How did we measure CO2 accurately pre-industry? Isn't the rise of industry kind of coincident with the ability to measure gases accurately? And if the trend as accuracy increased was up, wouldn't it suggest initial underestimates?

Or is there some reliable way we can measure in retrospect?