Drought, Forests and Agriculture in Connecticut
By Dr. David Miller
The University of Connecticut, 2002

According to the Hartford Courant (Wednesday, April 3, 2002), Connecticut officials have recently issued the first ever statewide drought advisory, asking residents and state agencies to voluntarily conserve water. The advisory was issued by the state departments of Environmental Protection, Public Health and Public Utility Control in response to low reservoir levels and streams that are drying up. Although there is a statewide concern for a “water supply” shortage, Connecticut forests and agriculture are showing little effect so far from this reduced precipitation. This article outlines some of the reasons for this disparity and suggests that there is little to worry about in terms of natural and cultivated plant growth at this time.
Over long periods it is drier than average about half the time. Therefore natural vegetation experiences reoccurring “dry” periods, of varying intensity, throughout their development. Therefore the forest communities acclimatize and develop resistance to large scale damage by all but the most severe and prolonged droughts. The droughts are part of the natural environment to which the forest biological community is constantly reacting and changing. Horticultural plants, especially ornamentals, on the other hand tend to be non-native and not “drought-hardened”. Therefore, without supplemental watering, most will suffer significant water stress damage in even short dry periods.
We, in the natural resources fields, generally define “water supply” drought intensity by the Palmer Drought Severity Index (PDSI). PDSI is calculated weekly for the United States by NOAA (Figure 1), The PDSI is a soil moisture algorithm calibrated for relatively homogeneous regions. It is used by many U.S. government agencies and states which rely on the PDSI to trigger drought relief programs. But its use for predicting plant vulnerability and damage is questionable. The PDSI values may lag emerging droughts by several months. The PDSI is a meteorological drought index and responds to weather conditions that have been abnormally dry or abnormally wet. We can see that our region, as of 30 March 2002 is only in moderate drought conditions according to the PDSI.

(Insert Figure 1 here)
The key to forest versus dry land agriculture vulnerability is not only the severity of the drought, but its timing. Forest productivity is impacted most by early growing season dry periods because this is the time of fastest tree growth. Whereas agriculture droughts tend to be most serious when the plants are forming or filling their seed, often in mid-summer in Connecticut. It is no accident that the relative amount of irrigated acreage of crops has been increasing in the relative wet Northeast. Even though rainfall supplements are only needed a few times a year, relief from just a few weeks of dry weather at a critical period during crop development can mean the difference between a high and low yield.
In Connecticut the general rainfall has been increasing gradually over the last 100 years as shown in Figure 2. Therefore we would expect the frequency of droughts as defined by the PDSI to be decreasing. As a result the forests might be expected to be increasing their productivity due to gradual decreases in the frequency and amount of water stress. But that is not likely. Over the century, the natural forest vegetation is most likely to be gradually acclimatizing to the higher rainfall. Thus the periodic droughts are likely to have similar effects on the natural plants as those earlier in the century because they are of similar intensity relative to the average. Since natural vegetation is acclimatized to these short term periodic weather cycles, ecosystems in general are not likely to be changed significantly by weather cycles (i.e. droughts or wet periods) with durations shorter than a decade.
(Insert Figure 2 here- This is figure 1 from the CT Rain Pub – the graph of annual Total Precip from 1895-1996).
(Insert Figure 3 here – this is the figure from the CT Rain Pub that has box and whisker graphs for each month of the year)
On average, Connecticut has ample rainfall all year around, about 4 inches each month, as shown in figure 3. But the variability, which is also shown in figure 3, indicates any given month can deliver no rainfall or enough rainfall to flood the landscape. Thus season-length droughts are quite frequent and if severe enough and the timing is right, can cause short-term reactions in the forest systems. In Connecticut forests the under story vegetation (e.g., relatively short, shade tolerent species, seedlings and saplings) is the most sensitive to reductions in rainfall. Shallow-rooted species, such as birch and dogwood, react quickly to emerging droughts by shedding leaves and showing signs of dormancy. This causes reduced growth and water use and increased mortality in the drought periods. While deep rooted trees, such as our large over story oaks, show some decrease in diameter growth during dry years but little decrease in transpiration. Only in sever multi-year droughts do our over story oaks demonstrate damage from which they cannot quickly recover. Below-ground root production, a basic requirement for optimum plant growth, is often reversed by root dieback during drought periods. Species composition is often dependent upon which species occupy the root space first after a drought.
Ranking our over story trees by their sensitivity to dry soil produces the same results as ranking by rooting depth from shallow to deep. For example, in general, the birches are more sensitive than poplars, which are more sensitive than maples, which are more sensitive than oaks. These relative sensitivities carry over into a variety of processes such as transpiration, photosynthesis and growth. In locations where less sensitive species make up much of the under story, mortality of the relatively more sensitivity species may result in long term changes in species composition.
With respect to dryland crops, timing is most important. For example, a few dry weeks during corn tasseling has a much larger effect on yield than any other time during the growing season. Whereas, in the midst of a several year drought, two or three inches of rain before and during tasseling may result in a respectable yield.
For more information on Plant Water Use and Droughts check the essay “Evapotranspiration and Drought: by Ronald Hanson on the USGS web page http://geochange.er.usgs.gov/sw/changes/natural/et/.
For more information on Rainfall and Dry Period Frequencies in Connecticut check the publication “Rainfall In Connecticut” by Miller et al. on the CT IWR web.
For more information on the Palmer Drought Severity Index and other Drought Indices check the paper by Dr. Michael J. Hayes, Climate Impacts Specialist, National Drought Mitigation Center. http://drought.unl.edu/ndmc/enigma/indices.htm