Quantitative measurement of the number of active surface sites on two-dimensional (2D) catalysts is one of the most crucial points in heterogeneous catalysis because it is used to determine the turnover frequency (TOF), which refers to the catalytic activity of model catalysts. However, because of the difficulty in identifying the effective active surface area on 2D heterogeneous catalysts, there is still the assumption that each metal atom is an active site. To shed light on these issues and to bridge the activity gaps between 2D and three-dimensional (3D) heterogeneous catalysts, we present an experimental approach that uses Pt nanoparticle (NP) arrays on a thin silicon wafer probed with CO pulse chemisorption, a widely used surface-sensitive technique, to determine the number of active sites and the area of the effective active surface. A Pt thin film and Pt NP arrays with two different NP sizes (i.e., 2.1 and 4.5 nm) were prepared as model systems for 2D catalysts. The effective active metal surface area determined using CO pulse chemisorption for these 2D catalysts is 53-79% of the apparent metal surface area that was obtained by measuring the surface area based on scanning electron microscopy images. This discrepancy between the active and apparent surface area is attributed to the presence of hydrocarbon contamination and organic capping layers on the catalysts. The results indicate that estimating the active sites of 2D catalysts by apparent surface area is reasonably in agreement with the number measured by chemisorption that is used to characterize 3D nanocatalysts. This experimental technique on 2D catalysts can be expected to provide information for extracting the true TOF of product molecules on 2D catalysts in gas-phase catalytic reactions.