METEOROLOGY. 363 



METEOROLOGY. 



Bjerknes, V., University of Leipzig, Leipzig, Germany. Preparation of a 

 work on the application of the methods of hydrodynamics and thermody- 

 namics to practical meteorology and hydrography. (For previous reports 

 see Year Books Nos. 5-14.) 



Dr. Th. Hesselberg left the service of the Carnegie Institution of 

 Washington at the end of the year 1915, having been appointed 

 Director of the Norwegian Meteorological Institute. He has been 

 succeeded by Mr. Holtsmark. 



Previous reports have referred to what we have called frictional 

 resistance to atmospheric motions. The observations having given 

 the change of motion during a certain time as well as some of the 

 forces producing this change (pressure-gradient, deviating force of 

 the earth's rotation), the problem in its most general form was this: 

 to find the unknown force which, added to the knov»'n forces, produces 

 the knov/n change of motion. This unknown force is of a complex 

 nature, but is, for the sake of brevity, referred to simply as friction. 

 It has turned out to be possible to obtain fairly constant average 

 values of this force (c/. previous reports). These average values 

 being found, we have tried to treat the reverse problem: to determine 

 the change of motion during a short interval of time (3 hours), all 

 forces, including friction, being considered as known. As for this 

 problem we should want the true local and momentary values of the 

 friction, but we know only certain average values, we do not claim 

 that the solution must necessarily succeed; but it is desirable to try 

 it as the most direct way of obtaining further insight into the nature 

 of ''atmospheric friction." 



A great difficulty, however, in this case is the incompleteness of 

 meteorological data. We have begun Vvdth observations from the United 

 States of America, kindly copied for us by the U. S. ¥/eather Bureau. 

 They give the meteorological elements at all stations for every hour 

 during the following periods: 1905, January 1, 2, 3; 1905, November 

 27, 28, 29; 1911, November 10, 11, 12. 



In order to avoid complications due to the topography of the land, 

 we chose for our investigations the region Umited on the west by the 

 Rocky JNIountains, on the east by the Appalachian Mountains, on 

 the north by Canada, and on the south by the Gulf of Mexico. 



The fu-st attempts led to apparently absurd results, showing that 

 the local and momentary values of the friction could differ enormously 

 from the average values. More detailed determinations of the fric- 

 tion for the defined region were therefore sought, and a chart of 

 friction was drawn, derived only from winter observations. In order 

 to realize as far as possible the conditions corresponding to average 

 friction, and in order to reduce at the same time the effect of the errors 

 of observation, the observed winds and the observed gradients were 

 averaged, on the one hand for periods of 3 hours, on the other hand 

 for areas of 250 square kilometers. We obtained in this way charts 



