As the effects of global climate change intensify, it is increasingly important to implement more effective water management practices, particularly in arid and semi-arid regions such as Meoqui, Chihuahua, situated in the arid northern center of Mexico. The objective of this study was to identify the optimal time-series model for analyzing the pattern of water extraction volumes and predicting a one-year forecast. It was hypothesized that the volume of water extracted over time could be explained by a statistical time-series model, with the objective of predicting future trends. To achieve this objective, three time-series models were evaluated. To assess the pattern of groundwater extraction, three time-series models were employed: the seasonal autoregressive integrated moving average (SARIMA), Prophet, and Prophet with extreme gradient boosting (XGBoost). The mean extraction volume for the entire period was 50,935 ± 47,540 m³, with a total of 67,233,578 m³ extracted from all wells. The greatest volume of water extracted has historically been from urban wells, with an average extraction of 55,720 ± 48,865 m³ and a total of 63,520,284 m³. The mean extraction volume for raw water wells was determined to be 20,629 ± 19,767 m³, with a total extraction volume of 3,713,294 m³. The SARIMA(1,1,1)(1,0,0)₁₂ model was identified as the optimal time-series model for general extraction, while a “white noise” model, an ARIMA(0,1,0) for raw water, and an SARIMA(2,1,1)(2,0,0)₁₂ model were identified as optimal for urban wells. These findings serve to reinforce the efficacy of the SARIMA model in forecasting and provide a basis for water resource managers in the region to develop policies that promote sustainable water management.

Water is a vital resource for sustaining life and for numerous processes within the transformation industry. It is a finite resource, albeit one that can be renewed, and thus sustainable management is imperative. To achieve this objective, it is necessary to have the appropriate tools to assist with the planning policies for its management. This paper presents a time series analysis approach to measure and predict the pattern of water consumption by humans throughout subsectors (domestic, commercial, public sector, education, industry, and raw water) and total water consumption in Meoqui, Chihuahua, Mexico with data from 2011 to 2023, applying calibration model techniques to measure uncertainty in the forecasting. The municipality of Meoqui encompasses an area of 342 km². The climate is semi-arid, with an average annual rainfall of 272 mm and average temperatures of 26.4 °C in summer and 9.7 °C in winter. The municipal seat, which has a population of 23,140, is supplied with water from ten wells, with an average consumption of 20 ± 579 m³ per user. The consumption of the general population indicates the existence of a seasonal autoregressive integrated moving average (SARIMA) (0,1,2)(0,0,2)₁₂ model. (Sen’s Slope = 682.7, p < 0.001). The domestic sector exhibited the highest overall consumption, with a total volume of 17,169,009 m³ (13 ± 93). A SARIMA (2,1,0)(2,0,0)₁₂ model was estimated, with a Sen’s slope of 221.65 and a p-value of less than 0.001. The second-largest consumer of total water was the “raw water” sector, which consumed 5,124,795 (30,146 ± 35,841) m³ and exhibited an SARIMA (0,1,1)(2,0,0)₁₂ model with no statistically significant trend. The resulting models will facilitate the company’s ability to define water resource management strategies in a sustainable manner, in alignment with projected consumption trends.

A variety of statistical techniques were assessed for their usefulness for analysing the pattern of geographical and temporal changes in groundwater levels in order to diagnose the water supply in Meoqui, Chihuahua, which is situated in dry North-Central Mexico. These included Facebook Prophet, Lasso, generalized linear regularized models (GLMNET), autoregressive-integrated moving average (ARIMA), and a hybrid approach that merged Prophet and XGBoost. This was conducted on the assumption that the levels varied during the year and that it was possible to perform statistical analyses that derived a model explaining the changes and allowing for spatiotemporal prediction. The data set encompassed the period between 2020 and 2025 and was obtained from the Junta Municipal de Aguas y Saneamiento (JMAS) in Meoqui. The data set consisted of eight wells from which water was extracted for human consumption. The ARIMA model was identified as the optimal method for generating predictions on a monthly and annual basis. Furthermore, an inverse distance weighted (IDW) interpolation approach was utilised to conduct a spatial analysis. This enabled the visualisation of the predicted spatiotemporal changes in groundwater levels. The mean overall level was determined to be 26 m ± 16, with a minimum of 3 m and a lower level of 64 m. Models were estimated, comprising a general model and models specific to each well type. The best model for general level was Facebook Prophet (MAE 6.31, MAPE 16.28, MASE 1.23, SMAPE 16.79, RMSE 7.25, R-sq 0.29). The Sen’s slope of the historic level was found to be 0.38 (p < 0.001), thus indicating a decline in the groundwater level. The spatiotemporal analysis indicated a monthly decline in water levels from February to August, followed by an improvement in levels until November, which were then maintained until January. The lowest levels were observed in the area associated with Well 5. The findings of this study offer valuable insights into the spatiotemporal patterns of groundwater in the region, which could inform the development of sustainable groundwater management policies.