Design of digital zenith camera in Institute of Geodesy and Geoinformatics of the University of Latvia started in 2009. Now, after tests of a prototype, improvements of design and manufacturing of the second version, it has reached operational status. The paper describes construction of camera, features of it‘s control software. Vertical deflection determination results are discussed.

Since the development of GNSS techniques, the determination of a precise quasi-geoid model has become even more actual. In terms of this project the staff of the Institute of Geodesy and Geoinformatics (GGI) has developed a new quasi-geoid model based on DFHRS (Digital Finite-element Height Reference Surface) approach additionally using astrogeodetic measurements – vertical deflections (VD), which can be observed by a Digital zenith camera. This paper evaluates a quasi-geoid model results based on vertical deflections, as a study area using the territory of Latvia: the standard deviation of the solution is equal to 0.006 m with observation residuals after the adjustment of minimum and maximum differences −0.012 and 0.012 accordingly. The standard deviation of quasi-geoid heights and h-H values from LGIA database is equal to 0.012 m with minimum and maximum differences −0.026 and 0.025 accordingly. The post-processed terrestrial VD observations have been compared to VD derivatives from EGM2008 and GGMplus geopotential models. The developed quasi-geoid has been compared to the national quasi-geoid model LV’14 and to the Scandinavian NKG2015.

According to the decision of IAG Reference Frame Sub-commission for Europe (EUREF) the EVRF2007 solution as the vertical reference has to be deployed in EU countries.The new height system LAS-2000,5 had been enacted as the European Vertical Reference System‘s EVRF2007 realization in Latvia and the new geoid model LV‘14 had been introduced by Latvian authority Latvian Geospatial Information Agency. However, the appreciation of the quality of quasi-geoid model LV‘14 is rather contradictious among the users in Latvia. The independent estimate and comparison of the two Latvian geoid models developed till now has been performed by the Institute of Geodesy and Geoinformatics. Previous geoid model LV98 which was developed for Baltic-1977 height system almost 20 years ago is outdated now. Preparatory actions described in order to fulfil the task of comparison the geoids in two different height systems. The equations and transformation parameters are presented in this article for the normal height conversion from Baltic-1977 height system to the Latvian realization named LAS-2000,5. The comparison is performed of both Latvian quasigeoid models – the new one LV‘14 and previous LV98. The quality of both models estimated by controlling the geoid heights at the properly densified GNSS/levelling network sites. The distribution of discrepancies in comparison with normal distribution N(x,μ,s) is depicted in corresponding figures. For LV‘14 quasi-geoid model the standard deviation of discrepancies is 3.2 cm, 75% of discrepancies x ≤ 3.2 cm. For LV98 quasigeoid model the standard deviation of discrepancies is 4.7 cm, 80% of discrepancies x ≤ 6 cm. Without doubt, the newly developed LV‘14 quasi-geoid model is of higher quality.

Since finalizing of design in 2016, the digital zenith camera of the University of Latvia was involved in a number of test observations as well as field observations at about 70 different sites. The paper presents analysis of observation results, estimation of instrument’s performance and accuracy. Random and systematic error sources are outlined. Impact of anomalous refraction on vertical deflection determination is discussed. Results of adaptation of GAIA reference star catalog for astrometric data reduction are reported.

No abstract