PART III.--EXERCISES IN THE CONSTRUCTION OF WEATHER MAPS.

CHAPTER IV.

THE DAILY WEATHER MAP.

The first daily weather maps were issued in connection with the Great Exhibition of 1851 in London. The data were collected by the Electric Telegraph Company and transmitted to London over its wires. These maps were published and sold daily (excepting Sundays) from Aug. 8 to Oct. 11, 1851. The first official weather map of the United States Weather Service was prepared in ma.n.u.script on Nov. 1, 1870, and on Jan. 14, 1871, the work of manifolding the maps for distribution was begun at Washington. Previous to the publication of this government map, Professor Cleveland Abbe had issued in Cincinnati, with the support of the Chamber of Commerce of that city, the first current weather maps published in the United States (Feb.

24 to Dec. 10, 1870). In France, daily weather maps have been published continuously since Sept. 16, 1863.

Two things are essential for the publication of a daily synoptic weather map; _first_, simultaneous meteorological observations over an extended area; and, _second_, the immediate collection of these observations by telegraph. The weather map of the United States is based on simultaneous observations made at about 150 stations in different parts of this country, besides several cooperating stations in Canada, Central America, Mexico, and the West Indies. At each of our stations, whose location may be seen on any weather map, the Weather Bureau employs one or more observers, who, twice a day, at 8 A.M. and 8 P.M., "Eastern Standard Time," make regular observations of the ordinary weather elements, _i.e._, temperature, pressure, humidity, wind direction and velocity, precipitation, cloudiness, etc. The instruments at these stations are all standard, but the completeness of the equipment varies according to the importance of the station. The 8 A.M. observations are the only ones now generally used in the preparation of weather maps. When the Weather Service was first established, tri-daily charts were for some time issued from the central office in Washington. On April 1, 1888, the number was reduced to two a day, and on Sept. 30, 1895, a further change was made, and now there is but one map a day.

The 8 A.M. observations, as soon as made, are corrected for certain instrumental errors, and the barometer readings are reduced to sea level.

The data are then put into cipher, not for secrecy, but to facilitate transmission and to lessen the chances of error, and are telegraphed from all parts of the country to the central office of the Weather Bureau in Washington. Besides sending their own messages to Washington, all the important stations of the Weather Bureau receive, by a carefully devised system of telegraphic circuits, a sufficient number of the reports from other stations to enable their observers to draw and issue local weather maps.

The observations are received at the central office of the Weather Bureau in Washington by special wires, and are usually all there within an hour after the readings were made. As the messages are received in the forecast room, they are translated from the cipher back again into the original form, and the data are entered upon blank maps. The official charged with making the forecasts then draws upon the maps lines of equal temperature, lines of equal pressure, lines of equal pressure-change and temperature-change during the past 24 hours. These several sets of lines, together with those showing the regions of precipitation during the past 24 hours, furnish the necessary data on which the forecasts can be based.

In other words, the forecast official has before him, on the several maps, a bird's-eye view of the weather conditions over the United States as they were an hour before, and also of the changes that have taken place in these conditions during the preceding 24 hours. Thus, by knowing the general laws which govern the movements of areas of high and low temperature, of fair and stormy weather, across the country, he can make a prediction as to the probable conditions which any state or section of the country will experience in 12, 24, or 36 hours.

In a later chapter some suggestions will be given for studies of forecasting.

The forecasts made in Washington, and printed on the Washington daily weather map, relate to all sections of the United States, and include predictions of cold waves, killing frosts, storm winds, river floods, and the like, besides the ordinary changes in weather conditions. These forecasts, as soon as made, are at once given to the local newspapers and to the press a.s.sociations. They are also sent by telegraph to all regular stations of the Weather Bureau, and to all stations at which cautionary or storm signals are to be displayed, along the Atlantic or Gulf coasts, and on the Great Lakes.

The Washington weather map is about 24 by 16 inches in size, and is newly lithographed each day. The total number of maps issued from the central office during the fiscal year ending June 30, 1898, was 310,250. In addition to these, there are now 84 stations of the Weather Bureau in different parts of the country, at which daily weather maps are issued and local forecasts made. These latter forecasts are made by a corps of local forecast officials, each of whom has to make the weather prediction for his own district. At first, and until within a few years, one predicting officer in Washington made all the forecasts for the country, but it was found better to have the country divided into geographical sections, over each one of which the meteorological conditions are fairly similar, and to have a local forecast official in charge of each section. These local forecast officials have the double advantage of being able to study the weather conditions over the whole country, as sent them by telegraph each morning, and also of knowing the special peculiarities of their own regions. This enables them to make more accurate predictions than can be made by an official who may be one or two thousand miles distant, in Washington.

The greater portion of the maps issued at the map stations outside of Washington are prepared by what is known as the chalk-plate process, suggested by Mr. J. W. Smith, local forecast official at Boston. This process is as follows: A thin covering of specially prepared chalk, 1/8 of an inch in thickness, is spread upon a steel plate of the size of the prospective weather map. On this chalk are engraved, by means of suitable instruments, the various weather symbols, the lines of equal pressure and of equal temperature, and the wind arrows. The plate is then stereotyped in the ordinary way, and printed on a sheet prepared for the purpose, which has a blank outline map of the United States at the top, and s.p.a.ce in the lower half for the forecasts, summary, and tables.

The size of the chalk-plate map itself is 10 by 6-1/2 inches; the size of the whole sheet, which includes also the text and tables, 16 by 11 inches.

Weather maps prepared by the chalk-plate process are now issued from 28 of the 84 stations which publish daily maps. At the remaining stations the maps are prepared by a stencil process, the size of the map being 13-1/2 by 22 inches. The total number of weather maps issued at the various stations during the fiscal year 1897-1898 was 5,239,300.

Besides recording the usual meteorological data, and publishing weather maps and forecasts, the various stations of the Weather Bureau serve as distributing centers for cold wave, frost, flood, and storm warnings.

These warnings are promptly sent out by telegraph, telephone, and mail.

Besides these usual methods of distributing forecasts, other means have also been adopted. In some places factory whistles are employed to inform those within hearing as to the coming weather; railway trains are provided with flags, whose various colors announce to those who are near the train fair or stormy weather, rising or falling temperature; and at numerous so-called "display stations," scattered all over the country, the forecasts are widely disseminated by means of flags.

CHAPTER V.

TEMPERATURE.

_A._ =Lines of Equal Temperature.=--Temperature is the most important of all the weather elements. It is therefore with a study of the distribution of temperature over the United States, and of the manner of representing that distribution, that we begin our exercises in map drawing. In carrying out the work we shall proceed in a way similar to that adopted by the officials of the Weather Bureau in Washington and at the other map-publishing stations over the country.

Enter on a blank weather map the temperature readings found in the first column of the table in Chapter VIII. These readings are given in degrees of the ordinary Fahrenheit scale [those which are preceded by the minus sign (-) being below zero], and were made at the same time (7 A.M., "Eastern Standard Time") all over the United States. Make your figures small but distinct, and place them close to the different stations to which they belong. This is done every morning at the Weather Bureau in Washington, when the telegraphic reports of weather conditions come in from all over the country. When all the temperature readings have been entered on the outline map, you have before you a view of the actual temperature distribution over the United States at 7 A.M., on the first day of the series. Describe the distribution of temperature in general terms, comparing and contrasting the different sections of the country in respect to their temperature conditions. Where are the lowest temperatures? Where are the highest? What was the lowest thermometer reading recorded anywhere on the morning of this day? At what station was this reading made? What was the highest temperature recorded? And at what station was this reading made?

Notice that the warmest districts on the map are in Florida, along the Gulf Coast, and along the coast of California. The marked contrasts in temperature between the Northwest and the Pacific and Gulf Coasts at once suggest a reason why Florida and Southern California are favorite winter resorts. To these favored districts great numbers of people who wish to escape the severe cold of winter in the Northern States travel every year, and here they enjoy mild temperature and prevailingly sunny weather. To the cold Northwest, on the other hand, far from the warm waters of the Pacific, where the days are short and the sun stands low in the sky, no seekers after health travel. This annual winter migration from the cities of the North to Florida and Southern California has led to the building of great hotels in favored locations in these States, and during the winter and spring fast express trains, splendidly equipped, are run from north to south and from south to north along the Atlantic Coast to accommodate the great numbers of travelers between New York, Philadelphia, Boston, Chicago, and other large northern cities, and the Florida winter resorts. Southern California also is rapidly developing as a winter resort, and rivals the far-famed Riviera of Southern Europe as a mild and sunny retreat from the severe climates of the more northern lat.i.tudes. The control which meteorological conditions exercise over travel and over habitability is thus clearly shown. Florida and Southern California are also regions in which, owing to the mildness of their winter climates, certain fruits, such as oranges and lemons, which are not found elsewhere in the country, can be grown out of doors, and these are shipped to all parts of the United States.

Let us take another step in order to emphasize more clearly the distribution of temperature over the United States on the first day of our series. Draw a line which shall separate all places having a temperature _above_ 30 from those having temperatures _below_ 30, 30 being nearly the freezing point and, therefore, a critical temperature. Evidently this will help us to make our description of the temperature distribution more detailed. If this line is to separate places having temperatures _above_ 30 from those having temperatures _below_ 30, it must evidently pa.s.s through all places whose temperature is exactly 30. Examine the thermometer readings entered on your map to see whether there are any which indicate exactly 30. You will find this reading at Norfolk, Va., Wilmington, S. C., Atlanta, Ga., Chattanooga, Tenn., Ft. Smith, Ark., and Portland, Ore. Through all these stations the line of 30 must be drawn.

Begin the line on the Atlantic Coast at Norfolk, Va., and draw it wherever you find a thermometer reading of 30. It is best to trace the line faintly with pencil at first, so that any mistakes can be easily rectified, and it should be drawn in smooth curves, not in angles. From Norfolk the line must run southwest through Wilmington, and then westward through Atlanta, pa.s.sing just north of Augusta, which has 31. From Atlanta the line goes northwest through Chattanooga, and thence westward, curving south of Memphis (28) and Little Rock (26), and then northwestward again through Ft. Smith.

In fixing the _exact_ position of the 30 line south of Memphis and Little Rock, the following considerations must be our guide: Memphis has 28; Vicksburg has 35. Neither of these stations has 30. Suppose, however, that you had started from Memphis, with a thermometer, and had traveled very rapidly to Vicksburg. The thermometer reading at starting in Memphis would have been 28, and at the end of your journey in Vicksburg it would have been 35, presuming that no change in temperature at either station took place during the journey. Evidently the mercury rose during the journey, and in rising from 28 to 35 it must, somewhere on the way, have stood at exactly 30. Now this place, where the temperature was exactly 30, is the point through which our 30 line ought to pa.s.s. How are we to determine its location? a.s.sume, as is always done in such cases, that the temperature increased at a uniform rate between Memphis and Vicksburg. The total rise was from 28 to 35 = 7. In order to find a temperature 7 higher than at Memphis, you had to travel the whole distance from Memphis to Vicksburg. Suppose you had only wished to find a temperature 5 higher.

Then, a.s.suming a uniform rate of increase between the two stations, you would have had to travel only 5/7 of the distance, and your thermometer at that place would have read 28 + 5 = 33. But a.s.sume you had wanted to find the place where the thermometer stood at 30. In this case you would have been obliged to go but 2/7 of the total distance from Memphis to Vicksburg, and at that point your thermometer reading would have been 28 + 2 = 30, which is the point we wish to find. In this way, then, when we do not find the _exact_ temperature we are looking for on the map, we can calculate where that temperature prevails by noting places which have temperatures somewhat higher and somewhat lower, and proceeding as in the case just described. Take another example. Little Rock, Ark., has 26; Shreveport, La., has 40. 40 - 26 = 14, which is the total difference.

From 26 to 30 is 4. Therefore a point 4/14 or 2/7 of the distance from Little Rock to Shreveport should have a temperature of 26 + 4 = 30, which is the point we wish to find, and through which our 30 line must pa.s.s.

From Ft. Smith the line cannot go north or northwest or west, because the temperatures there are all below 30. To the south the temperatures are all above 30. Evidently there is only one direction in which you can prolong the line, and that is to the southwest. Temperatures of 30 cannot be found north of El Paso (28), because there the temperature distinctly falls, Santa Fe having 4, Denver, -14, and Cheyenne, -23. Therefore temperatures _above_ 28 must be found south of El Paso. From Ft. Smith you may, therefore, continue the 30 line southwest and west, pa.s.sing close to El Paso, but to the south of it. In determining the further course of the 30 line, note that Yuma and all the California stations have temperatures above 30, while Winnemucca, Nev., has 13, and Portland, Ore., has exactly 30. From El Paso you may, therefore, continue the line to the northwest, pa.s.sing up through Central California parallel with the coast line, and to the east of all the California stations and of Roseburg, Ore., and thence running through Portland, Ore., ending just west of Seattle, Wash. Notice that the 30 line should be nearer to Sacramento, Cal., with 36, than to Red Bluff with 44.

Thus you have drawn the line which pa.s.ses through all places that have a temperature of 30 on the map under discussion. This may be called _a line of equal temperature_. _Isotherm_, a compound of two Greek words meaning _equal temperature_, is the name given in meteorology to such lines as this. You have drawn the isotherm of 30. All parts of the United States north and east of this line are below 30, while all districts south and west of it are above 30. You see, therefore, how much easier the drawing of this one line has made the description of the temperature distribution over the United States.

Carry this process a step further by drawing the line which shall pa.s.s through all places with a temperature of 40. This line begins at Jacksonville, Fla. (40), and runs west, pa.s.sing between Montgomery, Ala.

(33), and Pensacola, Fla. (46). Thence it turns to the northwest, pa.s.sing between Vicksburg, Miss. (35), and New Orleans, La. (48), and through Shreveport, La. (40). From Shreveport it turns to the southwest, pa.s.sing to the north and west of Palestine, Tex. (46), and down through San Antonio, Tex. (40). Its further exact location cannot be determined in Mexico, because there are no observations from Mexican stations, but the readings at Yuma, Ariz. (41), and at San Diego (42), Los Angeles (44), San Francisco (45), Red Bluff (44), and Cape Mendocino (43), all in California, show that the 40 isotherm may be started again just north of Yuma, and may be carried up through California, nearly parallel with the Pacific Coast, ending between Cape Mendocino, Cal. (43), and Roseburg, Ore. (37). You have now drawn the isotherms of 30 and of 40, and in order to avoid confusion, mark the ends of the first line 30 and the ends of the second line 40.

Isotherms on weather maps are drawn for every even 10 of temperature.

They are drawn in smooth curves and not in angular sections. Two isotherms cannot cross one another, for if they did you would have two temperatures, differing by 10, at the point of crossing, which is obviously impossible.

Complete the chart for this day by drawing the remaining isotherms, _i.e._, those for 50, 20, 10, 0, -10, -20, and -30, bearing in mind what has been said in regard to the determination of the positions of isotherms when the _exact_ temperature you are seeking is not given on the map.

The dotted lines in Fig. 18 show the positions of the isotherms when drawn. Notice how clearly the temperature distribution now stands out, and how simple the description of that distribution has become. Observe that the isotherms, although more or less irregular, show a good deal of uniformity in their general courses, and this uniformity is a great a.s.sistance in drawing them. Study the distribution of temperature on this map, and the positions of the isotherms, very carefully.

Construct isothermal charts for the remaining days of the series. Use a new blank map for each day, and take the temperature observations from the table in Chapter VIII. Proceed as in the case of the first day. Draw the isotherms for every even 10 of temperature, taking care to study the course of each line before you begin to draw the line. The charts when completed form a series in which the temperature distribution over the United States is shown at successive intervals of 24 hours.

[Ill.u.s.tration: FIG. 18.--Isotherms. First day.]

In order to bring out the temperature distribution on the maps more clearly, color (with colored pencils or water colors) all that portion of each map which lies within the -20 isotherm a dark blue; that portion which is between the 0 isotherm and the -20 isotherm a somewhat lighter shade of blue, and those districts which are between 0 and +30 a still lighter blue. The portion of the map above 30 and below 40 may be left uncolored, while the districts having temperatures over 40 may be colored red. In the map for the third day the district which has temperatures below -50 should be colored darker blue than any shade used on the other maps, or black, in order to emphasize the extremely low temperatures there found. Figs. 19-24, on which the isotherms are shown, also ill.u.s.trate the appearance of these maps when the different temperature areas are colored, as has been suggested.

[Ill.u.s.tration: FIG. 19.--Temperature. First Day.]

[Ill.u.s.tration: FIG. 20.--Temperature. Second Day.]

[Ill.u.s.tration: FIG. 21.--Temperature. Third Day.]

[Ill.u.s.tration: FIG. 22.--Temperature. Fourth Day.]

[Ill.u.s.tration: FIG. 23.--Temperature. Fifth Day.]

[Ill.u.s.tration: FIG. 24.--Temperature. Sixth Day.]

Study the maps individually at first. Describe the temperature distribution on each map. Ask yourself the following questions in each case: Where is it coldest? Where warmest? What is the lowest temperature on the map? What is the highest? At what stations were these readings made?

Then compare the successive maps and answer these questions: What changes have taken place in the intervening 24 hours? In what districts has the temperature risen? What is the greatest rise that has occurred? Where? In what districts has the temperature fallen? What was the greatest fall in temperature and where did it occur? Has the temperature remained nearly stationary in any districts? In which? You will find it a help in answering such questions to make out a table of all the stations, and to indicate in columns, after the names of the stations, the number of degrees of rise or fall in temperature at each place during the 24-hour interval between the successive maps. When the temperature is higher at any station than it was on the preceding day, note this by writing a plus sign (+) before the number of degrees of rise in temperature. When the temperature has fallen, put a minus sign (-) before the number of degrees of fall. Thus, New Orleans, La., had a temperature of 48 on the first day. On the second it had 33. Therefore the change at New Orleans was -15 in the 24 hours. At Key West, Fla., the change was +11 in the same time.

Write a brief account of the temperature distribution on each day of the series, and of the changes which took place between that day and the one preceding, naming the districts and States over which the most marked falls and rises in temperature occurred, with some indication of the amount of these changes. Note especially the changes in position, and the extent, of the districts with temperatures below -20; between 0 and -20, and between 30 and 0. Write out a clear, concise statement of the temperature distribution and changes shown on the whole set of six maps.

=Cold Waves.=--The series of charts for these six days furnishes an excellent ill.u.s.tration of a severe cold wave.