Freeze Dates and Growing Season Length for the Southern Piedmont, 1911-1986

Southeastern Geographer Vol. 28, No. 1, May 1988, pp. 34-45 FREEZE DATES AND GROWING SEASON LENGTH FOR THE SOUTHERN PIEDMONT, 1911-1986* Philip W. Suckling INTRODUCTION. Climate is highly variable on a wide range of spatial and temporal scales. The identification of climatic trends depends upon the particular climatic element under consideration. For example, Brinkmann has shown that mean temperature patterns and long-term trends in the length of the growing season are not necessarily linked. (J) This suggests that a full understanding of climatic change and variability requires careful consideration of individual climatic elements. For agricultural and landscaping applications, consideration of the last spring-freeze date, first fall-freeze date, and overall growing season length is very important. Variations in this aspect of climate have been studied for several regions of the United States, including the Corn Belt, (2) Wisconsin, (3) Virginia, (4) Montana, (S) Indiana, (6) Illinois, (7) Massachusetts , (S) Minnesota, (9) and New York. (JO) Spring-freeze dates were addressed for part of the Southeastern United States in a recent study, but there is a relative lack of research on freeze dates for the southern part of the country. (JJ) The purpose of this paper is to extend the previously published last spring-freeze date study by a few years and to include an assessment of first fall-freeze dates for the region. This will provide a more complete assessment of any long-term trends in the occurrence of freeze dates as well as an indication of any change in the overall growing season length. DATA. This study utilizes records from 43 climate stations in Alabama, Georgia, South Carolina, and southern North Carolina (Fig. 1). Most of these sites are located within the southern portion of the Piedmont physiographic region, while some are situated in the adjacent Coastal Plain and Appalachian Plateau. Data were obtained from the Climatological Data publications of the U.S. Weather Bureau (pre-1974) and the Na- * The technical assistance of Peter Yaukey is gratefully acknowledged. Dr. Suckling is Associate Professor of Geography at the University of Georgia in Athens, GA 30602. Vol. XXVIII, No. 1 35 Salisbury • Hendersonville ,f~ WinthropÌ"\ »Monroe Lumberton \ Valley Head«\ Saint Bernard Talladega· Clanton· Selma· Union Springs· Troy· Ozark· /Walhalla#Santuck' CheravN " ToccoaKAndersonWinnsboro? ? ^\ Newberry.. Darlington \ \ »Rome Gainesville \Gre*nwood Athens·\. Saluda Washington ¦._ »Aiken ilo Fig. 1920 1930 1940 1950 I960 1970 1980 YEAR 5. Ten-year running means of last spring-freeze and first fall-freeze (dashed lines) and growing season length (solid line) for Monroe, dates NC. Vol. XXVIII, No. 1 39 Greenwood, SC rr LU CO LU v^^ 240 0 G" 33 230 m O 220¡1 210Sin 1200^g 1 90 I920 I930 I940 I950 YEAR I960 I970 I980 Fig. 6. Ten-year running means of last spring-freeze and first fall-freeze dates (dashed lines) and growing season length (solid line) for Greenwood, SC. Newnan, GA CO LU QC LJ CO IxI y H250 a r 33 240 m O 230 ? 220 B% Q > 210 £g 2 200 Fig. 1950 I960 1970 1980 YEAR 7. Ten-year running means of last spring-freeze and first fall-freeze dates (dashed lines) and growing season length (solid line) for Newnan, GA. 40 Southeastern Geographer Hawkinsville, GA 330h CO LU 260 m Q or lu 230 « ? CO Lu 1920 1930 1940 1950 YEAR I960 1970 1980 Fig. 8. Ten-year running means of last spring-freeze and first fall-freeze dates (dashed lines) and growing season length (solid line) for Hawkinsville, GA. tions in these trend curves were evident. Growing season length was generally shorter for recent decades after distinct peaks during the 1920s/1930s and around 1950. As was the case for Monroe, the results for Selma, located on the southwestern edge of the study region, did not indicate distinct trends, although the curves exhibited greater variations than did those for Monroe (Fig. 9). Given that the graphs for Greenwood, Newnan, and Hawkinsville suggested some trend towards shorter growing seasons, mean and standard deviation values for the freeze dates and growing season lengths were calculated and compared for the first (1911-1948) and second (1949-1986) halves of the study period for all 43...