Abstract: Large-scale,high-enthalpy arc heated wind tunnels are the most reliable ground test facilities to test thermal protection materials and heat shield structures for space vehicles. Flow conditions in the facility need to be monitored.Currently,the facility conditions are defined using either the anticipated surface temperature with an assuming emissivity or the expected heat flux level.While this is useful to evaluate relative performance of the facility,it is not sufficient for quantitative measurement of the flow conditions.Temperature is one of the most important thermodynamic quantities in determining arc heated wind tunnel because it is a key parameter in determining arc-heater operating status and chemical reactions.Therefore,the development of accurate quantitative diagnostic techniques is necessary for better understanding the complex physics involved in the arc heated wind tunnel.The design of a tunable diode laser absorption spectroscopy (TDLAS)system to probe gas parameters during a bow shock wave ahead of a wa-ter cooled copper cylinder is presented in this paper.TDLAS is an effective method for measuring gas temperature and concentration in many fields due to its advantage of non-intrusive,high sensitivity, gas-specific and quick response.In our studies,an atomic oxygen absorption line near 777.2nm is utilized for detecting the arc-heated plasma using scanned-wavelength direct absorption mode with 100Hz repetition rate.The value of temperature is inferred directly from the Doppler broadening component of the absorption lineshape.Moreover,the number density of atomic oxygen is also determined through the integrated absorbance assuming local thermal equilibrium conditions. The agreement of the experimental observations and theoretical calculations shows that the ther-mal equilibrium assumption is valid.The current experimental results of this study illustrate the high potential of TDLAS measurements for routine and economical monitoring of arc heated wind tunnel operating status (gas temperature)as well as the time-resolved flow conditions in front of the model.