Auditing Construction-Dewatering Pumping Logs for Monitoring Prioritization: A Data-Limited Case Study from Loja, Ecuador

Routine construction-dewatering logs can support monitoring decisions when their processing rules and inferential scope are explicitly defined. This study audited 594 well-period records from 11 wells at a single urban construction site in Loja, Ecuador. Reading dates span 618 days (15 July 2024–25 March 2026), with the first monitoring interval beginning on 10 July 2024. Construction drawings indicate well lengths of 15.5–19.0 m. Reconstructed quality-control rules classified 406 records as Valid, 113 as Valid with caution, and 75 as Not recommended. Among 519 usable observation pairs, 509 (98.1%) had non-five-day intervals, resulting in a median bias of +60.0% when a default five-day divisor was applied. The inclusive 13-date panel comprised 111 Valid and 32 caution observations and showed pronounced inter-well ordering (Kendall’s W = 0.946). Sensitivity analyses based on quality thresholds and reduced-well configurations quantified the dependence of results on record-retention rules. Ordinary Theil–Sen screening was complemented by an irregular-time, dependence-aware log-GLS analysis and conditional detectability estimates; therefore, temporal signals are interpreted as exploratory rather than diagnostic. Recalculated Hydro-Operational Monitoring Priority Scores (HOMPS) placed Wells 10, 2, 8, and 5 in the highest sample-defined priority tier. Well 10 remained the highest-priority well under every component-omission scenario, providing partial support for H3. The study provides a transparent and reproducible framework for auditing routine dewatering records and prioritizing monitoring attention while explicitly avoiding unsupported inference about aquifer stress.