Long-term historical data on the presence of invasive pests can inform current pest management. By examining correlations between shifts in pest occurrences and various potential drivers, we can better inform decision-making and management strategies. However, the availability of such long-term data is often limited. We introduce a data rescue protocol to recover difficult-to-access pest information from periodical documents such as annual forestry health reports. We applied this to annual forestry reports on the island of Ireland from 1970-2020. This protocol resulted in an open-access dataset of pest dynamics and their management for the island. We combined the pest dataset with auxiliary weather data to estimate the effects of surveying effort, control measures and weather upon pest outbreak dynamics. Two first-order auto-logistic regression models were used to model rates of introduction and rates of eradication in an insect model and a bacteria, chromista and fungi model. The results provide evidence that multi-year systematic surveillance efforts have diminishing returns for estimating the rate of introduction, and surveillance effort requirements differ between specific pest groups. Control measures had a minor positive effect on the rate of local pest eradication, with high variation in the data. Our methodology for data collection and analysis serves as a blueprint for other regions of the world and other invasive species assemblages where data is physically available but not ready for analysis.