Aim: Understanding the source water utilization of rice-based cropping systems helps develop improving water management strategies for paddy management. We investigated the effects of altered flooding regimes and crop diversification on plant root water uptake on a fully-replicated field trial at the International Rice Research Institute in the Philippines. Methods: All potential water pools, e.g., plant and soil extracted water, were analyzed for their water stable isotopic compositions (δ²H and δ¹⁸O). We determined the relative contributions from different water sources to root water uptake (RWU) of rice plants by applying a multi-source mixing model (Stable Isotopes Analysis in R, SIAR). The sensitivity of the model to the incorporation of prior information based on in-situ measurements of soil water content and root length density was investigated as well. Results: The modeling results showed that wet rice plants mainly extracted surface ponded water (~56–72%) during both wet and dry seasons followed by soil surface (0–0.02 m) water (~17–19%) during growth. Dry rice extracted ~40–50% of its water from shallow soil (0–0.5 m) and ~35% from 0.1 to 0.3 m depth when the plants were matured. Conclusions: The mixing model results were better constrained with the additional information on soil water content and root length density. The relative contributions of the soil water sources to RWU decreased with depth and reflected the exponential shape of the root density profile. The main water source for wet rice was surface ponded water (independent of the season), whereas shallow soil water was the main source for dry rice.