideration is if the number of farms is increasing or decreasing.  Overall, the total count of farms increased in the five years from 2002 to 2007.  You’d probably prefer to be in a region with stable or increasing farms, rather than in an area with decreasing farms – these regions are probably suffering from encroaching urbanization.

An alternate view is here, which shows the change in number of acres being farmed.  A decrease in the number of farms, or an increase for that matter, might be due to one farm being split into two, or two farms merged into one, so the change in total number of acres being farmed is in some respects more helpful.

It is also helpful to understand what type of farm use applies in an area.

This map shows the percentage of crop type farms as measured against total farmland, and this map shows percentage of pastureland as measured against total farmland..  The two maps are more or less reciprocals or opposites of each other.

If you’d like further details about the type of farming, there are other maps on this page, and at the bottom of the page, links to a massive number more.  Here’s just one more of the many other useful insights offered by the USDA that can help you better understand the implications of one region compared to another.

These maps and the data they represent give an interesting and valuable additional perspective on ‘good’ and not so good areas to locate, but are relatively silent on the subject of population densities, so you need to balance the data here with other data, possibly the Index of Relative Rurality (see above), and/or the status of the region as being a frontier region or not, discussed next.

Frontier Communities

Although Dr Hart’s analysis of classification methodologies lists over 30 different ways to distinguish rural from urban, he is clearly interested in the concept of ‘frontier’ regions, although he is quick to point out, as we would too, that the word ‘frontier’ sometimes has irrelevant and unhelpful connotations.  We don’t mean frontier in any sort of ‘Davy Crockett’ or warlike sense at all, merely in the sense of being far away from urban living and lifestyles.

There is a National Center for Frontier Communities and they have come up with a three factor calculation for evaluating a region’s frontier status – population density, distance in miles from a market or service area, and travel time in minutes to that same market or service area.

They have asked states and in particular, the State Offices of Rural Health, to define, county by county, which counties they believe to be ‘frontier’ counties.  Thirteen states don’t define any of their counties as frontier counties, 19 use the three factor calculation, and 16 use other criteria, so the results are hardly homogenous and consistent.

The results are not very useful for two reasons.  First, they are county by county, and secondly, they are either ‘yes’ or ‘no’ rather than on a continuum with varying values.  However, you can see them here, and there is also a very interesting derivative map, showing the change in frontier status over the past 20 years.  The second map gives a good perspective of where urban sprawl is happening and moving.

In addition to this county-wide categorization, the Center is developing a new definition for ‘frontier’ and has published a draft series of four maps that are massively more valuable.

Firstly, they show varying degrees of frontier qualification, rather than the simplistic yes/no status of their current classification.  Secondly, they are more ‘granular’ – they show details down to zip code level rather than no smaller than county level.

You can see their four maps here.  In case it isn’t obvious, FAR Level One is a superset that includes all level two, three and four regions too.  Each additional level shows less and less frontier regions, and this is very helpful because you can see not only what your immediate area is categorized as, but what the adjoining areas may be as well.

Their definition for FAR Level Four is surprisingly close to something that would almost be suitable for locating a retreat – we’d doubtless prefer greater distances from larger metro areas, so we’d feel more comfortable in the middle of each block of FAR Level Four region.


This is the second part of a two-part article on how to determine the rural vs urban nature of an area.  If you’ve not already done so, we recommend you should read the first part – How Rural is Your Retreat’s Region – too.

We ended the first part by pointing out that identifying an ideal location for a retreat involves both art and science.  Now that we’ve looked at three different scientific methodologies, we continue to feel that there’s a measure of art involved, while the science component is much more complex than generally acknowledged.

Most of all, while we’ve looked at some detail into the rural vs urban differentiation in this two-part article, this is only one of very many different factors and considerations you need to consider when choosing your ideal retreat.

Please also see our more general series of articles on choosing and evaluating the best location for your retreat.