In Ukraine, due to financial difficulties, the planned census is often postponed from year to year. The country is forced to rely on static data, which is sometimes completely untrue. In addition, it is not possible to count the number of inhabitants everywhere. In particular, Ukraine has not known for several years exactly how many people live in the occupied territories. As a result -the wrong distribution of the budget, which entails another financial crisis and a number of other troubles. At the same time there several satellite based products allowing to estimate the population. This study provide validation of satellite based population products delivered by JRC and NASA for the territory of Ukraine. To verify the correctness of the satellite based products, such as Global Human Settlement Layer (GHSL) and NASA population GPWv4 data collection have been compared to official statistics on the number of the largest cities of Ukraine.

No abstract

1 st July 2021 the law on the creation of land market start effect in Ukraine. As a result, land appraisal became cornerstone task in Ukrainian agriculture sector. The official methodology on land appraisal includes use of soil fertility characteristics combined with coefficients related to the distance to the infrastructure objects or settlements and placing of field in specific functional areas, like recreational, or areas with high level of radiation pollution. In this study we collected open source infostructure geospatial information and characteristics of fields obtained from remote sensing data -crop types and Normalized Difference Vegetation Index to build land price predictive model trained on the official land market information. This work designed to investigate potential of geo-informational technologies and remote sensing in the land appraisal use. We separated all available ground truth land price data into three groups by fields size -very small, small, medium and big. We found different relationships between field characteristics and prices. For very small fields the most important features are area, altitude, slope, bonitet and distances to elevators, villages and roads. For small fields the most important are bonitet, altitude, area and distances to cities and roads. For medium and big field's area, slope, distance to cities, roads and historical NDVI.

The annual harvest growth is very important for the development of the agricultural sector of each country. To ensure its growth, there is a need for modern monitoring of agricultural land indicators. These indicators are usually obtained directly from local agronomists, but this method of collecting information is too long and complicated. As a state-ofthe-art alternative is the use of remote sensing data from satellites. Using remote sensing technologies, it is possible to predict the yield of major crops in the studied areas by classical methods of machine learning, in particular, regression analysis. In this paper, a regression model for yield forecasting for the territory of Kyiv region (Ukraine) will be create based on vegetation indices obtained from satellites data. The main purpose is to assess the reliability of the obtained model. The paper will refute or confirm the feasibility and effectiveness of the satellite data using in regression models for crop yield forecast for Kiev region.