Population growth worldwide will bring with it an increasing number of problems regarding drinking water demand and its relationship with availability. According to the water statistics for Mexico in 2019, published by the Comisión Nacional del Agua (National Water Commission, ^{Conagua [2019]}), Mexico receives approximately 1 449 471 million cubic meters of water annually through precipitation. Of this total, approximately 72.1% returns to the atmosphere through evapotranspiration, 21.4% flows through rivers and streams, and the remaining 6.4% infiltrates into the subsoil, naturally recharging aquifers. When accounting for cross-border outflows and inflows, Mexico has a 451 584.7 Hm³ annual renewable freshwater availability (^{Conagua, 2019}).

• ^{Conagua [2019]}
  Estadísticas del agua en México 2019, 2019
  Conagua. 2019. Estadísticas del agua en México 2019. Comisión Nacional del Agua, Secretaría de Medio Ambiente y Recursos Naturales, Mexico.
• ^{Conagua, 2019}
  Estadísticas del agua en México 2019, 2019
  Conagua. 2019. Estadísticas del agua en México 2019. Comisión Nacional del Agua, Secretaría de Medio Ambiente y Recursos Naturales, Mexico.

Surface water availability in Mexico is a critical issue influenced by climatic variability, population growth, and unsustainable water management practices. Studies such as those by ^{Mekonnen and Hoekstra (2016)} highlight the global and regional water scarcity challenges, with Mexico facing significant stress due to uneven distribution and overexploitation of surface water resources. ^{López-Morales (2017)} provides a comprehensive overview of Mexico’s water resources, highlighting that surface water availability is highly seasonal and spatially uneven, with most resources concentrated in the southern regions, while the north of the country faces severe water scarcity. Climate change further complicates this scenario, as highlighted by ^{Magaña et al. (1997)}, who projected reduced precipitation and increased evaporation rates, thereby threatening the long-term sustainability of surface water. Additionally, ^{Hernández-Espriú et al. (2014)} discussed the interplay between surface and groundwater systems, noting that over-reliance on groundwater in regions like Mexico City indirectly impacts surface water availability. Collectively, these studies underscore the urgent need for integrated water management strategies to address Mexico’s surface water challenges in the face of the growing demand and climate variability.

• ^{Mekonnen and Hoekstra (2016)}
  Four billion people facing severe water scarcity

  Science Advances, 2016
  Four billion people facing severe water scarcity
• ^{López-Morales (2017)}
  El estado del agua en México: retos, oportunidades y perspectivas

  El agua en México: actores, sectores y paradigmas para una transformación social-ecológica, 2017
  El estado del agua en México: retos, oportunidades y perspectivas
• ^{Magaña et al. (1997)}
  Assessment of current and future regional climate scenarios for Mexico

  Climate Research, 1997
  Assessment of current and future regional climate scenarios for Mexico
• ^{Hernández-Espriú et al. (2014)}
  The DRASTIC-Sg model: An extension to the DRASTIC approach for mapping groundwater vulnerability in aquifers subject to differential land subsidence, with application to Mexico City

  Hydrogeology Journa, 2014
  The DRASTIC-Sg model: An extension to the DRASTIC approach for mapping groundwater vulnerability in aquifers subject to differential land subsidence, with application to Mexico City

Despite these substantial resources, Mexico faces a complex water challenge, including regional imbalances in water availability, basin overexploitation, and increasing demand. The 2030 water agenda (^{Conagua, 2012}) emphasizes these disparities, particularly in critical basins. While regions such as Grijalva-Usumacinta (HR 30) use only a small fraction of their available water, others, like the Lerma and Bravo basins, exceed 100% of their natural surface water availability. This overuse threatens ecosystems and undermines sustainable economic development. Currently, Mexico’s national water demand stands at 78 400 Hm³ annually, of which 11 500 Hm³ are unsustainably sourced. Projections suggest that, without intervention, this gap could double within the next 20 years. Addressing these challenges requires an integrated water management strategy that not only tackles water scarcity but also considers issues of water quality, governance, and sustainability. These efforts must integrate the interconnected social, economic, and environmental dimensions of water resources. Collaborative action among government agencies, stakeholders, and communities is essential to ensure equitable access to clean water, promote environmental sustainability, and build resilience to future water-related challenges.

• ^{Conagua, 2012}
  Agenda del agua 2030. Avances y logros 2012, 2012
  Conagua. 2012. Agenda del agua 2030. Avances y logros 2012. Comisión Nacional del Agua, Secretaría de Medio Ambiente y Recursos Naturales, Mexico.

This paper examines surface water availability in Mexico across 757 basins, organized into 37 hydrological regions, using the methodology outlined in the Official Mexican Standard NOM-011-CONAGUA-2015 (^{Semarnat, 2015}). It is essential to note that while numerous manuals are available for estimating water availability in a basin, these primarily focus on natural availability or natural runoff. This is distinct from water availability for administrative or legal purposes, as outlined in ^{Silva-Hidalgo et al. (2013)}. For instance, ^{López-García et al. (2017)} conducted a water balance to determine natural availability under climate change scenarios for the Galeana Valley aquifer in the state of Nuevo León, Mexico. Similarly, ^{Loor (2017)} performed a watershed balance in Ecuador, and ^{Ordóñez-Gálvez (2011)} presented a methodology to estimate the natural surface water balance in Peru. The United Nations Educational, Scientific and Cultural Organization (UNESCO) incorporated the consumptive use variable (Uc) into the surface water balance of the Valley of Mexico basin, and calculated its water balance variability and uncertainty components (^{Aparicio-Mijares et al., 2006}),

• ^{Semarnat, 2015}
  Norma Oficial Mexicana NOM-011-CONAGUA-2015. Conservación del recurso agua. Que establece las especificaciones y el método para determinar la disponibilidad media anual de las aguas nacionales

  Secretaría de Medio Ambiente y Recursos Naturales, Mexico. Diario Oficial de la Federación, March 27, 2015
  Norma Oficial Mexicana NOM-011-CONAGUA-2015. Conservación del recurso agua. Que establece las especificaciones y el método para determinar la disponibilidad media anual de las aguas nacionales
• ^{Silva-Hidalgo et al. (2013)}
  Metodología para la determinación de disponibilidad y déficit de agua superficial en cuencas hidrológicas: aplicación al caso de la normativa mexicana

  Tecnología y Ciencias del Agua, 2013
  Metodología para la determinación de disponibilidad y déficit de agua superficial en cuencas hidrológicas: aplicación al caso de la normativa mexicana
• ^{López-García et al. (2017)}
  Disponibilidad hídrica bajo escenarios de cambio climático en el Valle de Galeana, Nuevo León, México

  Tecnología y Ciencias del Agua, 2017
  Disponibilidad hídrica bajo escenarios de cambio climático en el Valle de Galeana, Nuevo León, México
• ^{Loor (2017)}
  Estudio del balance hídrico superficial de las cuencas hidrográficas sector San Juan y San Bartolomé en el área de intercepción con el bosque y vegetación protector Aguarongo, 2017
  Estudio del balance hídrico superficial de las cuencas hidrográficas sector San Juan y San Bartolomé en el área de intercepción con el bosque y vegetación protector Aguarongo
• ^{Ordóñez-Gálvez (2011)}
  Balance hídrico superficial. Contribuyendo al desarrollo de una cultura del agua y la gestión integral de recurso hídrico. Cartilla técnica, 2011
  Balance hídrico superficial. Contribuyendo al desarrollo de una cultura del agua y la gestión integral de recurso hídrico. Cartilla técnica
• ^{Aparicio-Mijares et al., 2006}
  Evaluación de los recursos hídricos. Elaboración del balance hídrico integral por cuencas hidrográficas. Documento Técnico del phi-LAC No. 4, 2006
  Evaluación de los recursos hídricos. Elaboración del balance hídrico integral por cuencas hidrográficas. Documento Técnico del phi-LAC No. 4

Although most publications focus on the natural water balance, countries such as Chile, Spain, Mexico, and the USA have normative documents for water resource management. These documents are based on the fundamental equation (dV)(dt)^-1 = E - S, which states that the volume change (V) is equal to the inputs (E) minus the outputs (S) of water over a specific period (t) (^{Aparicio-Mijares et al., 2006}). The primary difference lies in the methodology used to estimate the data in the balance equation. Unlike the natural water balance, in the administrative or regulatory balance, consumptive use (Uc) is considered an output (S).

• ^{Aparicio-Mijares et al., 2006}
  Evaluación de los recursos hídricos. Elaboración del balance hídrico integral por cuencas hidrográficas. Documento Técnico del phi-LAC No. 4, 2006
  Evaluación de los recursos hídricos. Elaboración del balance hídrico integral por cuencas hidrográficas. Documento Técnico del phi-LAC No. 4

This paper uses a robust methodology to assess average annual surface water availability, quantifying the delta between runoff volume and existing downstream water commitments. Crucially, it extends beyond mere quantification to pinpoint basins exhibiting optimal potential for hydroelectric development by 2034. This is achieved through a meticulous evaluation of topographic characteristics, interwoven with a comprehensive consideration of ecological, social, and infrastructural constraints. By dissecting these multifaceted dynamics, this paper aims to make a significant contribution to a sustainable and resilient water management strategy for Mexico. It seeks to ensure responsible and equitable stewardship of natural resources, while simultaneously illuminating viable pathways for sustainable energy production. Figure 1 illustrates Mexico’s division into 37 hydrological regions, representing the country’s primary watersheds or basins. These regions define water flow dynamics, availability, and catchment boundaries, serving as a critical framework for water resource management, conservation, and infrastructure planning.

As outlined in the Official Mexican Standard NOM-011-CONAGUA-2015 (^{Semarnat, 2015}), which provides specifications and methodology for determining the average annual availability of national waters in Mexico, average surface water availability is estimated using the following equation:

• ^{Semarnat, 2015}
  Norma Oficial Mexicana NOM-011-CONAGUA-2015. Conservación del recurso agua. Que establece las especificaciones y el método para determinar la disponibilidad media anual de las aguas nacionales

  Secretaría de Medio Ambiente y Recursos Naturales, Mexico. Diario Oficial de la Federación, March 27, 2015
  Norma Oficial Mexicana NOM-011-CONAGUA-2015. Conservación del recurso agua. Que establece las especificaciones y el método para determinar la disponibilidad media anual de las aguas nacionales

where: D is the average annual surface water availability in Hm³, A _b is the average annual runoff volume downstream in Hm³, and R _xy is the average annual volume committed downstream in Hm³.

The average runoff volume downstream (A _b ) is further calculated using Eq. (2):

where A _r is the average annual runoff volume from upstream basin; C _p is the average annual natural runoff volume; R _e is the annual returns volume; I _m is the annual import volume; E _x is the annual export volume; U _c is the annual surface water volume extraction (U _ca : annual surface water volume extraction through titles currently registered/assigned in the Water Rights Public Registry [REPDA, for its Spanish acronym], U _cb : annual surface water volume extraction from titles in the registration process at REPDA, and U _cc : annual volume corresponding to reserves and regulated areas); E _v is the average annual evaporation volume in reservoirs and water bodies; and A _v is the average annual storage volume variation in reservoirs (all variables are reported in Hm³).

Eq. (2) defines A _b as water balance inflows and outflows within a basin. Positive variables represent the water volume entering the basin, while negative variables indicate the water volume leaving it. To calculate water availability, both natural and anthropogenic factors that influence these variables are considered. The behavior of these variables over time was analyzed, and projected values were compared to historical records. The percentage change for each variable was then applied to the latest values published in a 2020 availability study (^{SINA, 2020}).

• ^{SINA, 2020}
  Detalle de las cuencas hidrológicas en México (2020), 2020
  SINA. 2020. Detalle de las cuencas hidrológicas en México (2020). Sistema Nacional de Información del Agua 3.0. Available at: https://sinav30.conagua.gob.mx:8080/SINA/?opcion=cuencas

Hydrological regions in Mexico are composed of multiple hydrological basins, and water availability was calculated both by basin and by hydrological region using Eq. (1). The D value in Eq. (1) was calculated for each hydrological basin. However, the methodology established in NOM-011-CONAGUA-2015 specifies that this calculation must be applied to the entire hydrological region, as it is performed from the downstream basin to the upstream basin. It is important to note that NOM-011-CONAGUA-2015 was designed for its application to a single hydrological basin, which must have a single outlet. In contrast, a hydrological region comprises multiple hydrological basins, and not all hydrological regions in Mexico drain into a single outlet. Many hydrological regions in Mexico have multiple outlets draining into the sea, which does not align with the requirements of NOM-011-CONAGUA-2015.

In this analysis, natural variables include precipitation and temperature, and anthropogenic variables include water usage or consumptive uses (U _c ). These inputs are essential for calculating the water balance within a basin (Cp). The methodology to estimate consumptive water uses and the climatological variables required for assessing water availability in Mexico for 2020 and 2034 are detailed below.

^{Bonsal et al. (2020)} conducted a review analyzing freshwater supply and demand in the Canadian Cordillera, highlighting both historical and future changes in water availability resulting from glacial melt. They found that projected impacts are greater on the seasonality of water flow, with increases in winter and decreases in summer, especially under high-emission scenarios. Southern regions, such as Saskatchewan and Okanagan, will face greater vulnerabilities due to summer water scarcity and growing economic demands. In the north, changes in the landscape, such as permafrost thaw, will impact water quantity and quality. To project future water use to 2034, we analyzed primary sector water demands like agricultural, industrial, and urban public uses. Trends in production systems, population growth, and economic activity were taken into consideration. Regression analysis was used to obtain these trends and extrapolate them to 2034. The projection used to estimate the value by 2034 is based on historical data from 1976 to 2020. However, it is important to acknowledge that long-term projections are subject to uncertainty due to external factors that may impact the results. Therefore, the presented results should be interpreted as a projection or scenario based on a trend, rather than an exact prediction.

• ^{Bonsal et al. (2020)}
  Western Canadian freshwater availability: Current and future vulnerabilities

  Environmental Reviews, 2020
  Western Canadian freshwater availability: Current and future vulnerabilities

Data on planted areas for various crops in irrigation districts and units was sourced from Conagua and the Servicio de Información Agroalimentaria y Pesquera (Agri-Food and Fisheries Information Server, SIAP). Analyzing trends in planted areas by crop type is essential for projecting future water demands.

Population growth data was obtained from the Instituto Nacional de Estadística y Geografía (National Institute of Statistics and Geography) (^{INEGI, 2020}) and the Consejo Nacional de Población (National Population Council, Conapo). Trends in population growth were used to project future water demands.

• ^{INEGI, 2020}
  Censo de población y vivienda 2020, 2020
  INEGI. 2020. Censo de población y vivienda 2020. Instituto Nacional de Estadística y Geografía, Mexico. Available at: Available at: https://inegi.org.mx/programas/cObv/2020/ (accessed September 2021).

Data on gross domestic product and production was sourced from INEGI. Trends in economic activity were used to project future water demands.

After calculating water volumes by use and by basin for the years 2020 and 2034, the growth rate between these years was estimated by dividing the volume in 2034 by the volume in 2020 and subtracting 1[(Vol₂₀₃₄)(Vol₂₀₂₀)^-1] - 1. This growth rate (see Table SI in the supplementary material) of the consumptive use is then applied to the uses included in the 2020 surface water availability studies. It is important to note that the water volume for consumptive uses does not differentiate between surface water and groundwater sources; the primary interest is to estimate the percentage increase or decrease in consumptive uses.

^{Boulanger et al. (2005)} analyzed long-term trends in precipitation within the Paraná-Plata basin. They highlighted a positive increase in the precipitation total index (PTI) from the late 1960s to the early 1980s. They observed a significant increase in precipitation from November to May in southern Brazil and Argentina. Changes in the El Niño-Southern Oscillation (ENSO) characteristics have influenced the variability of precipitation, making it difficult to define robust statistical relationships between ENSO and precipitation in the basin. Additionally, they emphasized the limited usefulness of linear statistical forecast systems for predicting impacts at the local or regional scale.

• ^{Boulanger et al. (2005)}
  Observed precipitation in the Paraná-Plata hydrological basin: Long-term trends, extreme conditions and ENSO teleconnections

  Climate Dynamics, 2005
  Observed precipitation in the Paraná-Plata hydrological basin: Long-term trends, extreme conditions and ENSO teleconnections

To analyze the climatic variables of the 757 basins, this work utilized the results of ^{Ramírez-Villa et al. (2022)}, who obtained precipitation and temperature records from the CLImate COMputing project (Clicom). This database, encompassing 5442 meteorological stations across Mexico, provided the foundational data. They obtained historical precipitation (1976-2018) and temperature data (1976-2015). To ensure data integrity, they implemented a rigorous quality control process, leveraging the CLIMATOL software (v. 3.1) within the R environment. This process incorporated the ^{Paulhus and Kohler (1952)} method, addressing anomalous values to facilitate data homogenization and rectify missing data. To address the temporal variability in station operation, ^{Ramírez-Villa et al. (2022)} implemented a rigorous data selection process, adhering to World Meteorological Organization guidelines (^{WMO, 2011}) to identify stations with data completeness of at least 80% from 1976 to 2018. Additionally, they calculated the annual accumulated precipitation for each selected station and Thiessen polygons, as stipulated by NOM-011-CONAGUA-2015, to estimate the average rainfall distribution. Polygons corresponding to stations within each basin were extracted, along with their influence areas, enabling the calculation of cumulative annual precipitation per basin (Eq. [3]).

• ^{Ramírez-Villa et al. (2022)}
  Conformación y análisis de climatologías de precipitación y temperaturas para las cuencas hidrológicas de México, 2022
  Conformación y análisis de climatologías de precipitación y temperaturas para las cuencas hidrológicas de México
• ^{Paulhus and Kohler (1952)}
  Interpolation of missing precipitation records

  Monthly Weather Review, 1952
  Interpolation of missing precipitation records
• ^{Ramírez-Villa et al. (2022)}
  Conformación y análisis de climatologías de precipitación y temperaturas para las cuencas hidrológicas de México, 2022
  Conformación y análisis de climatologías de precipitación y temperaturas para las cuencas hidrológicas de México
• ^{WMO, 2011}
  Guide to climatological practices. WMO No. 100, 2011
  WMO 2011. Guide to climatological practices. WMO No. 100. World Meteorological Organization, Geneva. Available at: Available at: https://community.wmo.int/en/activity-areas/climate/draft-fourth-edition-guide-climatological-practices-wmo-no-100 (accessed March 2022).

A linear regression model, trained on data from 1976 to 2018, was used by ^{Ramírez-Villa et al. (2022)} to project precipitation and temperatures to 2034, employing Excel-based models. Acknowledging the inherent uncertainty of regression models, this study includes the calculation and presentation of statistical metrics such as R² and root mean square error (RMSE), comparing observed historical data with model-generated predictions (shown in Table SI in the supplementary material). These predictions were taken as projections or scenarios due to low R² and RMSE values in some basins. For example, in basin 3708 (Sierra Madre), with an average annual rainfall of 461.7 mm, the R² value is 0.005. On the other hand, in basin 2305 (La Punta), the R² value is 0.435.

• ^{Ramírez-Villa et al. (2022)}
  Conformación y análisis de climatologías de precipitación y temperaturas para las cuencas hidrológicas de México, 2022
  Conformación y análisis de climatologías de precipitación y temperaturas para las cuencas hidrológicas de México

where P _i is the annual accumulated precipitation for each meteorological station located within a basin, A _i is the influencing area of the basin, and n is the total polygons number associated with stations surrounding each basin.

To analyze the temperature parameter, a selection process identified 1512 meteorological stations with complete data spanning from 1976 to 2015. This selection was necessary due to a significant reduction in available station data after 2015. A comprehensive data quality assessment was conducted with this robust dataset, followed by the calculation of annual average maximum and minimum temperatures for each station. Subsequently, two interpolations were carried out for each year (maximum and minimum annual temperature) from 1976 to 2015 using Kriging regression. This method generated a mesh with a resolution of 0.05º for the entire Mexican territory. Additionally, an elevation adjustment model was used to improve the quality of the results due to the high correlation between these variables. The temperature average value in each Mexican basin from 1976 to 2015 was obtained using Eq. (4):

where T _i represents the temperature grid points per year located within each basin (in ºC), and n is the total number of polygons located within the basin.

Similar to the procedure for obtaining precipitation projections, temperature projections were obtained. A linear regression Excel model was used for each of the 757 hydrographic basins, utilizing their historical records, to estimate the maximum and minimum annual average temperature values (in ºC) for 2034 (see Table SI in the supplementary material).

^{Koshida et al. (2015)} examined the impact of climate variability and projected climate change on water availability in Canada. This research, the first in a three-part series, reviewed and compared different approaches to estimating water availability, categorized into three types: climate-based, hydrological, and water supply/demand indicators. Climate-based indicators use variables such as precipitation and evapotranspiration to calculate water balances. Hydrological indicators focus on river flow and runoff. Supply/demand indicators compare the volume of available water with water use. The study provides insights into the current state of water availability estimates in Canada. Water volume per local basin (Cp) is the total water volume that originates within a specific hydrological basin and is available for potential use. This volume is measured in cubic hectometers (Hm³) and is calculated based on factors such as precipitation, temperature, evaporation, and the basin's physical characteristics. To estimate Cp, three sources were considered:

• ^{Koshida et al. (2015)}
  Climate and water availability indicators in Canada: Challenges and a way forward. Part I. Indicators

  Canadian Water Resources Journal/Revue Canadienne des Resources Hydriques, 2015
  Climate and water availability indicators in Canada: Challenges and a way forward. Part I. Indicators

• Historical Cp: values from the 2020 availability study (^{Semarnat, 2020}).

• Turc’s method: this method estimates Cp based on precipitation, temperature, and basin area (^{Sánchez, 2022}).

• The runoff coefficient (Rc) method: this method estimates Cp based on precipitation, basin area, and a runoff coefficient.

• ^{Semarnat, 2020}
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

  Secretaría de Medio Ambiente y Recursos Naturales, Mexico. Diario Oficial de la Federación, September 21, 2020
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos
• ^{Sánchez, 2022}
  Evapotransporación. Hidrología, hidrogeología, 2022
  Evapotransporación. Hidrología, hidrogeología

Both the Turc’s and Rc methods were applied to historical and projected climate data to estimate Cp for the current scenario and the projected 2034 scenario.

This method estimates Cp by calculating the difference between precipitation and actual evapotranspiration (ETR). Eq. (5) demonstrates that ETR is functionally dependent on both annual precipitation and average annual temperature.

where E is the annual specific runoff, P the annual precipitation, and ETR the actual ETR in the basin (all in mm).

ETR is calculated with the following expression:

If P > 0.31 L then

where L is obtained as:

where, in turn, P is the annual precipitation (in mm) and T is the average temperature (in ºC).

If P < 0.31 L then

Natural runoff average annual volume (Cp, in m³), is obtained with the expression:

where A represents the basin area (in m²) and E the specific annual runoff (in m).

The volume per basin was calculated with Turc’s method, where the most important variables are precipitation and temperature, as well as the area of each basin.

The Cp volume was based on the average precipitation data from 1976 to 2018 and temperature data from 1976 to 2015. Using the same methodology, Cp was also estimated using precipitation and temperature projections for 2034. The percentage change between these values was then determined using Eq. (10):

where % change is the percentage of change between an average annual value and that projected for 2034, C _p2034 is the volume per local basin estimated for 2034 by Turc’s method (Hm³), and Cp average is the volume per local basin estimated with average annual climatic values by the Turc’s method (Hm³).

As outlined in NOM-011-CONAGUA-2015, Eq. (11) estimates C _p by the Rc method using a runoff coefficient that depends on land use and soil type. Rc values were either obtained directly from the Conagua database or estimated based on data from similar basins. In this study, land use and soil type were assumed to remain unchanged through 2034.

where Cp is the volume per local basin (m³), P is the average annual rainfall (m), A is the basin area (m²), and Rc is the runoff coefficient (dimensionless).

To maintain temporal consistency across our analysis, a percentage change calculation, as defined by Eq. (10), was performed. However, this calculation was specifically applied to the average annual precipitation data spanning from 1976 to 2018. This deliberate restriction of the time period was implemented to ensure uniformity with both the temperature parameter dataset and the projected precipitation values for 2034.

Three scenarios were considered for estimating surface water availability.

• 2020 availability: Values from the 2020 availability study (^{Semarnat, 2020}).

• 2034 availability (%change using Turc’s method): The percentage change calculated using Turc’s method was applied to the 2020 Cp value to estimate the corresponding Cp for 2034. Consumptive use (Uc for 2020) was adjusted based on its projected growth rate for 2034. Other variables in Eq. (2), such as R _e , I _m , E _x , E _v , and A _v , were assumed to remain constant.

• 2034 availability (%change using the Rc method): Similarly, the percentage change derived using the Rc method was applied to the 2020 Cp value to calculate the corresponding Cp for 2034. Consumptive use (Uc for 2020) was adjusted based on its projected growth rate for 2034. Other variables in Eq. (2), such as R _e , I _m , E _x , E _v , and A _v , were assumed to remain constant.

• ^{Semarnat, 2020}
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

  Secretaría de Medio Ambiente y Recursos Naturales, Mexico. Diario Oficial de la Federación, September 21, 2020
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

Using the values obtained from the availability water document from ^{Semarnat (2020)} and the percentage changes from the Turc’s and Rc methods, the availability of surface water for 2034 was estimated. Eqs. 1 and 2 were applied to compute the two scenarios for 2034 based on the percentage changes from Turc’s and Rc. The same growth rate for Uc was used in both methods.

• ^{Semarnat (2020)}
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

  Secretaría de Medio Ambiente y Recursos Naturales, Mexico. Diario Oficial de la Federación, September 21, 2020
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

Basins characterized by surface water flows greater than or equal to 2 m³ s^-1 and slopes exceeding 2% were identified as potential sites for hydroelectric development. Legal and ecological constraints, including protected areas and environmental flow requirements, were considered.

The average annual available water volume was converted to flow rate in cubic meters per second (m² s^-1). Basins with slopes greater than 2% were derived from the slope map at a scale of 1:250000, published by the Instituto Nacional de Ecología (National Institute of Ecology) (^{INEGI, n.d.}).

• ^{INEGI, n.d.}
  Espacio y datos de México. Mapas. Pendientes del terreno 1:250:000,
  Espacio y datos de México. Mapas. Pendientes del terreno 1:250:000

The agricultural sector is the primary water consumer in Mexico. To ensure sustainable water use, it is crucial to improve water efficiency and adopt sustainable agricultural practices. Water use patterns vary across hydrological regions, requiring region-specific water management strategies. Figure 2 illustrates the significant increases in agricultural water consumption projected for the Balsas, Bravo, Sinaloa, and Nazas-Aguanaval regions. Industrial water use will remain concentrated along the central axis, with increasing demand in northwestern states such as Chihuahua, Sonora, Sinaloa, and Durango. Urban public water use will continue to be highest in basins with large cities. By 2034, the Bajo Atoyac River basin in the Balsas region and the Río Blanco in the Papaloapan region will have the highest extraction volumes, while the Valley of Mexico basin will remain the primary basin for urban public water use. Given the large number of hydrological basins, the results are presented by hydrological regions, which group multiple basins into broader, more manageable categories for analysis.

Climate change is expected to have a significant impact on Mexico’s water resources. Studies by ^{Boulanger et al. (2005)} and ^{Koshida et al. (2015)} highlighted the potential consequences of shifting climate patterns on water availability. While long-term climate forecasts inherently carry uncertainty, it is crucial to recognize the potential impacts of climate change on Mexico’s water resources. Factors such as altered rainfall patterns, rising temperatures, and increased frequency of extreme weather events could exacerbate water scarcity and disrupt hydrological cycles. Our climatic variable findings are detailed in the following sections.

• ^{Boulanger et al. (2005)}
  Observed precipitation in the Paraná-Plata hydrological basin: Long-term trends, extreme conditions and ENSO teleconnections

  Climate Dynamics, 2005
  Observed precipitation in the Paraná-Plata hydrological basin: Long-term trends, extreme conditions and ENSO teleconnections
• ^{Koshida et al. (2015)}
  Climate and water availability indicators in Canada: Challenges and a way forward. Part I. Indicators

  Canadian Water Resources Journal/Revue Canadienne des Resources Hydriques, 2015
  Climate and water availability indicators in Canada: Challenges and a way forward. Part I. Indicators

Precipitation patterns across Mexico exhibit significant regional variability, with the Baja California Peninsula, along with the northwest and north-central regions, experiencing the lowest rainfall, while the southeast region, including Veracruz, Oaxaca, Tabasco, Chiapas, and Campeche, receives the highest precipitation. The projected precipitation for 2034 shows significant regional variability, with some areas projected to experience substantial increases (e.g., Rh 30: +41%) and others facing decreases (e.g., Rh 1, Rh 4, Rh 7: -48%). These changes underscore the need for region-specific water management strategies to address the challenges of water scarcity and flooding. These changes underscore the increasing variability in precipitation patterns, with some regions facing heightened water stress while others experience higher rainfall. Figure 5 illustrates the percentage change in precipitation between the 1976-2018 average and the projected values for 2034, emphasizing the need for region-specific water management strategies to address these shifts. Additionally, the linear regression model’s performance across 757 basins reveals significant disparities, with basins like 1804 (Río Nexapa) showing relatively strong predictive accuracy (R² = 0.33 and RMSE = 78.55), while others, such as 2302 (Tepanatepec), 2321 (Coatán), and 3030 (Paredón), exhibit poor performance (R² < 0.05, RMSE > 500). These values reinforce what is described in the methodology; the precipitation and temperature values projected for 2034 have a high level of uncertainty, so they are only considered as a scenario (Figs. 3 and 4).

• In 101 basins, maximum temperatures are projected to decrease slightly by 1% (0.5 ºC).

• In 656 basins, maximum temperatures are projected to increase by 3% (almost 1 ºC).

• Decreases in maximum temperature are concentrated in northern Mexico, Michoacán, coastal Oaxaca, and Campeche, and the southern border with Guatemala.

• Historical minimum temperatures are concentrated in the areas between the Sierra Madre Occidental and the Sierra Madre Oriental, together with Baja California Norte.

• In 299 basins, minimum temperatures are projected to decrease by 3% (-0.4 ºC), while in 458 basins, they are projected to increase by 3% (0.4 ºC).

• Decreases in minimum temperature are observed in northwestern Mexico, particularly in the mountains of Sinaloa, Chihuahua, and Durango.

• Increases in minimum temperature are observed in the Valley of Mexico basin and parts of Campeche.

Similarly, Figures 6 and 7 show the percentage changes in maximum and minimum temperatures, respectively, over the same period.

Turc’s and Rc methods were employed to estimate Cp under both historical and projected climate conditions, as well as to calculate the corresponding percentage changes. Turc’s method typically results in larger changes in Cp compared to Rc, which accounts solely for variations in precipitation. Figure 8 illustrates the water volume per local basin (Cp), as reported in the study published by ^{Semarnat (2020)}.

• ^{Semarnat (2020)}
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

  Secretaría de Medio Ambiente y Recursos Naturales, Mexico. Diario Oficial de la Federación, September 21, 2020
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

Figure 8 also shows that in 2020, most basins generated a runoff lower than 500 Hm³ per year, except for the Tarahumara mountains, the Balsas River, and the eastern slopes of the Sierra Madre. The change percentage predicts a decrease in Cp in Baja California Peninsula and northwestern territories, and an increase in the south-central region (Fig. 9). The Rc method shows a less significant change, but also indicates a decrease in Cp in the northwest (Fig. 10).

For 2020, basins without water availability were primarily concentrated in the following hydrological regions: 8 Sonora Norte, 9 Sonora Sur, 24 Río Bravo Conchos, 12 Lerma, and 18 Balsas, as well as basin 3405 Río Santa María 1 together with several basins in the southwestern area of hydrological region 25 San Fernando Soto la Marina. Figure 11 shows surface water availability for 2020, as published by ^{Semarnat (2020)}.

• ^{Semarnat (2020)}
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

  Secretaría de Medio Ambiente y Recursos Naturales, Mexico. Diario Oficial de la Federación, September 21, 2020
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

In the 2034 scenario, using Turc’s method, projections show 154 basins (25%) without availability, 266 basins (36%) with availability (but less than 100 Hm³), and 337 basins (38%) with more than 100 Hm³. For the 2034 projection, using the percentage change calculated by Turc’s method (Fig. 12), basins without availability are expected to be concentrated in the northwest of the country and Rh 29 Coatzacoalcos, with additional scattered basins without availability distributed throughout the national territory. Figure 12 displays the surface water availability projected for 2034, calculated using Turc’s method.

Figure 13 shows the surface water availability for 2034, calculated using the Rc method. Additionally, Figure 14 illustrates the comparative behavior of water availability between 2020 and the projected scenarios for 2034. These figures collectively provide a comprehensive comparison of the results obtained in this study with the official water availability data published by the Mexican government, as well as the projected scenarios based on the applied methodologies. Similarly, projections with the runoff coefficient method (Fig. 13) show that basins without availability will also be concentrated in the northwest of the country, though to a lesser extent. Additional affected areas include Rh 30 Grijalva-Usumacinta, 18 Balsas, and 35 Mapimí. With the Rc method, in the 2034 scenario, there are 103 basins without availability (23%), 334 basins (42%) with availability (but less than 100 Hm³) and 320 basins (35%) with more than 100 Hm³.

Figure 14 graphically contrasts the calculated results with the 2020 water availability data published by ^{Semarnat (2020)}, providing a clearer understanding of trends and variations. The basins without availability, as reported in the 2020 study, increase slightly when the estimated percentage changes derived from the Turc’s and Rc methods are applied. In both scenarios, the number of basins with availability of less than 100 Hm³ decreases, while the number of basins with availability exceeding 100 Hm³ increases. Nationwide, the total number of basins without availability rose significantly from 91 in 2020 to 154 when applying the percentage change from Turc’s method and to 103 when using the Rc method.

• ^{Semarnat (2020)}
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

  Secretaría de Medio Ambiente y Recursos Naturales, Mexico. Diario Oficial de la Federación, September 21, 2020
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

To identify potential sites for hydropower development, basins with surface water availability (D) exceeding 2 m³ s^-1 and slopes greater than 2% were considered. Additionally, downstream flow (AB), as defined in NOM-011-CONAGUA-2015, was utilized to pinpoint basins suitable for further analysis. This approach enables the identification of sites with hydroelectric potential, which can accommodate either large dams or smaller infrastructure, such as pipelines running parallel to the riverbeds. Figures 15 and 16 present the results of the calculations, employing each previously mentioned method, based on the scenery’s water availability for 2034.

Figure 17 offers a comprehensive visual representation of the intricate relationships between hydrological regions, projected water availability for 2034, and ecologically critical areas. To assess the potential impacts of water availability on sensitive ecosystems, this study incorporated existing designations of ecologically critical areas, including Protected National Areas (ANPs, for their acronym in Spanish), Ramsar wetlands, and Important Bird Conservation Areas (AICAs, for their acronym in Spanish). The AB 2034 analysis, utilizing Turc’s method (as depicted in Fig. 12), identified basins with sufficient water flow to meet future demands, while also revealing the highest number of basins projected to experience water scarcity by 2034. These results were then overlaid with a map of protected areas, encompassing NPAs, Ramsar wetlands, and AICAs, emphasizing the imperative need for sustainable infrastructure planning. This integrated approach, facilitated by the calculations and development shown in Figure 17, provides a crucial tool for understanding the intersection of water resources and ecological preservation.

• Infrastructure prioritization: Focusing on regions with projected water availability exceeding 2 m³ s^-1, especially in hydrological regions such as HR 12, HR 25, HR 26, HR 30, and HR 31.

• Protecting critical ecosystems: Avoiding or minimizing the impact of infrastructure projects on Ramsar wetlands, ANPs, and AICAs to safeguard biodiversity.

• Adaptive water management: Implementing strategies to address challenges posed by climate variability and anthropogenic pressures, ensuring resilience in water resource management.

Reconciling the potential and restrictions in HR 30: Apparent abundant water resources in Hydrological Region 30 (HR 30), encompassing the Grijalva-Usumacinta basin, coupled with the strict limitations on hydroelectric development along the Santo Domingo River, highlight the intricate equilibrium required in water resource management. This scenario vividly illustrates the dual mandates of energy generation and ecological preservation. While HR 30 exhibits the volumetric capacity to meet projected future water demands, it simultaneously represents one of Mexico’s most ecologically sensitive and significant areas. Consequently, a nuanced strategy that prioritizes both sustainable energy solutions and the protection of critical ecosystems, is essential. The proposed Santo Domingo hydroelectric project within HR 30 was halted by the Secretaría de Medio Ambiente y Recursos Naturales (Secretariat of Environment and Natural Resources, Semarnat) due to its potential ecological and social repercussions, demonstrating the government’s commitment to prioritizing ecological connectivity and biodiversity conservation over energy development. This decision embodies a broader strategic framework aimed at:

Ensuring ecological flows: Maintaining sufficient water flow to sustain local biodiversity and essential ecosystem services.

Mitigating social conflict: Respecting indigenous rights and addressing community opposition to large-scale infrastructure projects.

Upholding conservation commitments: Adhering to national and international legal frameworks that promote sustainable development.

By effectively balancing these imperatives, sustainable infrastructure planning can unlock the potential of regions like HR 30 while safeguarding Mexico’s vital ecosystems. This integrated approach fosters responsible resource management and ensures the ecological integrity of the nation’s hydrological basins.

In this work, the results of the analysis of historical precipitation and temperatures conducted by Ramírez-Villa et al. (2020) and their projection to 2034, based on a linear behavior, were utilized. The growth rates of water volume in agricultural, industrial, and urban public consumptive uses were analyzed. Using two equations (Turc’s and Rc), the volume per local basin (Cp) was estimated and applied in Eqs. 1 and 2, to calculate the available volume of surface water by 2034. Subsequently, the basins with the possibility of developing hydroelectric projects were identified.

R² and RMSE values obtained from the linear regression applied to historical precipitation and temperature data provide results with a low probability of occurrence through 2034; however, they can be considered a scenario or an average value. On the other hand, if these trends in rainfall and temperature behavior, as well as water use behavior, continue, surface water availability scenarios can be obtained through 2034.

The results of this work enable a first analysis of the surface water resources available in 2020 and their scenarios by 2034. For example, according to the ^{Semarnat (2020)} study, 321 953.36 Hm³ of water are generated in Mexico by natural runoff (Cp), and 192 022.94 Hm³ are extracted for consumptive uses (60%). There is sufficient water in the country, but the problem arises when the analysis is carried out by hydrological region. In 2020, the natural runoff in the north was 81 095.08 Hm³, with a Uc of 78 830.86 Hm³ (97.2%). In the rest of the regions (south), the natural runoff was 240 858.28 Hm³ and the Uc of 113 192.08 Hm³ (47%). That is, in the north, almost 100% of surface water is consumed for different purposes, resulting in severe water stress.

• ^{Semarnat (2020)}
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

  Secretaría de Medio Ambiente y Recursos Naturales, Mexico. Diario Oficial de la Federación, September 21, 2020
  Acuerdo por el que se actualiza la disponibilidad media anual de las aguas nacionales superficiales de las 757 cuencas hidrológicas que comprenden las 37 regiones hidrológicas en que se encuentra dividido los Estados Unidos Mexicanos

The northern regions experience severe water stress, characterized by high local basin (Cp) values and consumptive use (Uc). The substantial exploitation of available surface water across various sectors intensifies pressure on already scarce resources.

The southern regions, which face significant barriers to infrastructure development, exhibit lower water utilization rates. These challenges include irregular precipitation patterns, limited investment, environmental constraints, social resistance, and the recognition of indigenous rights. Furthermore, limited hydropower potential necessitates the exploration of alternative energy sources. Developing hydropower infrastructure presents a significant challenge, demanding a careful equilibrium between energy generation and ecological and social considerations. Southeastern basins, like the Usumacinta and Grijalva, exhibit substantial hydropower potential, yet they also face considerable hurdles.

Legal requirements for ecological flows, as demonstrated by the environmental flow decrees for the Usumacinta River, frequently clash with hydropower operational demands. Furthermore, indigenous rights and local opposition create additional complexities for the implementation of infrastructure. Achieving sustainable hydropower development requires a commitment to prioritizing biodiversity conservation, respecting indigenous rights, and ensuring equitable distribution of resources. This approach aims to harmonize energy needs with the preservation of vital ecosystems and the well-being of local communities. This paper underscores the urgent need for stronger regulatory interventions to combat illegal surface water extraction and restore ecological balance. Strict enforcement of ecological flow requirements in perennial channels is crucial for protecting ecosystems. The remaining flows should be allocated for consumptive uses, prioritizing human consumption and essential agriculture.

In conclusion, Mexico faces complex challenges in managing its water resources and meeting its energy demands. However, proactive and integrated measures that prioritize sustainability, equity, and resilience can overcome these obstacles. By embracing innovative solutions and fostering the responsible use of natural resources, Mexico can secure long-term environmental health and equitable access to resources for all its regions.