Reliable, on-line and high frequency data are essential for increasing our knowledge regarding nitrate (and nitrite) reactivity in a clay formation surrounding a repository for nitrate-containing nuclear waste. In this paper, the performance and applicability of an on-line UV spectrophotometer and pH and redox probes for real-time monitoring of in situ nitrate and nitrite reactivity in Opalinus Clay water has been evaluated, as part of the Bitumen-Nitrate-clay interaction (or BN) experiment at the Mont Terri Rock Laboratory (Switzerland). Standardly, the spectrophotometer uses a multiwavelength-based algorithm combined with Partial-Least-Square regression analysis. This technique provides highly reliable measurements of the nitrate and nitrite concentrations (relative standard deviation of 0.3–1.7% for NO3 − and 0.1–3% for NO2 −) in artificially prepared Opalinus Clay water, even in the presence of high amounts of other UV absorbing species (e.g. chloride). Although a severe sensitivity of the nitrite measurements to turbidity was shown using the multiwavelength-based algorithm, accurate nitrite measurements based on UV absorbance spectra are still possible, even in turbid conditions, when applying a novel single wavelength-based algorithm including normalisation of the UV spectra for turbidity. A similar new algorithm was developed for accurate measurements of nitrate concentrations, higher than the limit of quantification using the multiwavelength-based algorithm and even in turbid conditions. Furthermore, as no organics or significant amounts of chloride are leaching from the pH and redox electrodes during long periods, its polymeric interface is proven very stable, demonstrating its long-term in situ applicability. Based on the results of an in situ test in the BN borehole in the Opalinus Clay, the combined on-line equipment provides detailed and accurate monitoring data of the in situ nitrate and nitrite reactivity. The probes immediately react upon changes in the nitrate and nitrite concentrations, enabling the detection and identification of intermediate and/or fast processes, which are frequently missed when only performing off-line analyses on sampled solutions. This is especially valuable for in situ experiments such as the BN experiment where the sampling frequency is necessarily limited and microbial processes can occur at a high rate. This on-line equipment is thus an efficient tool to assess the reactivity of nitrate and nitrite under in situ conditions.