Little Big Drought

BAD DROUGHT, UNLUCKY VIRGINIA

AS OF LATE SUMMER, Northern Virginia and the Shenandoah Valley lay under the siege of one of the most profound droughts of the 20^th century, and the Piedmont—from the Fall Line to the Blue Ridge—wasn’t far behind.

Streamflow reached all time record-low flow in the Rappahannock, farm ponds turned to puddles (or completely dried up), and pastures turned to tinder. Shenandoah Valley corn turned bantam, and what stood absorbed enough nitrate to become toxic to cattle. Fish died in the tepid backwaters of Chesapeake Bay, creating a hovering stink that grew peculiarly concentrated as night air descended.

As if this weren’t all bad enough, TV couldn’t let go of the story. From ABC to the Weather Channel, it was story Numero Uno in August. Most folks turn on the tube to get away from reality, but the constant beating of the big drought story left no escape.

And now for your perhaps “Politically Incorrect” Advisory take: Too bad it’s here, but it’s what happens much of the time somewhere in the Nation.

That’s right. In terms of national severity and extent, this drought is nothing special. In fact, many “droughts” are barely noticed, because of the peculiar way we measure them. Others are pretty obvious, such as this one, especially when it’s in your backyard.

DROUGHT INDEXING

What’s a drought? Webster says, “A prolonged period of dryness. He defines “dryness” as “lacking or deficient in moisture. By that definition, most deserts are in a state of semipermanent drought, and most forests are not. And that’s probably true.

But that’s not the way climatologists measure drought. No, we have the at-least-debatable idea that drought is measured primarily by the departure from average moisture at any spot.

In other words, if a place like Washington DC receives 60% of its average rainfall, it’s likely to be in a big time drought, even though that means that it still gets 13 times the average rainfall of Death Valley, which averages a paltry 1.78 inches per year.

In fact, some of the rainiest places in the country, such as Washington’s Olympic Peninsula, can receive scores of inches of rain per year, but, because this is substantially below normal, still be in a “severe drought.”

There’s a certain logic around this. The implication is that the natural vegetation, as well as the human infrastructure put in its place, are adapted to “normal” conditions for that spot, and, the more abnormal things become, the more they fall apart. In other words, Death Valley Scotty is adapted to Death Valley and Senators to Washington, but if either experiences below normal rainfall they’re unhappy.

The Palmer Drought Severity Index

Other things besides lack of rainfall go into a drought. That’s because water either precipitates (as rain) runs off (as a creek or a river), goes into the ground, or evaporates. Each of these processes determines the local moisture departure from normal.

Evaporation is exceedingly important in this equation. In Virginia, a pan of water left out in December will evaporate about one half on an inch during the entire month, on average. In July, this figure goes up to seven inches—fourteen times as much. That’s because evaporation is highly dependent upon temperature, which itself is highly dependent upon sunlight (the central theorem of climatology is “the sun warms the earth”).

Both latitude—which determines how high the sun is in the sky during summer—and temperature determine evaporation. All of the evaporation variables are also linked in a “positive feedback loop. In general, the sunnier it is, the warmer it is, which promotes more evaporation, which makes it drier. The drier it is, the hotter it gets, as less and less of the sun’s energy is required for evaporation, the hotter it gets, the drier it is, etc.

Over forty years ago, Weather Bureau (that’s the bureaucratic ancestor of the National Weather Service) meteorologist Wayne Palmer tried to put all of these variables—evaporation, temperature, rainfall, and soil storage—together in an index to describe the local moisture status. Not surprisingly, his mathematical device is known as the Palmer Drought Severity Index, or PDSI.

The PDSI is referenced to zero. In other words, a long period of nearly average precipitation and temperature gives a reading of right around 0.0. As temperature goes further above normal and/or precipitation goes down, so does the PDSI. (When temperatures are cooler than average and/or rainfall is heavier, the index goes up.)

All of the factors that determine the overall moisture give a large range of readings. These values are then statistically “normalized,” so that equally unusual departures from normal at two different places give the same result when interpreted by the Palmer Index. In this fashion, 25 inches of rain per year in Washington DC (60% of average) has an analogous effect on the Palmer Index there as does a bit over one inch for the entire year at Death Valley.

“CLIMATOLOGICAL DIVISIONS”—DIVVYING UP VIRGINIA

Those with an eye for news might note that drought stories mentioning the Palmer Index repeatedly refer to the “Western Piedmont,” “Tidewater, ” or “Southwestern Virginia.” That’s because these are three of the six “Climatological Divisions” (CDs) that make up Virginia.

CDs have been around for a long time. In the early part of the century, Virginia had only three, labeled “Tidewater,” “Middle Virginia” and “The Great Valley.” The last one included some pretty mountainous stations rather far from the Shenandoah Valley, such as Monterey, Pennington Gap and Wytheville.

This scheme was revised in fall of 1956 to the current six CDs, in a national attempt to impose more climatic “sense.” Many divisions were assigned names that also signaled some type of climatic homogeneity that related to geography. For example, the dividing line between the new “Tidewater” and “Eastern Piedmont” districts is the Fall Line, meaning that the Tidewater region is largely coastal plain while the Piedmont is more upland.

Virginia’s six districts are still somewhat of an anomaly. Most states (for reasons yet to be divined) have nine. But all CD’s share one thing in common: Their Palmer Index values for each CD are referenced to zero and given a standard deviation of ±2.0, with –2.0 indicating “Moderate Drought.”

As a result of this process, the mean and standard deviation of the Palmer Index for each region in the U.S. are 0.0 and ±2.0, respectively. Statistically, about one-sixth of all readings are above 2.0, and about one-sixth are below. Where the index is negative, we’re on the dry side of zero, while positive values are wet.

Values between –2.0 and –2.99 are called “moderate drought. It is for the reader to decide whether 17% of the time (or one in six occurrences), counts as a “drought,” but that’s the way it works.

INTERPRETING THE PALMER DROUGHT SEVERITY INDEX
(The “Official” and the “Unofficial”)

Here’s a handy reference chart for the Palmer Index, including the “official” meaning, and the relative frequency of each category.

VALUE	“OFFICIAL” MEANING	“UNOFFICIAL” COMMENT
–4.0 or less	“Extreme” drought	Averages 4.4% of U.S each year; Your chance of being in it is one in 40.
–3.0 to –3.99	“Severe” drought	Averages 10% of U.S. each year; Your chance of being in it or worse is one in 14.
–2.0 to –2.99	“Moderate” drought	Makes newspaper does little else. This or worse in 21% of U.S. each year.
+1.99 to –1.99	“Near Normal”	Occurs less than 2/3 of the time; therefore, more than 1/3 of all readings are “abnormal” (!).
+2.0 to + 2.99	“Unusually” moist	Averages 1/6 of the time, 1/6 of the area.
+3.0 to +3.99	“Very moist” spell	Analogous to “Severe” drought.
+4.0 or more	“Extremely” moist	Analogous to “Extreme” drought.

“Moderate drought” is more a wake-up call to people like State Climatologists who are assigned public worrying duties in their states. It also is good for a few byline stories in the paper, along with the requisite photo of stunted corn. If you look hard enough, this can usually be found within 100 miles of any decent sized daily newspaper every summer.

Put another way, According to the situation, on the average (where things rarely are), 17% of this fine nation (The Palmer Index is normally only calculated for the 48 contiguous states.) should experience a moderate or worse drought at any given time. That’s a lot of land area, and surely enough to worry about almost every year.

Shaded areas are Climatic Divisions which were classified as “Extreme Drought” by the Palmer Index as of August 14, 1999.

When the index is between –3.0 and –3.99, the drought designation is “severe. Statistically, the index is at or below this value about 7% of the time, or one in 14 years, whether you are in a rain forest, a swamp like DC or in a desert like Death Valley. We again leave it to our readers to consider the rhetorical definition (“severe”) vs. the reality. Actually, around here –3.0 is where people really begin to notice. In the summer, –3.0 is likely to be accompanied by some pretty lousy pasture conditions and a lot of brown lawns.

Shaded areas are Climatic Divisions which were classified as “Extreme Drought” by the Palmer Index in January 1931.

At or below –4.0 is “extreme” drought, which, in Virginia, almost certainly translates into regional crop losses and low water tables. One way to make an eyeball calculation of –4.0 or worse conditions is that farm ponds are usually near (or at) bottom; this situation is common in the central and northern Shenandoah Valley this summer. Statistically, readings are at or below –4.0 around 2.5% of the time (or, on the average, 2.5% of the U.S. is in extreme or worse drought at any one time).

The Big Drought—Shaded areas are Climatic Divisions which were classified as “Extreme Drought” by the Palmer Index in August 1934. Note the big difference between this drought and the situation in 1999.

STATISTICS AND REALITY

We’re dreadfully familiar with the old saw that statistics can be used to lie about anything. But sometimes, blind faith in statistics themselves is the easiest way to come up with the not-quite-right answer.

It’s true that the statistically expected percent of the U.S. that experiences extreme drought should be around 2.5%. That assumes that readings are independent from month-to-month and from place-to-place. They’re not. If there’s a big drought on this month, the chance that it’s still going to be here next month is very high, because it takes a lot of rain to end a bad drought. Ditto for the CD next door. If it’s bad here, the same probably applies to your neighbor

As a result of all this, the actual percent of area covered by the various classes of drought average a bit higher than the theoretical values resulting from the statistical manipulation of the Palmer Index. We find, for example, that on the average, 21% of the nation shows a PDSI of –2.0 or lower (“moderate drought” or worse) compared to the theoretical expectation of 17%.

The same applies to extreme drought. On the average, 4.4% of the nation is in this category. That’s a lot of the U.S. lower 48 states.

COMPARATIVE FIGURES FOR 1999

Before we get much further, lets stipulate that it has been, very dry in the Shenandoah Valley and Northern Virginia. (The Piedmont has been generally merely very dry.) Farmers in the Valley have experienced major logistical and financial difficulties and deserve all help that they can get.

But, in terms of severity and extent, on the national scale, this drought has been pretty small potatoes. Smaller potatoes than average, to be precise. As of August 28, less than 3% of this fair nation was experiencing extreme drought (Palmer Index of –4.0 or lower), while the long-term average is 4.4%!

Our map, from August 14, shows the areas of extreme drought. One notable aspect of the current situation is that Washington, DC seems to be just about the bulls-eye. Perhaps that has something to do with this drought’s “importance.”

On August 7, President Clinton called this the “worst agricultural drought this century.” It turns out this depends upon what “worst” means. But in terms of physical extent, for droughts in the Mid-Atlantic region, it can’t hold a shovel to 1930, when virtually the entire northeastern quarter of the nation was in extreme drought, as shown by our map.

Finally, it’s always instructive to compare your drought to 1934. That year, almost half of the nation was in the grip of extreme drought, as shown by our figure. It hardly seems that the current drought can be fairly labeled as worse than that whopper.