The 2007-2009 Drought

by Peter Corrigan

The Drought of 2007-2009 which affected most of the Blacksburg/Roanoke WFO Hydrologic Service Area (HSA) was the worst since the 2000-2002 drought. By some measures it surpassed that drought, certainly in terms of duration if not intensity. This article provides a retrospective of this historic drought and provides some insight into the causes.

Dry conditions first began affecting the HSA in early 2007 with a fairly dry winter but it was the very dry May of 2007 that really began to see drought taking hold across the area. That month the average rainfall using all 77 stations from the NWS Cooperative Network was 2.07 inches or less than 50 percent of the normal May rainfall of 4.42 inches. For this article all references to normal monthly precipitation in the HSA is for the 30-year climate normals from 1971-2000 as defined by the National Climatic Data Center (NCDC). The dry May caused the introduction of D0 (Abnormally Dry) and then D1 (Moderate Drought) conditions on the weekly issuance of the U.S. Drought Monitor (USDM) into the southern portions of the HSA in mid-May. For more information on the USDM and the specific drought categories please visit the Drought Monitor Page.

The drought intensified quickly in June 2007 when D2 or Severe Drought was introduced USDM – June 5) into far northwest North Carolina across portions of Ashe and Watauga counties. Although the drought was confined to this extreme southwest corner of the Blacksburg HSA through much of mid-summer it expanded considerably during the very dry August of 2007. That month HSA average rainfall was only 1.63 inches with many stations recording less than 1 inch of rainfall. To make matters worse August 2007 was one of the hottest months ever recorded in this region, with numerous heat records being established. By early September, D3 or Extreme Drought reached up into nearly all the North Carolina counties while much of the southeastern U.S. was locked up in a devastating D4 or Exceptional Drought that made headlines across the nation. September 2007 continued to be very dry across the HSA and by early October saw the first appearance of the worst drought category, D4 in the North Carolina counties of the HSA (see Figure 1 below). The August-September 2007 two-month period was the driest on record at both Blacksburg and Bluefield, WV.

Drought Monitor from October 16th, 2007

Figure 1. USDM – October 16, 2007: D4 edges into the southern counties of the HSA

One notable factor in prolonging the drought was the lack of tropical activity in any portion of the southern Appalachians and mid-Atlantic during the fall of 2007.  The remainder of 2007 saw D3 or D4 seemingly a permanent feature over the Carolinas while the Virginia and West Virginia counties remained in much better shape with very little drought in our far northern HSA. The year 2007 was the driest across the Blacksburg HSA since 2001 with only 82 per cent of normal precipitation. At the official observation point at Roanoke Airport 2007 was the driest (29.86 inches) year since 1948, when records at the airport were begun. Danville had its 4th driest since 1948 with 33.87 inches. In addition to the dryness, 2007 was also very warm, the warmest on record at both Roanoke and Bluefield, WV. The worst drought conditions over the Carolinas were realized in late December 2007 (see Figure 2 below) and early January 2008 with all the counties in D3 or D4 drought. The Virginia and West Virginia counties continued to escape the worst conditions however. The Palmer Drought Severity Index (PDSI) reached a monthly average of -3.97 for Virginia climate division 6 (Southwest Mountains), -3.96 for North Carolina division 3 (Piedmont) and -3.06 for NC division 2 (NW mountains). The PDSI varies from +4 to -4 with anything -3 to -4 representing Severe Drought.

Drought Monitor of December 25th, 2007

Figure 2. USDM – December 25, 2007: D4 occupies nearly all the NC counties of the HSA

2008 began about where the previous year left off, with a very dry January. HSA-wide precipitation was only 1.46 inches or less than half of normal. D3/D4 drought persisted most of the winter in North Carolina while D1/D2 pushed into Virginia. The period from April 2007 through March 2008 was the driest April-March period on record in North Carolina. Fortunately, April 2008 saw rains near normal and by mid-April D2 was confined to only a very small portion of the HSA in North Carolina. Conditions worsened yet again however, as June 2008 was a fairly dry month (2.77 inches versus normal of  3.92 inches) and Severe Drought  (D3) returned back to into the far southern HSA. Short of a tropical storm, droughts can be difficult to break during summer months as water demand is very high. Thus, the summer of 2008 saw a slow deterioration in the drought situation, again worst over the mountains of North Carolina. Figure 3 below shows the very low streamflows across the southeastern U.S. by July 2008 reaching well north into Virginia.

Streamflow Map of July 2008

Figure 3.  Map of streamflow for July 2008. Dark red shades are below average flows.

It was the passage of tropical remnants from Hurricane Fay in late August that provided very badly needed moisture at a time when the drought had perhaps reached its nadir. Prior to Fay numerous streams and rivers were at all time low flows for the date, but anywhere from 2 to 8 inches of rain from Fay provided a 1-category improvement in the drought in nearly all areas by the beginning of September. D2 was pushed back into the mountains of North Carolina for much of the fall. October 2008 was another very dry month however, with 1.65 inches or less than 50% of normal rainfall (normal 3.38inches). But November was also dry and by mid- to late month D2 had expanded back into southwest Virginia as far north as Roanoke and into far southeastern West Virginia. December finally brought some relief with above normal rains, followed by a normal January. The combination of 2 consecutive normal or above winter months was enough to bring even the southwestern HSA out of Severe Drought by the middle of December (December 16 USDM). Thus, using D2 criteria the drought persisted from June 5, 2007 to December 8, 2008 with D2 or worse conditions in at least some portion of the HSA. This was a period of 79 weeks or roughly a year and a half. The duration of D1 or worse continues with a small pocket persisting in the far southwest corner of Ashe and Watauga counties where it actually began back in May, 2007. That is 100 weeks and counting as of the April 7, 2009 USDM.

Comparing this drought with the 2001-2002 drought is somewhat difficult but it has clearly been longer although the areal extent of D4 was not quite as great. That drought reached its worst point during the summer of 2002 when D4 or Exceptional Drought covered portions of south central VA and north central North Carolina. However, the duration of D2 or worse was “only” 51 weeks from November 20, 2001 to November 5, 2002. The drought ended gradually during a wet October- November, 2002.

What Caused the Drought?

The most compelling and perhaps simplest explanation is to ascribe the drought to La Niña which is also known as the Cold Phase of the El Niño/Southern Oscillation or ENSO. The onset of this drought which began in May or June 2007 corresponded remarkably well with the near simultaneous onset of an official La Niña as defined by the U.S. Climate Prediction Center (CPC). The CPC definition is for a -0.5oC or greater sea-surface temperature anomaly in a specified area of the central Pacific Ocean persisting over a 3-month average. This so-called Oceanic Niño Index or ONI reached
-0.4o in the July-August-September 2007 three-month period and -0.7o in the August-September-October period, shortly after drought began in the southeastern U.S.  The ONI reached its lowest value of the ENSO event at -1.4o in January-February-March 2008 which was its lowest value since 1999-2000 (another period of significant drought) when it reached a value of -1.6o. An ONI greater than +(-) 1.5 is considered a strong ENSO event, while +(-) 1.0 is a moderate event. The figure below shows major jet stream patterns and associated weather anomalies during moderate to strong El Niño and La Niña events across the U.S. The tendency for dry winter conditions in the southeastern U.S. during moderate to strong La Niña events has now been fairly well established and certainly this relationship held true during parts of the 2007-2008 drought. The La Niña weakened considerably during the second half of 2008 and the ONI actually reached 0 or neutral by July-August-September. Interestingly, La Niña made somewhat of a comeback in recent months reaching -0.8 during the recent winter. This may have contributed to the very dry February 2009, which saw only 1.19 inches of precipitation or about 35 percent of normal. Forecasts from the CPC indicate a weakening of La Niña this spring and summer. Of course other climate and meteorological factors are undoubtedly at work in determining seasonal precipitation and drought patterns in any region and it is not suggested that ENSO variations alone can  explain more than a portion.

El Niño and La Niña - related winter features over North America

Figure 4. Seasonal (cool season) anomalies and atmospheric patterns associated with moderate to strong Warm (El Niño) and Cool (La Niña) ENSO  events across the United States.