Recent results from animal irradiation studies have demonstrated the potential of ultra-high dose rate irradiation (also known as FLASH) for reducing radiation toxicity in normal tissues. However, despite mounting evidence of a 'FLASH effect', a mechanism has yet to be elucidated. This article hypothesizes that the radioprotecting effect of FLASH irradiation could be due to the specific sparing of hypoxic stem cell niches, which have been identified in several organs including the bone marrow and the brain. To explore this hypothesis, a new computational model is presented that frames transient radiolytic oxygen depletion (ROD) during FLASH irradiation in terms of its effect on the oxygen enhancement ratio (OER). The model takes into consideration oxygen diffusion through the tissue, its consumption by metabolic cells, and its radiolytic depletion to estimate the relative decrease in radiosensitivity of cells receiving FLASH irradiation. Based on this model and the following parameters (oxygen diffusion constant [Formula: see text]  =  2 · 10-5 cm2 s-1, oxygen metabolic rate m  =  3 mmHg s-1, ROD rate L ROD  =  [Formula: see text] mmHg Gy-1, prescribed dose D p  =  10 Gy, and capillary oxygen tension p 0  =  40 mmHg), several predictions are made that could be tested in future experiments: (1) the FLASH effect should gradually disappear as the radiation pulse duration is increased from  <1 s to 10 s; (2) dose should be deposited using the smallest number of radiation pulses to achieve the greatest FLASH effect; (3) a FLASH effect should only be observed in cells that are already hypoxic at the time of irradiation; and (4) changes in capillary oxygen tension (increase or decrease) should diminish the FLASH effect.