164 B. V. KOTLIAREVSKII 



Much less frequently there are cases of linear or near -linear fields, having 

 points of inflection but no extreme points. For these fields, gQ and Zq cannot 

 be determined. In these cases, keeping to the framework of the proposed 

 method, only negative conclusions can be drawn from the results. Let us 

 assume that the parameters of the survey under consideration, a — 0.6 mgal, 

 a = 6 km and p — 2 mgal from the nomograms (Figs. 11 and 13) we find 

 that such a survey makes it possible to determine anomalies characterized 

 by the values g^ — 11.8 mgal and Zp = 8.5 km with an accuracy of E^= 15% 

 and D^ =30%. However, there are no such anomalies on the map, conse- 

 quently they do not exist in actual fact. Given other indices of accuracy, 

 we find the new values of gQ and Zq, etc. With regard to the actual accuracy 

 of this survey, another method of determination should be used. This 

 method is based on the use of the first and second order differences A^q 

 and ZI2 as gravity field characteristics. 



SUMMARY 



In planning the work, a considerable problem is the selection of the 

 values of E^^ and D^ since they determine the parameters of the survey. 

 These values should obviously be different for detailed and regional or 

 exploratory surveys. It is difficult, however, to give actual figures for any 

 of these forms of survey. After studying certain production data, the author 

 came to a preliminary conclusion that for regional and exploratory surveys, 

 the values E^ = 15% and D^ = 25+30% give quite reliable results. 

 However, with respect to the regional and exploratory surveys, these require- 

 ments are too high. Let us consider in particular, the data of the experi- 

 mental example. The survey parameters in this example are exceptionally 

 unfavourable. As a result, the relative errors are very large, being E^ = 50 % 

 and D^ = 69%. Despite this, of the 50 extreme points of the true field, only 5 

 points were not noticed in the observed field. All the remaining extreme 

 points, although distorted in their shape and amplitude, are reliably diffe- 

 rentiated on the observed curve. Consequently, if the survey is only requi- 

 red to give localization of anomalies, at the planning stage the values of E^ 

 and Dj^ can be considerably increased. To solve this problem finally, it is 

 necessary to analyse a large amount of the actual material, as was done by 

 the author. 



At the start of the article, a mention was made of the method of 

 evaluating the accuracy of the survey according to the value of the mean 

 square error in the force of gravity at the observation points, and on the 

 method for determining the cross -sections of the isoanomalies according 



