Water Resources in Rockland - Planning in a Changing World

Drought in the Hudson River Valley

storm cloudsHistorical Drought Frequency

The Hudson River Valley in New York State has had average annual precipitation since 1895 of about 1065 mm/year (~ 42 inches/yr), or about 89 mm/month (~3.5 inches/month).  However, droughts in this region are not uncommon, and because of the urban/suburban populations in some of these counties, economic and social impacts of droughts can be substantial. The most detailed information available to address historical drought occurrences is from daily and monthly records of precipitation and temperature recorded at numerous stations in the region and collected by the National Oceanic and Atmospheric Administration (NOAA). There are more than 3 dozen stations currently reporting monthly precipitation in the Hudson Valley. A longer view of drought is also possible using tree ring records stretching back 500 years. Click here for April 25, 2010 presentation on Recent Hudson Valley Droughts in Historical Perspective.

Scientists, planners, and farmers define drought in many different ways. We look at droughts based on three different definitions here:


Droughts in the Past Century

New York Climate Division 5 Map

New York Climate Division 5 extends from the headwaters of the Hudson N. of Albany south to Rockland County and includes 14 counties in part or in whole.  From a single reporting precipitation station in 1895, the network of precipitation monitoring stations grew to a high of 81 sites in the fifties and then diminished to 40 in 2000. (view larger map )


Drought Defined as Departure from Average Precipitation

While there are many different ways of measuring and defining droughts, a simple one to look at is simply the departure from the average precipitation, typically taken by climatologists to be the most recent 30 year period for which data are available. The charts and table below show droughts based on Hudson valley precipitation data from 1895 - 2005. We chose this data set because of the length of the record and because NOAA has systematically addressed the missing data problem.

The base period used is 1971-2000 and droughts are defined to begin in any month when the 12 month running mean is less than the 30 year average for that month and to end in the first month where the negative departure from the 30 year mean ends. The charts and tables that follow show all droughts defined in this way that lasted more than 6 months, of which there were 28 lasting a total of 70.8 years during the 110 year period analyzed.

The 5 year running mean of annual precipitation deficits makes it graphically clear that the 1971-2000 period was considerably wetter than much of the 75 years that preceded it. During this wet end-of-century period Rockland County’s population grew 25%. The precipitation deficit by month chart gives a much more granular feel for precipitation variability in the Hudson Valley and the chart of precipitation deficit vs. drought length illustrates how moderate the recent droughts have been.

5 year running mean of annual precipitation deficits

Droughts derived from NYS Climate Division #5 exceeding 6 months in length

(view table)

Droughts based on >6 mon period of 12 month running mean negative departures from climatology (1971-2000)
Droughts derived from NYS Climate Division #5 exceeding 6 months in length
Drought # Year
Start End Cum Deficit
# Months Cum Deficit
# Yrs
1 1896 Dec-1895 Jun-1897 -249 19 249 1.58
2 1900 Aug-1899 Jul-1901 -190 24 190 2.00
3 1906 Feb-04 Feb-08 -478 49 478 4.08
4 1910 Sep-08 Apr-12 -791 44 791 3.67
5 1916 Jun-12 Nov-20 -1188 102 1188 8.50
6 1923 Jun-21 Aug-24 -502 39 502 3.25
7 1926 Oct-24 Oct-27 -473 37 473 3.08
8 1930 Nov-28 Oct-32 -715 48 715 4.00
9 1935 Mar-35 Apr-36 -87 14 87 1.17
10 1941 Aug-39 Nov-42 -475 40 475 3.33
11 1944 Sep-43 Apr-45 -150 20 150 1.67
12 1947 Jul-46 Dec-48 -168 30 168 2.50
13 1950 Mar-49 Jun-51 -345 28 345 2.33
14 1954 Jan-54 Oct-54 -70 10 70 0.83
15 1955 Dec-54 Jul-55 -160 8 160 0.67
16 1957 Oct-56 Sep-58 -347 24 347 2.00
17 1959 Dec-58 Jun-60 -204 19 204 1.58
18 1966 Jun-61 Aug-71 -2264 123 2264 10.25
19 1981 Jan-80 Mar-83 -553 39 553 3.25
20 1985 Feb-85 Jun-86 -178 17 178 1.42
21 1986 Sep-86 Mar-87 -93 7 93 0.58
22 1988 Dec-87 May-89 -233 18 233 1.50
23 1992 Jul-91 Mar-93 -86 21 86 1.75
24 1993 Jun-93 May-94 -25 12 25 1.00
25 1995 Mar-95 Mar-96 -110 13 110 1.08
26 1998 Sep-97 May-98 38 9 -38 0.75
27 1999 Nov-98 Mar-00 -84 17 84 1.42
28 2002 May-01 Nov-02 -220 19 220 1.58
Total 28       -10401 850   70.8
Total <6 months       78   6.5

Palmer Drought Severity Index (PDSI)

PDSI Drought Categories (National Drought Mitigation Center)

4.0 or more extremely wet
3.0 to 3.99 very wet
2.0 to 2.99 moderately wet
1.0 to 1.99 slightly wet
0.5 to 0.99 incipient wet spell
0.49 to -0.49 near normal
-0.5 to -0.99 incipient dry spell
-1.0 to -1.99 mild drought
-2.0 to -2.99 moderate drought
-3.0 to -3.99 severe drought
-4.0 or less extreme drought

The most commonly used drought definition in the US is the Palmer Drought Severity Index (PDSI). The PDSI is a soil moisture formula which was the first comprehensive drought index developed in the US. The index, developed in 1965, is based on the water balance equation and it takes into account the available water content of the soil as well as precipitation and temperature. Human impacts are not considered. The objective of the PDSI is to provide standardized measurements of moisture conditions, allowing meaningful comparisons between different locations and time periods. (Dr. Michael Hayes, National Drought Mitigation Center). The eleven PDSI drought categories shown here go from extreme drought through near normal to extreme wet conditions. By definition, zero values (and those between -1.0 and +1.0) on this scale represent "average" conditions for combined precipitation and temperature data. Positive values greater than +1.0 indicate wetter than average conditions, and negative values less than -1.0 indicate drought conditions.


Drought in the Northeast

% of Northeastern States in Severe or Extreme Drought

One of the most extensive sources of information about drought frequencies and intensities for the US is provided by NOAA Regional Climate Centers. The center for 12 northeastern states (ME, NH, VT, MA, CT, RI, NY, PA, NJ, MD, DE, WV) is located at Cornell University. Using monthly data from the precipitation and temperature stations for the climate divisions in each of these states from 1895 to the present, they have derived extensive statistical data on droughts based on monthly values of the Palmer Drought Severity Index (see: http:/ /met-www.cit.cornell.edu/drought/ ). This graph gives a visual picture of the frequency of severe or extreme droughts on a regional scale. (view fullsize graph )


Hudson Valley Droughts since 1895

This table show periods during which the Hudson Valley has experienced severe or extreme drought since 1895 as defined by the Palmer Drought Severity Index. There have been 15 such periods, or about one every 7 1/2 years, lasting in total 70 months, an average of about 5 months per drought occurrence. (view full table)

Droughts in the Hudson Valley based on Meteorological Records
Derived from Northeast Regional Climate Center (Cornell)
Hudson Valley Climate Division #5
Periods of 2 or More Months of Severe or Extreme Drought
Start to End Duration Lowest Lowest
Mon/yr Mon PDSI Mon
11/1908 - 1/1909 3 -3.50 Dec-08
11/1909 - 12/1909 2 -3.61 Dec-09
10/1910 - 1/1911 4 -3.20 Oct-10
4/1911 - 7/1911 4 -4.02 May-11
12/1930 - 1/1931 2 -3.15 Dec-30
10/1941 - 2/1942 5 -3.76 Nov-41
4/1942 - 5/1942 2 -3.12 Apr-42
10/1949 - 12/1949 3 -3.97 Dec-49
8/1957 - 11/1957 4 -3.54 Nov-57
10/1963 - 12/1963 3 -3.74 Oct-63
5/1964 - 9/1966 29 -6.66 Nov-64
1/1967 - 2/1967 2 -3.13 Jan-67
4/1985 - 5/1985 2 -3.38 Apr-85
8/1995 - 9/1995 2 -3.14 Sep-95
12/2001 - 2/2002 3 -3.60 Feb-02
Jan 1895-Mar 2006 70    

Periods of 2 or more months in severe or extreme drought in the Hudson Valley since 1895

Rockland County Drought Definitions

The Rockland County Department of Health has defined multiple criteria for declaring Drought Watches and Drought Emergencies.

In the first table (right) a series of precipitation thresholds are defined, based on departures from a 20 year average of ppt measured at Lake DeForest 1959-1979. These thresholds are used as one basis for declaring a drought watch, a drought alert, a drought warning, a drought emergency or a severe drought emergency.(Rockland County Sanitary Code, Article V, as summarized by Lyons et al, 2005)


Precipitation Thresholds for Rockland County drought declarations
Level of Emergency  Precipitation Threshold
STAGE I Previous two months < 60% of average
STAGE II Previous three months < 60% of average
STAGE III Previous four months < 60% of average
STAGE IV Previous six months < 60% of average
STAGE V Previous 12 months < 65% of average

A second criterion is available as an alternate basis for declaring a drought watch, a drought alert and a drought warning - the total capacity of water available for augmentation of base flow in the Ramapo River. Thresholds are based on the curves shown below:

Water available for base flow augmentation of Ramapo River (click for fullsize graph)

A third criterion is available as an alternate basis for declaring a drought alert, a drought warning, a drought emergency or a severe drought emergency - the level of water in Lake DeForest. Thresholds are based on the curves shown below:

Water level of Lake DeForest (click for fullsize graph)

Rockland County Drought Declarations

To get a sense of the number of Rockland County Drought Declarations that would have been declared historically if the current criteria were in place Lyons et al analyzed Central Park precipitation data from 1875-2001 and found that precipitation thresholds for Stage III emergencies were met more than 50 times and precipitation thresholds for Stage IV emergencies (which have never been declared) were met more than 16 times. (click for fullsize graph)


Five centuries of tree ring records

Because the impact of variations in water on tree growth can be seen and studied in tree rings, we are able to extend our knowledge of historical drought occurrences in the Hudson Valley by looking at the record of tree growth over the past 450 years.

Trees can grow to be hundreds to thousands of years old and can contain annually-resolved records of climate for centuries to millennia

tree rings


Tree-ring-Based Drought Index for the NY Metro Area

The same Palmer Drought Severity Index used in analyzing data for the past century has been calculated from 1550 to 1990 based on tree rings from the New York Metropolitan area. In addition to clearly showing the regional drought in the '60s, the data identify similarly severe droughts in 1570, 1630, 1700 and 1820. (view full-sized chart)

Tree-ring-Based Drought Index for the NY Metro Area


The estimate of the PDSI average value for July derived from tree ring data provides a single value for each calendar year and that data is plotted in black line segments connecting each year value for the entire time series of more than 400 years. Another set of July PDSI values was derived based on smoothing (similar to averaging) of each individual year over a ten year period to remove large year-to-year variations to facilitate observations of general trends. This smoothed set of values is plotted in red to help identify the major periods of drought or higher than average rainfall over the last four centuries. (Cook, unpublished data)


Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer