In phytoplankton primary production studies, samples for the determination of photosynthesis versus irradiance relationship (P–I) are usually incubated at several irradiance levels during a fixed time period, commonly 2–4 h. However, it is not clear if the use of this fixed incubation time is appropriate to study the P–I relationship in any given ecosystem. The aim of this work was to study the influence of incubation time on the P–I relationship in natural phytoplankton populations from three different coastal ecosystems: an open coastal area, an estuary, and a coastal lagoon. Physical and chemical variables, phytoplankton biomass, species composition, and P–I curves were analysed. The results showed that, in the coastal area and in the estuary, P–I relationships were time dependent, whereas in the coastal lagoon different incubation periods produced the same P–I curve. An underestimation of daily primary production, ranging from 13% to 42.5%, was calculated when data from standard incubation times (2–4 h) were used in ecosystems where P–I curves present a dynamic time-dependent behaviour. This work suggests that the response of the P–I curves to the incubation time varies with the characteristics of the ecosystem and is related to the light regime to which phytoplankton cells are adapted.