Design of Groundwater Level Monitoring Networks for Maximum Data Acquisition at Minimum Travel Cost

Groundwater monitoring networks represent the main source of information about water levels and water quality within aquifers. In this paper, a method is proposed for the optimal design of monitoring networks to obtain groundwater-level data of high spatial relevance at a low cost. It uses the estimate error variance reduction obtained with the static Kalman filter as optimization criteria, while simultaneously evaluating the optimal routes to follow through the traveling salesman problem. It was tested for a network of 49 wells in the Calera aquifer in Zacatecas, Mexico. The study area was divided into three zones, and one working day (8 h) was taken to visit each one, with an average speed of 40 km/h and a sampling time of 0.5 h. An optimal network of 26 wells was obtained with the proposal, while 21 wells should be monitored if the optimal routing is neglected. The average standard error using 49 wells of the original network was 35.01 m, an error of 38.35 m was obtained for 21 wells (without optimal routing) and 38.36 m with the 26 wells selected using the proposal. However, the latter produce estimates closer to those obtained with the 49 wells. Following the proposal, more field data can be acquired, reducing costs.