The metrological support for monitoring of ground layer of the atmosphere with ASKRO gamma spectrometers requires determining their sensitivity in the measurement geometry of technogenic radionuclides. The application of numerical modeling, the method of similarity of radiation fields, and a combined method in evaluating sensitivity of a spectrometer is examined. It is shown that the model method overstates the sensitivity and does not permit checking the spectrometer. To decrease the error of estimation, a combination of measurements of the efficiency of the spectrometer for a point source and a Monte Carlo calculation of the incident flux is used. An estimate of the total relative error of determination of the sensitivity is given and it is shown that it can be verified. The estimates obtained for the sensitivity increase the minimum measured volume activity of the spectrometer at energies less than 1 MeV.Metrological support for measurements of the volume activity of technogenic radionuclides in the atmospheric layer at the ground by means of the spectrometric channel ASKRO includes certification of the gamma spectrometer and methods for checking its main measuring characteristics, development of methods for performing measurements and running monitoring of the data, collection of control sources and standard measure of activity. The sensitivity and limit of admissibility of the relative main error are, according to GOST 21496-89 [1], the main metrological characteristics of the means of measurement of the volume activity of radionuclides in a gas. The need to determine them in the geometry of a semi-infinite source (referred to below as a 2π source) with respect to all technogenic radionuclides raises difficulties in the certification of the spectrometer and verifying it with respect these parameters under experimental conditions. The information of the spectrometric channel ASKRO serves for establishing a correspondence between the volume activity of the ith technogenic radionuclide q i and control levels (control correspondence), its variation in time (monitoring), the probability of evaluating the presence in the ground layer of the atmosphere (discovery). The δq i error in evaluating the concentration is related with, first and foremost, the accuracy δS ij with which the sensitivity of the spectrometer with respect to the jth line of the ith radionuclide is evaluated in the geometry used for the measurements:where ΔX ij is the excess count rate above the background in the region of the total absorption peak of the jth characteristic line of the radionuclide [2].If the concentration q i of a radionuclide is less than the minimum measured value, the relative estimate of the excess of the count in the jth energy range N ij above the a posteriori average count N ij , calculated on the stationariness interval, (N ij -N ij )/N ij 1/2 is used to calculate the probability of its presence in the ground layer of the atmosphere [3]. The sensitivity of the spectrometer with respect to the ith radionuclide in explicit...