The goal of this study is to identify the current state of in situ observations and remote sensing data and methods used to assess biomass and net primary production (NPP) in West Siberian natural ecosystems, and consider perspectives for future developments. The natural ecosystems of the boreal region mainly consist of two classes: wetlands and forests, where one is very different from the other, requiring different methods for biomass assessment. Basically, two methods are available to estimate NPP and biomass in regional terrestrial ecosystems: (1) extrapolating the local field measurements up to a larger region, using the vegetation or land cover maps and (2) modeling productivity and plant biomass at regional and grid point scales, with or without the use of remote sensing data and techniques. The first method was predominantly used to estimate wetland biomass, having an extensive dataset of direct in situ measurements in both the above- and below-ground fractions of biomass. So far, no direct methods based on remote sensing data have been elaborated for biomass estimations in wetland ecosystems and soil carbon inventories. In forest ecosystems, the biomass can be estimated by processing satellite data from high-resolution radiometers (AVHRRs). The radar or LIDAR remote sensing approaches hold great promise for direct observations of the three-dimensional structure (3D) of the above-ground vegetation that can be used for relatively straight-forward calculations of carbon storage, but the method works only in low to medium biomass ecosystems. The SAR-based biomass retrievals were found to be fairly uncertain in mature forests with high biomass values, as the Synthetic Aperture Radar (SAR) signal often saturates at ~70 tonnes/ha. The estimation errors in terms of RMSE are typically found at 25–30 % of the mean biomass. The methods should be further refined to reduce uncertainties and to make them operational over the vast region of Siberia.