There have been criticisms in years past that the climatological conditions of the United States have not received that care and attention which their importance demanded. Much has been done to remedy defects in this respect, although, as is well known here in Washington, the general law which forbids the printing of any works without the direct authority of Congress, has been an obvious bar to great activity on the part of the Signal Office. Within the year the rainfall conditions of twelve Western States and Territories have been published with elaborate tables of data and fifteen large charts, which set forth in considerable detail the rainfall conditions for that section of the country. In addition the climatic characteristics of Oregon and Washington have been graphically represented; and rainfall maps,—unfortunately on a small scale,—have been prepared, showing for each month, the average precipitation of the entire United States, as determined from observations covering periods varying from fifteen to eighteen years.
In Missouri, Professor Nipher has prepared normal rainfall charts for that State, unfortunately on rather a small scale. In New York, Professor Fuertes, and in Michigan, Sergeant Conger, of the Signal Service, have commenced maps showing, by months, the normal temperatures of their respective States on maps of fairly open scale. Work of a similar character has been carried on in Pennsylvania under the supervision of Professor Blodget, well known from his climatological work. In other directions and in other ways, work of a similar character is in progress.
Without doubt too much is anticipated from pending or projected irrigation enterprises in the very arid regions of the West. These unwarranted expectations must in part result from a failure on the part of the investors to consider the general question of these enterprises, in its varied aspects, with that scientific exactness so essential in dealing theoretically with extended subjects of such great importance.
Everyone admits the correctness of the statement that the amount of water which flows through drainage channels to the sea, cannot exceed the amount which has evaporated from adjacent oceans and fallen as precipitation on the land. Further it is not to be denied that the quantity of water available in any way for irrigation must be only a very moderate percentage of the total rainfall which occurs at elevations above, and perhaps it may be stated considerably above, that of the land to be benefited.
Elsewhere it might be appropriate to dwell in detail upon the importance of cultivated land in serving as a reservoir which parts slowly with the water fallen upon or diverted to it, and in avoiding the quick and wasteful drainage which obtains in sections devoid of extensive vegetation or cultivation; and also that water thus taken up by cultivated lands must later evaporate and may again fall as rain on other land. But the pertinence of meteorological investigations in connection with irrigation and this annual address, relates much more directly to important questions of the manner by, and extent to which, precipitation over the catchment basins of the great central valleys fails to return in direct and visible form, through the water courses, to the Gulf of Mexico.
The inter-relation of rainfall and river outflows is one of peculiar interest, in connection with the important matter of irrigation now under consideration in this country.
Probably more attention has been paid to this subject in the valley of the Seine, by Belgrand and Chateaublanc, than in any other portion of the globe. One of the curious outcomes of Chateaublanc's observations, is one bearing on the maximum value of the floods in the Seine for the cold season, from October to May, by which he says that the reading of the river gauge at Port Royal is equal to 12.7 minus the number of decimetres of rainfall which has fallen on an average throughout the catchment basin during the preceding year. This curiously shows that the intensity of the winter floods of the Seine is inversely proportional to the quantity of rain of the preceding year.
Sometime since, John Murray, Esq., in the Scottish Geographic Magazine, treated generally the question of rainfall and river outflows. The annual rainfall of the globe was estimated to be 29,350 cubic miles, of which 2,343, falling on inland drainage areas, such as the Sahara desert, etc., evaporate. The total annual discharge of rivers was estimated at 7,270 cubic miles. In the case of European drainage areas between a third and a fourth of the rainfall reaches the sea through the rivers. The Nile delivers only one thirty-seventh of the rainfall of its catchment basin, while tropical rivers in general deliver one-fifth.
The Saale river of Germany, from late data based on 45 rainfall stations in its catchment basin, during the years 1883 to 1886, discharged 30 per cent. of its rainfall.
During the past year Professor Russell, of the Signal Office, has determined carefully the rainfall and river outflow over the most important part of the United States, the entire catchment basin of the Mississippi river and its tributaries. This work was done as preliminary to formulating rules for forecasting the stage of the water several days in advance on the more important of the western rivers in the United States. The river outflows at various places on the Mississippi and Missouri and Ohio rivers, were tabulated from data given in the reports of the Mississippi and Missouri River Commissions. The tables were largely derived from the results of the measurement of current velocities. As gauge readings were taken at the time of discharge or outflow measurements, the discharges or outflows can be told approximately at other times when only the river gauge readings are known. The results for the outflow of rivers derived from measurements made under the supervision of these commissions, are of a high order of accuracy, and it is not probable that the results deduced from the gauge readings are much in error. Of 1881 and 1882, during which years measurements were made, 1881 was a year of great flood in the Missouri river, while the Mississippi river was not flooded. The year 1882, on the other hand, was marked by a great flood in the lower Mississippi river, with a stage in the Missouri much above the average. The rainfall in the six great valleys of the Mississippi, during the entire years 1881 and 1882, was charted from all observations available, and its amount in cubic miles of water calculated with the aid of a planimeter.