The purpose of the research is to analyse and assess existing approaches in investigation of interconnections between climate and underlying surface. Land use/land cover (LULC) influences climate formation via physical and chemical properties (albedo, roughness, height, chemical composition etc.). Climate in its turn affects land cover by means of meteorological parameters (air temperature and humidity, precipitation, wind etc.) and causes both cyclic and irreversible changes in land cover. In addition, anthropogenic factors exacerbate surface-climate interactions through? for example, LULC change that usually causes an additional release of chemical compounds. The paper distinguishes three main directions of the “climate - LULC” interactions research that is conducted mainly with application of satellite monitoring products, observation dataset, geographic information systems (GIS) and numerical modelling. The first direction implies monitoring and research of cyclic changes and transformation of LULC influenced by natural and anthropogenic factors, using different GIS-based satellite and surface meteorological observation databases. Despite significant technical progress and great amount of studies conducted for detecting dynamics of LULC change for different time intervals, the problems of dealing with cloudiness and shadows from orographic and other objects still remain. The second direction investigates the influence of LULC change on the chemical composition in the atmospheric boundary layer and on the regional climate. Numerous researches were dedicated to the influence of different kinds of surface such as forests, grasslands, croplands, urban areas etc. on climate characteristics and also on fluxes, for example, CO2. The effect of midlatitude forests on climate remains to be one of the challenging and urgent issues. The third direction relates to LULC change modelling and regional climate modelling. For the last decade a spatial resolution of models was considerably increased and, as a result, representation of interaction between atmosphere and land improved. Online integrated numerical atmospheric models are found as the most promising ones. They include "meteorological parameters – atmospheric chemical composition" feedbacks and can consider LULC on global and regional scales. However, some issues still need improvement, namely radiative transfer, cloud microphysics, cloud-aerosol-precipitation interactions, as well as parametrizations of some types of land and their interaction with the atmospheric boundary layer.