Remarkable Surface Synoptic Maps from the 1930s

BY JOHN N. RAYNER

Illustrated example from the 1930s of an unmined data source for research into mesoscale synoptic weather phenomena.

n the mid-1960s Guy-Harold Smith, an emeritus ers in 1862 when the U.S. Department of Agriculture professor at The Ohio State University, showed me (USDA) was established. Although this was down to Isome fascinating weather maps. Dr. Smith, a car- 21% by 1930, the devastating effects of the of tographer and former chair (1933-63) of geography the early 1930s raised public awareness and Congress at the university, explained that he had been involved created The Soil Conservation Service (SCS) in 1935 in analyzing data from the Muskingum River basin of under the USDA as a replacement for the Soil Erosion eastern Ohio for the Soil Conservation Service in the Service (SES) (Harlow 1994). The latter had been estab- late 1930s. He had kept copies of one particular set of lished in 1933 under the Department of Interior, largely maps, for 1500 LT 12 October 1937. They depicted at as the result of the work of Hugh Hammond Bennett high resolution a clearly defined surface warm front ly- (1881-1960), a soil scientist in the USDA (Helms 1998). ing across the region. Fortunately I had the foresight to Bennett was made chief of the new unit, a position he trace all of the data and maps onto 450 mm x 600 mm held until he retired in 1951. Incidentally, the Weather sheets. These are now reproduced in digital form. The Bureau had initiated the USDA's work in soils in 1892 present location of the originals is unknown. (Buie 1980), the year after it was transferred from the USDA from the Army Signal Corps, where it was SOURCE OF THE DATA. A significant portion of known as the Weather Service. the U.S. public has always had a concern for the natural In the 1930s a number of research projects in- environment. Perhaps that was inevitable in the early volving the conservation of natural resources were days of the nation when most people worked on the initiated or were already in progress. Concern over land. Indeed, 58% of the people were considered farm- the need for employment during the Depression produced additional financial support from the Federal Government. Several of these projects under AFFILIATIONS: RAYNER—Geography and Atmospheric Sciences, the USDA were unified by its secretary in 1935 (Buie The Ohio State University, Colombus, Ohio 1980). For example, 10 erosion-control experiment CORRESPONDING AUTHOR: John N. Rayner, 5185 Ashford stations, including Zanesville, Ohio, were trans- Road, Dublin, OH 43017 E-mail: [email protected] ferred to the SES. In the same year, the Emergency DOI: 10.1175/BAMS-86-11-1603 Conservation Work camps (manned by the Civilian Conservation Corps) were also transferred to the SES In final form 14 July 2005 ©2005 American Meteorological Society from the Forest Service. Buie (1980) indicated that these camps and the Works Progress Administration

AMERICAN METEOROLOGICAL SOCIETY NOVEMBER 2005 BAFft I 1603

Unauthenticated | Downloaded 10/05/21 12:47 PM UTC (WPA) crews provided manual labor for the research PROJECT, OHIO. 1937-42. Approx. 82,000 items. projects. Fastened into 229 loose-leaf notebooks, plus loose Helms (1992) stated that by mid-1936 there were sheets, and arranged by subject as shown in Appen- 147 demonstration projects, 48 nurseries, 23 ex- dix J. Entries include . .. (available online at www. periment stations, 454 Civilian Conservation Corps colby.edu/sts/97guide/naral I4.html) camps, and over 23,000 WPA workers on the job. It was in this environment that sufficient funding and The format of the original maps in this large personnel were made available for the collection of collection appear to be similar. Each displays the closely spaced weather measurements, one set of hand-plotted data and isopleths for the watershed, which is presented in this paper. which is approximately 400 mm wide x 450 mm Jenkins (1976) gives an excellent description of long on paper. This represents an area about 150 km the Muskingum project. Following is a summary of east-west x 190 km north-south..The Muskingum his findings. River basin occupies 20,577 km2. The set of maps The SCS established the climatic study in April presented here is missing from the archive [R. Smith 1937, in cooperation with the Weather Bureau and 2004, National Archives (where the records may be the WPA, from which funds for labor were ob- reviewed), personal communication]. tained. Dr. C. Warren Thornthwaite, who was chief The original maps were six in number. A base map of Climate and Physiographic Research in the SCS outlined the basin with cities, streams, and bound- (1935-42), directed the project that was supervised aries. Five plotted and analyzed sheets for 1500 LT by Leonard B. Corwin. 12 October 1937 consisted of temperature in whole Weather stations were established at approxi- degrees Fahrenheit, wind with direction vectors and mately 4-mile intervals throughout the watershed speed given in whole miles per hour, cloud amount with more than 500 WPA-paid weather observers. in tenths, precipitation for 1430-1500 LT in hun- Each station was equipped with thermometers, hy- dredths, including "T" for trace, and precipitation grometers, recording rain gauges, and instruments for 1230-1500 LT. for determining wind velocity and direction. The Although the index at the National Archives in- observers made half-hourly reports to project offices dicates that there were 500 stations in the watershed, at New Philadelphia and Akron, Ohio, where the data were converted to color maps on an assembly line basis. Clear days produced over 60,000 weather readings, which were augmented on rainy days to nearly 90,000. As Jenkins (1976) states,

the hundreds of maps,... when assembled consecu- tively, [produced] the effect of a motion picture ... They illustrated practically all of the atmospheric conditions and showed how storms behave . . . [I]t was possible to trace the individual storms as they entered and crossed the watershed, to observe their behavior, determine their water contribution to each part of the entire area, and ultimately their effect upon soil erosion and flood control.

The project ended in February 1941, and much of the equipment was given to the Weather Bureau. The extensive records from the Muskingum River watershed are currently held at the National Archives in College Park, Maryland, and are described by the National Archives and Records Administration (RG-114) Records of the Soil Conservation Service:

79. WEATHER OBSERVATION MAPS OF THE FIG. I. The location of observing stations and key places MUSKINGUM WATERSHED MICROCLIMATIC for the Muskingum watershed.

1604 I BAF15- NOVEMBER 2005

Unauthenticated | Downloaded 10/05/21 12:47 PM UTC less than 400 were plotted on each TABLE 1. List of places. map. Some stations appeared on one or two maps and not on others. Name Symbol Latitude Longitude

Akron A 41 °05'N 8I°30'W DIGITAL MAPS. The hand plots Cambridge C 40°00'N 8I°35'W were overlaid with a rectangular grid to ascertain the relative coordi- Canton O 40°50'N 8I°25'W nates of each station, the watershed Coshocton Hydrologic St. H 40°I6'N 8I°52'W boundary, and the cities. Then, the Mansfield F 40°45'N 82°30'W data were added to each station file. Marietta M 39°25'N 8I°30'W As a preliminary step the numeri- Newark N 40°05'N 82°25'W cal values were printed on maps by computer in order to check for er- New Philadelphia P 40°30'N 8I°39'W rors. Analysis was performed under Zanesville Z 39°55'N 82°00'W GNU/Linux using PGPLOT-5.2, a software program written and made freely available by Tim Pearson at the California front has been superimposed on the temperature, Institute of Technology. cloud, and precipitation maps as a dashed line. One Comparison of the computer maps with the possible reason for the warm tongues extending original subjective analyses reveals a few significant north of the wind front is that the front may have differences, especially in the 2.5-h precipitation map. extended farther to the north earlier, and some In all cases, these differences appear to be due to the warm air remained as the system moved southward. cartographic license taken by the original analyst. Earlier maps in the sequence may confirm this. Because the pattern produced by the half-hour pre- Daily maximum and minimum temperature records cipitation data is very similar to that of the 2.5-h one, for Columbus, Ohio, which lies on Interstate 70 that map is omitted. about 50 km to the west of Newark, reveal that the The locations of the observing points are presented warm air existed in the region for less than 1 day in Fig. 1. Their distribution is fairly uniform over the watershed, but not in a 4-mile-interval array as suggested by Jenkins (1976). The National Archives indicates that a station index and maps are avail- able. To these points have been added the watershed boundary and the major cities and places mentioned in the text. These are listed in Table 1. Interstate 70 runs west-east 13 km south of Newark through Zanesville, and 1 km south of Cambridge, Ohio. Interstate 77 runs north-south through Akron, Can- ton, New Philadelphia, and 1 km east of Cambridge. The Muskingum River runs southward to Marietta, Ohio, where it joins the Ohio River. By far, the most impressive feature of these maps is the warm front (Fig. 2) that runs circuitously from west-southwest to east-northeast through the center of the watershed. Temperatures vary generally from 48°F (9°C) in the north to 74°F (23°C) 160 km to the south, but the gradient is as high as 10°F (5.5°C) over 8 km at the front. There is no apparent cause for the irregular line of the front. It does not conform at all to the relief, which is on the order of 90 m within the basin. The wind map (Fig. 3) clearly delineates the front, but its line does not coincide with the oscillations FIG. 2. Temperature for the Muskingum watershed at in the temperature front. For comparison the wind 1500 LT 12 Oct 1937.

AMERICAN METEOROLOGICAL SOCIETY NOVEMBER 2005 BAflS* | 1605

Unauthenticated | Downloaded 10/05/21 12:47 PM UTC Unauthenticated | Downloaded 10/05/21 12:47 PM UTC

FIG. 3. Winds for the Muskingum watershed at 1500 FIG. 4. Cloud amount for the Muskingum watershed at LT 12 Oct 1937. 1500 LT 12 Oct 1937.

(Table 2). Clearly, more research into the way the were other regions where data were collected in the elements of frontal boundaries move across the late 1930s. Are there similar interesting maps to be Earth's surface is needed. discovered? These maps alone raise research ques- As might be expected, the cloud shield was fairly tions with regard to frontal boundaries to which continuous over the front (Fig. 4). Unfortunately there current technology might be applied. are no records of cloud type. The raw data and maps for the Muskingum have The precipitation for the previous 2.5 h, which not been published previously, although the weather varied from a trace to 0.38 in. (9.65 mm), showed a data and measurements of other variables, such as typical pattern for overrunning warm air with all runoff, were used in subsequent research projects. the rainfall several kilometers north of the wind For example, the superviser and lead researcher, front (Fig. 5). Thornthwaite, used them for "a climatic index of soil-moisture conditions" (Thornthwaite et al. COMMENT. These are remarkably detailed maps 1938). Mather and Sanderson (1996) elaborate on of a front. Even with the enormous amount of infor- Thornthwaite's central role in this project and how mation that is available from remote sensing systems it related to his broader research into the hydrologic today the author has not seen comparable published cycle. He also probably used the data as a basis for his images. This is only one set from 3.5 yr of data for classification of climate (Thornthwaite 1948), which the Muskingum watershed. The occurrence of a front has been used extensively worldwide over the last in the region will be infrequent, but it should not be 57 yr. Unfortunately, in the latter article he did not unusual. Also, from the Helms (1992) account there identify the sources of his data. This one example set of maps from the National Archives reveals that there is a wealth of valuable TABLE 2. Temperatu res at Columbus (°F). synoptic weather and associated data from nearly Oct 9 10 II 12 13 14 15 70 yr ago, which is still awaiting researchers.

Max 48 57 58 73 48 46 51 ACKNOWLEDGMENTS. Several very useful sugges- Min 39 38 47 45 37 32 31 tions with detailed references were made by the referees.

1606 | BAFFT november 2005 REFERENCES Buie, E. C., 1980: A History of Depart- ment of Agriculture Water Resources Activities. U.S. Government Printing Office, 172 pp. Harlow, J. T., 1994: History of the Soil Conservation Service national resource inventories. Soil Conserva- tion Service, U.S. Government Printing Office, Tech. Publication, 11 pp. Helms, D., Ed., 1992: Readings in the History of the Soil Conservation Service. United States Department of Agriculture, Soil Conservation Service, Econom- ics and Social Sciences Division, U.S. Government Printing Office, 176 pp. , 1998: Natural Resources Conservation Service brief history. A Historical Guide to the US. Government, G. T. Kurian, Ed., Oxford University Press, 434-439. Jenkins, H., 1976: A Valley Renewed: A History of the Muskingum Watershed Conservancy. The Kent State University Press, 206 pp. Mather, J. R., and M. Sanderson, 1996: The Genius of C. Warren Thornthwaite. University of Oklahoma Press, 226 pp. FIG. 5. Precipitation for the Muskingum watershed at xhornthwaite) C w>> 1948: An approach toward a ratio_ 1230-1500 LT 12 Oct 1937. . . .r.. r .. ^ n rr nal classification or climate. Geogr. Rev., 38, 55-94. , B. Holzman, and D. I. Blumenstock, 1938: Climatic The author has attempted to incorporate all of these and research in the soil conservation service. Mon. Wea. believes that the paper has been enhanced as a result. Rev., 66, 351-368.

In its ongoing commitment to education, the AMS has established the Vonnegut/Schaefer Freshman Scholarship, an endowed scholarship that will be awarded to a student entering his or her first year of undergraduate study in the atmospheric or related sciences. Bernie Vonnegut and Vince Schaefer were colleagues and friends who for many years made significant contributions to science and meteorology and continuously shared their zest for learning and encouragement to their students. To help launch the scholarship, a longtime AMS member (and the ate student of Bernard Vonnegut) has pledged a two-for-one challenge match of $50,000 over the next two years. For every dollar contributed to the Vonnegut/Schaefer Freshman Scholarship, it will be matched with a $2 gift. To contribute, please send your check (made payable to the Ameri- can Meteorological Society) to AMS, attn: Vonnegut/Schaefer Scholar- ship, 45 Beacon Street, Boston, MA 02108. Questions may be directed to a to make Stephanie Armstrong, Director of Development, 617-227-2426 ext. 235; ' '^uouac [email protected].

AMERICAN METEOROLOGICAL SOCIETY NOVEMBER 2005 BAflS* | 1607

Unauthenticated | Downloaded 10/05/21 12:47 PM UTC N9RTF1EAST SN9WST9RMS Volume I: Overview

PAUL J. K9CIM & L9UI5 W. UCCCLLIMI Published by the A; Mete i

N9RTF1EAST 5H?W5T?RRI5 Volume I: Overview Volume II: The Cases Paul J. Kocin and Louis W. Uccellini

Northeast Snowstorms offers the most comprehensive treatment on winter storms ever compiled: more than 50 years of professional experience in the form of a two-volume com pendium of insights, examples, photographs, over 200 color figures, and a DVD of added material.

American Meteorological Society; 818 pages, hardbound; AMS code MM54

Price: $100.00 list/$80.00 member/ $60.00 student member

Submit your prepaid orders to AMS, Attn: Order Dept, 45 Beacon St., Boston, MA 02108-3693. Please make checks AMERICAN payable to the American Meteorological Society. If using a credit METEOROLOGICAL card (MasterCard, VISA, and American Express accepted), you \ SOCIETY may place your order by phone to 617-227-2426 x204 or x237 or by email to [email protected].

1608 I BAF15- NOVEMBER 2005

Unauthenticated | Downloaded 10/05/21 12:47 PM UTC