More diverse ecosystems on land and in the ocean are thought to be more productive, stable, and resistant to change, but these relationships are highly variable across systems and scales. Although the productivity-diversity relationship (PDR) has been extensively explored on land, there are limited in-water observations of the PDR in marine ecosystems. In this work, the relationship between phytoplankton diversity and carbon biomass (as a proxy for productivity) was examined using a global in situ dataset. The shape of this relationship was evaluated for three metrics of phytoplankton diversity: pigment concentrations modeled from hyperspectral remote sensing reflectance (Rrs({lambda})), pigment concentrations measured by high performance liquid chromatography (HPLC), and 18S rRNA gene sequences. While gene sequencing methods provide higher resolution taxonomic information about phytoplankton communities, remote sensing methods collect higher resolution spatiotemporal information on global scales. By comparing these methods (Rrs({lambda})-modeled pigments vs. measured HPLC pigments vs. 18S rRNA gene sequences), this work demonstrates the variability in the relationship between phytoplankton diversity and carbon biomass based on the method of assessing these parameters, and establishes a baseline for in situ observations that has the potential to be extended to global observations from NASA's Plankton Aerosol Cloud ocean Ecosystem (PACE) satellite.