In this PhD, I have advanced the study of ground-based photometric observations in the near-infrared. Specifically, I have worked in the context of robotic exoplanet transit surveys in collaboration with the SPECULOOS Southern Observatory (SSO). Here, I targeted optimising the photometric precision of observing late M, L type stars by developing correction methods and a new instrument called SPIRIT.

My first original contribution was the development of a correction method for the induced effects from varying precipitable water vapour (PWV) in our atmosphere, specifically on differentially resolved light curves from SSO. This work succeeded in reducing false variability of time-series data from late M, L type stars on both long and short timescales, to the extent of removing false transit features.

In parallel, I performed a feasibility study of introducing new near-infrared instrumentation to SSO. One which would permit better photometric precision than the existing CCD Si based instrumentation, and likewise minimise the induced effects of PWV variability. An InGaAs based instrument, sensitive up to 1.62 µm was identified, and a custom wide-pass filter called zYJ was designed and manufactured to form SPeculoos' Infra-Red photometric Imager for Transits (SPIRIT). It proved to be a significantly lower-cost alternative to the traditionally used HgCdTe based instrumentation, as well as being better suited to robotic observatories.

On sky results of SPIRIT at SSO successfully demonstrated better photometric precision for stars below 2550 K than the existing instrumentation. It similarly demonstrated the benefit of seeing further into the infrared for minimising the observed variability of M, L type stars. Finally, the custom designed wide-pass filter, zYJ, successfully demonstrated a significantly lower sensitivity to PWV variability.

These results pave a new avenue for ground-based near-infrared robotic exoplanet transit surveys, as well as similar time-series focused astronomy. I conclude my work by suggesting viable routes to further improve the photometric precision of such new instrumentation.