Exposure.—We now consider what the daily sunshine record for a particular station means. An ideal exposure has an uninter rupted view of those parts of the horizon in which the sun rises or sets; and elsewhere the view of the sun must not be obstfucted by the ground, buildings, trees or any other obstacle ; but ideal exposures are not always to be obtained. In mountainous districts particularly it may be impossible to find a site in which the sun is not obstructed for an appreciable part of the day. In these cir cumstances it becomes a question whether the amount of sunshine recorded should be referred to the maximum possible for an un interrupted horizon or the maximum possible for the particular exposure. The answer to the question really depends upon the purpose for which the information is wanted. As a climatological factor of the locality the shadow cast by the surrounding hills is of importance, it is part of the difference between the fertility of the southern and northern slopes of hill country. This importance is, of course, in many respects exclusively local, and indeed the possible duration of sunshine at any station is a local characteristic which it is desirable to know. Consequently as evidence of the peculiarity of the site the recorded sunshine might be referred to the total possible with a free horizon. On the other hand, taking the record of sunshine as an indication of the clearness of the sky for the purposes of general meteorology, the screening of the sun by hills must be regarded simply as limiting the time during which observation is possible and the duration of the sunshine recorded should be referred to the possible duration at the particular site.
Sunshine Records for the British Isles.—The interest in the use of sunshine recorders is more widely extended in the British Isles than elsewhere, and it is, so far as the public are concerned, the most important meteorological element, but it is singular that up to the present a knowledge of the total amount of sunshine recorded during the day, the week, the month or the year is all that is apparently required.
In figure 3 the mean daily duration of sunshine for the month of July 1928 is represented by a number of isohels or lines of equal sunshine, the figures being given in hours. Figure 4 gives similar data for October 1928. Further, figure 5 represents the average weekly distribution of sunshine in different sections of the British Isles according to the average of twenty-five years. (X.) tions of New Mexico and Arizona, and portions of Nevada and the interior of southern California, have averages in excess of 8o%, Yuma, Arizona, having the maximum amount of recorded sunshine in the country, 88% of the possible.
Small areas of the northern portions from the Upper Lakes westward to Washington, much of the lower Ohio Valley and nearby portions of the Great Lakes, and small portions of the North Atlantic States have annual percentages ranging between 5o and 6o%, while only limited portions of the far North-west, the districts around Lake Superior, and the area from the upper Ohio Valley north-eastward to northern New England have aver ages less than 5o% of the possible.
In general during the colder months the percentages of the possible sunshine are less than during the warm season, but even then a large part of the central and southern areas west of the Mississippi River have amounts in excess of 6o% of the possible, while the warmer months mainly have decidedly high percentages, The United States.—Continuous records of sunshine in the
United States are kept at nearly two hundred different points well distributed over the country, the records being obtained by means of the Marvin electrical sunshine recorders, the character of which is described on page 569. Sunshine data may be expressed either in the actual hours of sunlight at each individual station, or in percentage of the hours of actual sunshine as compared with that possible. Both systems have their advantages, but in the contrast ing of different climates it appears that data on the percentages of possible sunshine actually received afford the better comparison of different regions, since the possible hours vary greatly for dif ferent seasons and localities, and their values must be made known before comparisons are feasible.
Considering first the annual sunshine, it becomes apparent that the United States, as a whole, is a country of abundant sunshine, nearly two-thirds of the entire area receiving annually on the average more than 6o% of the possible amount, a proportion ex ceeded by only a few of the localities commonly referred to as having sunny climates.
Over a large part of the south-west, extending westward from central Texas, and including the western portions of Oklahoma and Kansas, nearly to the Pacific Coast, the annual amounts average more than 7o% of the possible, and localities in the far south-west, including extreme western Texas, the southern por the summer months particularly having values ranging from 6o to 8o% in nearly all parts, and even exceeding 9o% of the possible in portions of interior California and nearby areas of Nevada.
The accompanying charts (Figs. 6 and 7) show the percentages of the possible sunshine in all parts of the country for January and July, and indicate the approximate sunshine conditions for winter and summer respectively. (C. F. M.) Sunshine Results for Other Parts of the World.—Maps showing the average annual distribution of sunshine over Europe and North America are given in Bartholomew's Physical Atlas, vol. iii. Atlas of Meteorology. Over Europe the largest totals, over 2,75o hours per annum, are shown over central Spain. In North America, values exceed 3,25o hours per annum in the New Mexico region. For other parts of the world the information available is not sufficiently extensive for the construction of charts. • Effect upon Sunshine Records of the Smoke of Great Cities.—Much discussion has taken place from time to time as to whether the climate of a locality can be altered by artificial means. One question of the kind to which the sunshine recorder gives an incontrovertible answer is as to the effect of the smoke of great cities in diminishing the sunshine in the immediate neigh bourhood. This is not a question which comes out merely by tak ing averages. The answer can be seen directly by comparing the daily cards at the different localities. Thus it appears that the di rect effect of the local contamination of the London atmosphere results in the diminution of the recorded sunshine for the whole year by 37%, and it is clear that the contamination extends in some degree as far as Kew, where the loss amounts to about I 0%. There is evidence of various kinds to show that the effect of the smoke cloud of cities can be traced sometimes for great distances.