124 



METEOROLOGY— FORECASTING 



as representative as possil:)le, i.e., luiiiiflueneeil by 

 purely local effects. Measurements are : 



1. Surface air temperature (of the unmodified air 

 over land, in the case of air moving off a land mass) ; 



2. Surface air humidity (also of the unmodified 

 air which can be expressed in terms of relative humid- 

 ity, specific humidity, dew point, wet bulb tempera- 

 ture, or vapor pressure) ; 



3. Sea surface temperature ; 



4. Wind speed and direction (preferably at 1,000 ft- 

 elevation) ; and sometimes 



5. Distance from land (of primary importance only 

 in the case of stal)ility conditions, when the air is 

 warmer than the sea surface). 



All these data may, of course, lie profitably supple- 

 mented by aerological soundings, weather maps, and 

 any other pertinent information available. 



Modification of Air by :^ca Surface. As a qualitative 

 description of the modification that the air undergoes 

 in moving over water^ three cases may be distinguished, 

 namely : 



1. Neutral equilihriuni (resulting when the initial 

 surface air temperature is the same as the sea temper- 

 ature). The temperature structure of the air remains 

 unchanged ; however, since the air is usually not com- 

 pletely saturated, moisture is supplied to the lower 

 layers by evaporation from the sea surface, in this way 

 causing a greater decrease of humidity with height, 

 which tends to establish an M distribution sucli that 

 the modified refractive index is either constant or 

 decreasing with height. In the case in which the air 

 is initially completely saturated im iiKKlilicatiini takes 

 place. 



2. Unstable eqaiUhrititii (resulting when the initial 

 surface air temperature is less than the sea surface 

 temperature). In this case the moisture content of the 

 air is always less than that corresponding to satura- 

 tion at the sea surface temperature, so tlmt the lower 

 layers of air suffer an increase in humidity as well as 

 in temperature. Owing to the greater sensitivity of 

 3/ to humidity than to teni]ierature this tends to bring 

 about a decrease of M with height in a layer of air 

 adjacent to tlie sea surface, while the unstable condi- 

 tions give rise to vertical mixing which keeps the 3/ 

 distribution close to standard above this layer so that 

 the duct is confined to lower levels than in the case of 

 neutral equilibrium. 



3. Stable equilibrium (resulting when the initial 

 surface air temperature is greater than the sea sur- 

 face temperature). If, in addition, the air is initially 



quite dry, i.e., has a moisture content less than that 

 corresi^onding to saturation at the sea surface tempera- 

 ture, then the resulting rapid decrease of moisture with 

 height plus the stable temperature distribution leads 

 to a tendency to surface duct formation. On the other 

 hand, if the initial moisture content of the air is rela- 

 tively large, moisture may be condensed out of the 

 surface layers of air thus tending to give rise to an 

 increase in humidity with elevation which when suffi- 

 ciently marked may counteract the effect of the stable 

 temperature distribution and so prevent the forma- 

 tion of a duct or even produce substandard conditions. 

 In either case, the stable structure of the air tends to 

 hinder vertical mixing so that modification from the 

 surface upward proceeds slowly and hence is highly 

 dependent on the distance traveled by the air over 

 the water. 



Necessary Calculations. To determine quantitatively 

 the possibilities of duct formation, the following items 

 can be readily calculated from any particular observed 

 set of initial conditions. 



1. Temperature e.vcess, which is merely the repre- 

 sentative surface air temjDerature (before modifica- 

 tion) minus the sea surface temperature. 



2. M deficit, defined as the value of M correspond- 

 ing to the sea surface temperature minus the value 

 of M determined from the representative surface air 

 temperature and humidity (before modification) ; 

 values of M can be ascertained from nomograms,'' 

 tables,' or directly from the formula.^ In the case of 

 M corresponding to the sea temperature, a 98 per cent 

 saturation is assumed; the 3 per cent vapor pressure 

 correction is subtracted to take into account the salin- 

 ity of the sea water. 



3. Batio of duct ividtli to M deficit, determined 

 from the chart in Figure 19 for a given temperature 

 excess and wind speed as measured at the 1,000-ft 

 level. 



4. Duct wicltti, found by multiplying the above 

 ratio (3) by the M deficit. 



Applicability of the Method. Since the procedure 

 is designed to take into account only the surface modi- 

 fication of air over water, its application is restricted 

 to the prediction of the first four types of M curves 

 described in Section 8.3.2 : standard, substandard, 

 simple surface trapping, and transitional. The causes 



''Nomograms are included and explained in the Appendix. 

 'Tables for computing M can be found in references S and 9. 

 'The formula expressing M in terms of pressure, temper- 

 ature, humidity, and elevation is given in the Appendix. 



