BLOG

WATER POLITICS

Via Science Direct, a new journal article on transboundary aquifer areas between Mexico and the U.S.:

• Previous article in issue
• Next article in issue

Keywords

Effective aquifer area

Groundwater management

Transboundary aquifer

1. Introduction

Transboundary water resources are becoming increasingly strategic as domestic reserves are getting exhausted (Biswas, 2011). Even more, groundwater resources, including those shared internationally, are estimated to provide around 40?% of worldwide drinking water needs (UN-Water, 2008). As surface water becomes less available, droughts become more prominent, and worldwide temperatures hit high record levels, shared groundwater resources are being used at unprecedented levels. The case of the United States (U.S.)–Mexico border is not an exception. The Rio Colorado and Rio Grande transboundary basins are experiencing the lowest international reservoir levels of the last one hundred years. This condition has put the border communities at an increasing risk facing a twofold percentage of projected population growth, increasing water demands for urban, industrial, and agricultural needs, and, at the mercy of the poorly regulated and mismanaged available transboundary groundwater resources.

Efforts to study and assess transboundary aquifers along the U.S.–Mexico border are very recent, data and research remain very limited, and binational formal attention has traditionally been negligible to promote a change in the status quo of current groundwater use in the border region (Carter et al., 2017, Callegary et al., 2018, Matherne and Megdal, 2023; Petersen-Perlman et al., 2021; Sanchez and Rodriguez, 2021).

In 2022, Sanchez & Rodriguez (2022), reported the first complete map of transboundary aquifers crossing the U.S.–Mexico border. A total of 73 hydrogeological units (HGUs) were identified, of which 28 were considered as transboundary aquifers. Previously, Sanchez et al. (2020) made the first attempt to identify priority areas (Effective Transboundary Aquifer Areas) within the boundaries of the HGUs across the State of Texas in the U.S., and Mexico. The Effective Transboundary Aquifer Area (ETAA) approach tries to delineate those “hot spots” where pumping zones are concentrated and therefore can be prioritized over the rest of the geographic extension. Ideally, it is generally recognized that transboundary groundwater management strategies should approach a unified and holistic system across the complete aquifer; however, this is not just often challenging as data sharing and harmonization between different countries is usually politically sensitive, but also because it might be impractical and rather unfeasible for decision-making process considering differences in legal and governances’ systems across the aquifer. Additionally, groundwater flow is usually slow and fragmented into various paths, often only some parts of TBAs are relevant for controlling cross-border groundwater impacts (Maass-Morales et al., 2024).

This type of assessment where a more refined scale of analysis is proposed in lieu of the complete aquifer system is very limited in the literature but seems to be getting more attention as transboundary groundwater resources management become more strategic and needed. The use of Transboundary Aquifer Management Zones proposed by Maass-Morales et al. (2024), zoning by Rivera et al. (2023) and Pétré et al. (2021), hot spots by Fraser et al. (2020), Effective Transboundary Aquifer Areas (ETAAs) by Sanchez et al. (2020), or “border strip” earlier by Kettelhut et al. (2010), are approaches designed to concentrate the prioritization process in smaller zones rather than whole aquifer areas. These different assessments try to identify those priority zones based on different criteria mostly related to groundwater development, flow direction, and potential impacts at transboundary scale (groundwater flows, overexploitation, or contamination). This paper attempts to build on the current framework of identification of priority areas by adding two new elements that were not considered in Sanchez et al. (2020) nor addressed in previous studies. First, instead of using a pattern of wells (visually driven) to identify an “effective pumping area” as it was first reported in Sanchez et al. (2020), we utilize well density data, which is a quantifiable approach and provides comparable capabilities among ETAAs. Second, we use depth contours, that combined with geological information, can provide additional insights of groundwater flows and consequently potential impacts at transboundary scale. This second element was somehow attempted by Kettelhut et al. (2010), by estimating the width of the strip of land of any border that incorporates the interactions of most of the physical, demographical, environmental, social, and institutional aspects of the transboundary aquifer. Though this approach indeed tries to identify those priority hot spots (the size of the strip of land), it does so by totally ignoring the limits of the aquifer as a whole or the boundaries of the system (geological boundaries), therefore potentially limiting the effective involvement of stakeholders in the area of attention that could be affected at different timeframes by different geological formations from which wells can be extracting. Additionally, the fact that there is no specific case study where this methodology has been applied, complicates its assessment and applicability to the broader possibilities of aquifers where data is usually limited (Maass-Morales et al., 2024; Eckstein and Eckstein, 2024).

The ETAA approach was developed as a process that considers the specific study case and conditions of the aquifers located in the U.S-Mexico border. We hypothesized that in this region, there is a correlation between higher density of wells and shallow depths, and therefore, the potential to have an impact at a local scale is higher than at a regional scale. In addition, the intensity of density across ETAAs varies as well, therefore, providing an extra layer of differentiation among different ETAAs that was not considered before in Sanchez et al. (2020). This new information provides more granularity on the analysis both at regional and local scales in terms of vulnerability to overexploitation and potential contamination of aquifers, which eventually can have an impact on border communities that depend heavily on groundwater to fulfill their social and economic needs.

Though the objective of the ETAAs does not pertain to measure potential impacts of groundwater flow in terms of water quantity or quality at transboundary level, the ETAA approach does identify a more refined area of study where these potential issues may arise with hydrological implications that derive into socio-economic consequences. For the purpose of this research, the concept of vulnerability pertains to how much the region could be affected by significant or major ETAAs (above 99 wells per 100?km²) combined with its level of groundwater dependency, as well as potential impacts of groundwater flows at transboundary scale. An ETAA located in a highly groundwater dependable region, such as the border cities of Arizona or New Mexico, are considered to have greater levels of vulnerability than those reported in the Lower Rio Grande Valley, for example.

The objectives of this paper are: First, to improve the methodology developed by Sanchez et al. (2020) on the HGUs between Mexico and Texas, U.S., and extend it to the rest of the HGUs identified between the States of California, Arizona, and New Mexico on the U.S. side and Baja California, Sonora, and Western Chihuahua on the Mexico side. We expanded the spatial analysis and its potential impacts at a local scale using density of wells and the available well depth data to portray contours of areas with similar depth as a proxy to potential groundwater flows (shallow, intermediate, or deep layers). And second, to identify those regions across the U.S border that represent the most vulnerable zones in terms of groundwater development and dependency.

In summary, results indicate that New Mexico and Chihuahua show the most dense and vulnerable ETAAs across the border region, followed by Arizona and Sonora, California and Baja California, and finally, Texas and Mexico. However, Texas and Mexico show the most consistent distribution of lower-density ETAAs covering around 60?% of the of the 40 HGUs. As we estimated, in the case of transboundary aquifers between Mexico and the U.S., there seems to be a correlation between high well density and shallow depths (e.g., Conejos-Medanos Aq./Mesilla Bolson) and therefore, the potential to have an impact at local scale is higher than at a regional scale. The evaluation of the magnitude of those impacts is yet to be studied, but the transboundary nature of those impacts is depicted in this paper.

2. Materials and methods

2.1. Transboundary hydrogeological units (HGUs)

The identification of ETAAs is performed over the geological extensions of the HGUs across the States of Chihuahua (CH), Sonora (SO), and Baja California (BC) in Mexico, and New Mexico (NM), Arizona (AZ), and California (CA) in the U.S., as reported by Sanchez and Rodriguez (2021). The ETAAs approach is primarily based on the understanding that, due to the geological heterogeneity and changes in lateral facies, not the entire aquifer area is exploitable. Therefore, the location and depth of all available active wells within the boundaries of the HGUs are used to identify an aquifer’s exploitable, productive, hot spot or pumping area, to provide alternative criteria to prioritize vulnerable or impacted regions with a more refined area of attention, particularly at the transboundary level. Well data used in this study provides the most robust field information on groundwater extraction or production for this analysis, as opposed to other proxy data that could be obtained using remote sensing techniques.

The study attempts to expand and improve the methodology used by Sanchez et al. (2020) by identifying both the density within the active pumping areas of each HGU and depth information as a proxy to confirm the well density patterns and potential source of geological unit. Well depth provides insights into the potential age of groundwater and therefore, flow systems at shallow versus deeper layers. Groundwater ages (time since recharge) and velocities are youngest and fastest in local shallower flow systems and are older and slower in intermediate to deeper regional systems (Alley et al., 1999). The boundaries of 40 HGUs (Fig. 1, Table 1) reported by Sanchez and Rodriguez (2021) were compared and analyzed using the updated ETAA approach. This effort was also performed on the previously reported HGUs between Mexico and Texas, U.S. for mapping purposes only to portray the complete map of ETAAs at the border using the same updated methodology.

1. Download: Download high-res image (533KB)
2. Download: Download full-size image

Fig. 1. Hydrogeological Units (HGUs) between Baja California, Sonora, and Western Chihuahua in Mexico and California, Arizona, and New Mexico in the U.S. Adapted from Sanchez and Rodriguez (2021).

Table 1. Hydrogeological Units (HGU) between Baja California, Sonora, and Western Chihuahua in Mexico and California, Arizona, and New Mexico in the U.S shown in Fig. 1.

HGU	Rock Type	HGU	Rock Type
1. Tijuana-San Diego Aq.	Alluvium	21. Baboquivari Mountains	Conglomerate
2. Tecate Aq.-Potrero Valley	Alluvium	22. Arroyo Seco Aq.	Gravel
3. La Rumorosa-Tecate Aq./ Jacumba Valley	Basalt	23. Rio Altar Aq.	Gravel
4. Laguna Salada Aq./ Coyote Wells Valley	Alluvium	24. Pajarito Mountains	Dacite
5. Valle de Mexicali-San Luis Rio Colorado/ Yuma-Imperial Valley	Alluvium	25. Nogales-Rio Santa Cruz Aq./Upper Santa Cruz Basin	Granodiorite
6. Tinajas Atlas Mountains	Volcanic Rock (Aphanitic)	26. Elenita_Huachuca Basin	Conglomerate
7. Puente Cuates Valley/ Lechuguilla Desert	Lava Flow	27. Rio San Pedro Aq./Upper San Pedro Basin	Sandstone
8. Cabeza Prieta Mountains	Alluvium	28. Mule Mountains	Limestone
9. Los Vidrios Aq.	Sedimentary	29. Rio Agua Prieta Aq./Douglas Basin	Alkaline Basalt
10. Sonoyta-Puerto Peñasco Aq.	Basalt	30. Perilla Mountains	Gravel
11. Agua Dulce Mountains	Gravel	31. Arroyo San Bernardino Aq./ San Bernardino Valley	Sandstone
12. Cerro Colorado Numero 3 Valley	Andesite	32. Guadalupe Mountains	Conglomerate
13. Quitobaquito Hills	Granite	33. Animas Basin	Conglomerate
14. La Abra Plain	Gravel	34. Janos Aq./Playas Basin	Dune Sand
15. Senita Basin	Sandstone	35. Alamo Hueco Mountains	Gabbro
16. Lukeville-Sonoyta Valley	Granodiorite	36. Ascension Aq./Hachita-Moscos Basin	Sandstone
17. Sierra de Santa Rosa – La Nariz	Granite	37. Josefa Ortiz de Dominguez Aq.	Basalt
18. The Great Plain	Granodiorite	38. Las Palmas Aq./Mimbres Basin	Dune Sand
19. Los Chirriones Aq.	Granite	39. Potrillo Mountains	Granite
20. San Simon Wash	Dacite	40. Conejos Menados Aq./Mesilla Bolson	Gravel

The use of the term “effective” as used by Sanchez et al. (2020) refers to the location of an area or areas within the boundary of each HGU that reports a pattern of water wells as an indicator of groundwater productivity that can be differentiated from the rest of the geographical extent of the HGU. For this research, the identified ETAAs that cross the international border are the priority of the analysis, however, there are also ETAAs that do not seem to be crossing the border or located at greater distances from the border that provides additional information to the overall picture of the pattern of groundwater development in a specific region.

2.2. Data sources

The data on water wells in Mexico were obtained from Registro Publico de Derechos de Agua (REPDA, 2023). Geological formation data was obtained from the Servicio Geológico Mexicano database (SGM, 2023). Information on wells in the State of New Mexico was obtained from the point of diversion (POD) dataset (NMOSE, 2023), in the State of Arizona, from the Arizona Groundwater Site Inventory (GWSI) (ADWR, 2023), and in the State of California, from the Groundwater Ambient Monitoring and Assessment (GAMA) online groundwater information system (SWCB, 2023). Data sources of the State of Texas are the same as those reported by Sanchez et al. (2020) and are only adapted in this analysis for mapping of ETAAs and visualization purposes of the complete border map.

2.3. Data processing

Due to the different data sources, including heterogeneity in units, different geographical coordinates, and lack of information in a variety of data fields, it was necessary to reconcile all datasets to be able to integrate the information and generate the visualization maps for the ETAAs.

To harmonize various types of water uses in the integrated dataset, water uses were classified into six categories: agriculture (includes livestock, irrigation, and animal feeding operations), public water supply (all supplies for municipal or residential use), industrial (industry and commercial activities), power, mining, and others (recreation, construction, etc.). A seventh category “Unknown” was assigned to wells with no water use information. Water use information was available for most of the wells located in the U.S., but limited information was available for wells in Mexico. Wells used for oil and gas activities (except for rig wells) and monitoring wells were discarded. Wells were also screened based on their status (active or inactive). Only active wells were considered as part of this analysis. Data on the status of wells in Mexico was not available and all wells were assumed to be active. The spatial distribution of wells used in this study and their attributes can be found in Fig. S1 and Table S1 in the supplementary materials.

2.4. Delineation of ETAAs

In this study, ETAAs within the 40 HGUs were identified based on the spatial density of the 43,413 active wells. The rationale of using well density as the basis for ETAA delineation was that if a certain number of active wells were found in a unit area, this unit area was then considered to be actively exploited, and several of such spatially contiguous unit areas together were considered an ETAA.

Using tools embedded in ArcGIS Pro, ETAAs were delineated within four groups of HGUs independently, that is, the same geospatial analysis was conducted separately for each HGU group. Group one (G1) integrates HGUs 1 through 5 primarily located in California and Baja California, and HGU 5 partly located in Arizona and Sonora. Group two (G2) consists of HGUs 6 through 18 located between southwestern Arizona and Sonora. Group three (G3) integrates HGUs 19 through 31 located across south-eastern Arizona and Sonora, and Group four (G4) includes HGUs 32 through 40 primarily located between New Mexico and Chihuahua, and HGU 32 partially located in the Arizona-Sonora region. The geospatial analysis applied for the ETAA delineation is described below.

The following steps were followed to reasonably corroborate the visualization inspections of water well concentration areas.

Select active wells located within G1, G2, G3, or G4 using the Select by Location tool. Use the wells selected as the input in the Point Density tool. To delineate realistic ETAAs, we had to determine a few parameter values regarding the neighborhood through trial and error. Parameter values determined through trial and error included the size of the output cell (i.e., unit area), the shape of the neighborhood, and the search radius around each cell that is used to calculate the density value. Based on the available well location data, we eventually set the cell size of the output density map as one km by one km, the shape of the neighborhood square, the search radius around each cell as five km, and the unit area to calculate density as one km².

The cell value in a resulting density map represented the number of wells per km², which was, in large, a decimal value. For easier interpretation, the cell value (i.e., number of wells per km²) was converted to number of wells per 100?km² so that most cell values were integers. Then, define and calibrate ETAAs based on cell values within HGUs, i.e., estimated number of wells per 100?km².

• a.
  For HGUs in G1 and G2, ETAAs included cells from the density map that had 10 (ten) or more wells per 100?km².
• b.
  For HGUs in G3 and G4, ETAAs were delineated by selecting the cells of the density map that had 20 or more wells per 100?km².
• c.
  For HGUs located between Texas and Easter Chihuahua, Coahuila, Nuevo Leon, and Tamaulipas, the methodology performed by Sanchez et al. (2020) was updated to this refined approach and the ETAAs were delineated by selecting the cells of the density map that had more than 10 wells per 100?km².

The ETAAs composed of selected cells were dissolved first using the Dissolve tool. This tool was applied to aggregate selected one km² cells into larger polygons. The output feature was cleaned up manually to remove polygons that were only one km² in area, i.e., polygons consisted of only one qualified unit area that was not adjacent to any other qualified unit areas. That said, ETAAs delineated in this study contained two or more contiguous unit areas.

The output polygons were then smoothed using the Smooth Polygon tool. This tool simplifies polygons by removing relatively extraneous vertices while preserving the essential shape of the polygons. For the present dataset, the default simplification algorithm, retain critical points (Douglas-Peucker), was applied, and simplification tolerance was set as five km. This tolerance was determined to accurately represent areas of the input ETAAs and effectively smooth the outlines of the input ETAAs, so they were not as jagged as the edges of the square cells.

In addition, for each HGU we also estimated a smooth surface of well depths based on depth records using the Diffusion Interpolation with Barriers tool. After creating spatially continuous well-depth images, well-depth contour lines were generated by joining locations (pixels) with the same estimated well depth using the Contour function. The based contour value was set at one meter and the contour interval was set at 50?m. To refine the level of analysis and prioritization of ETAAs, hot spots were divided into Major ETAAs and Minor ETAAs based on well density. Where well densities exceeded the 50th percentile (median) well density value equivalent to 99 wells per 100?km² (about one well per km²) were categorized as major, while those below the median value were considered minor. For the purpose of this research and due to limited well information, we assumed that all wells considered in this paper penetrate the complete aquifer thickness. Therefore, there are no differences accounted for this parameter in our methods design.

The interested reader in more technical details can revise other studies such as Johnston et al. (2001) and Silverman (1986), for more information regarding the spatial analysis tools used in this study.

3. Results

Figs. 2 through 7 show the HGUs from California to New Mexico in the U.S., and from Baja California to western Chihuahua in Mexico. Each figure shows the well-density pattern in each formation within and across the different HGUs represented as the darkest brown colors. Major ETAAs (well density above 99 wells per 100?km²) are delineated to differentiate them from minor ETAAs (density below 99 wells per 100?km²). Fig. 3, Fig. 5, Fig. 7 additionally show the well depth contouring across the HGUs. The purpose of these figures is to show four important elements: different water well concentrations (density) within the HGUs; well depth across the HGU as a proxy of potential shallow (local) or deeper (regional) groundwater flow, and finally the delineated major ETAAs and minor (non-delineated) ETAAs. Some of them are closer to the border than others, however, we consider them all as part of the ETAAs identification process. The analysis of the paper concentrates on major ETAAs, additional information on minor ETAAs can be found in supplementary material. Table 2 and Table S1 show a compilation of the results described below by Group and HGU. It includes the total area of each HGU compared to the ETAAs’ area, number of wells (total and within the ETAAs), Major ETAAs, depth of wells (range), geological features and primary groundwater use. Both tables also include the criteria used by Sanchez et al. (2021) to account for population categories to support the analysis of groundwater vulnerability and dependability. Fig. 8, Fig. 9 show the complete maps of all the HGUs on the border (including Texas) showing Major and Minor ETAAs. Major ETAAs are indicated with their corresponding identifiers.

1. Download: Download high-res image (547KB)
2. Download: Download full-size image

Fig. 2. Effective Transboundary Aquifer Areas (ETAAs) between California, U.S., and Baja California, Mexico. Additional properties of major ETAAs are listed in Table 2.

1. Download: Download high-res image (597KB)
2. Download: Download full-size image

Fig. 3. ETAAs with depth contours in HGUs located across California, the U.S., and Baja California, Mexico.

1. Download: Download high-res image (717KB)
2. Download: Download full-size image

Fig. 4. Effective Transboundary Aquifer Areas (ETAAs) in Eastern Arizona, U.S., and Sonora, Mexico. Additional properties of major ETAAs are listed in Table 2.

1. Download: Download high-res image (962KB)
2. Download: Download full-size image

Fig. 5. ETAAs and depth contours in HGUs located in Eastern Arizona, U.S., and Sonora, Mexico.

1. Download: Download high-res image (639KB)
2. Download: Download full-size image

Fig. 6. Effective Transboundary Aquifer Areas (ETAAs) in New Mexico, U.S., and Chihuahua, Mexico. Additional properties of major ETAAs are described in Table 2.

1. Download: Download high-res image (860KB)
2. Download: Download full-size image

Fig. 7. ETAA and depth contours in HGUs located in New Mexico, U.S., and Chihuahua, Mexico.

Table 2. Properties of HGUs and Major ETAAs between Baja California, Sonora, Chihuahua, Coahuila, Nuevo Leon, and Tamaulipas in Mexico, and California, Arizona, New Mexico, and Texas in the U.S.

HGUa ID	HGU/Aquifer	Geologic ageb	Dominant Water usesc	# wells in HGU, average depth	HGU area (km2)	HGU Pop classd	ETAAe ID	# of wells in ETAA	Average well depth in ETAA (m)	ETAA area (km2)	# of wells per 100?km2
GROUP 1 ETAAs between California, U.S., and Baja California, MX
1	Tijuana– San Diego Aq.	Late Cretaceous	PUB	461, 118?m	749	3	e01_1	428	115.23	412.07	104
4	Laguna Salada Aq./ Coyote Wells Valley	Quaternary	PUB	182, 62?m	3814	2	e04_1	118	82.93	119.51	99
5	Valle de Mexicali–San Luis Rio Colorado Aq./ Yuma–Imperial Valley	Quaternary	PUB/AG	3148, 63?m	28,023	3	e05_1	2058	62.91	1773.23	116
GROUP 2 ETAAs between Arizona, U.S., and Sonora, MX
22	Arroyo Seco Aq.	Late Cretaceous to Early Tertiary	PUB	1059, 173?m	5589	1	e22_1	120	166.94	96	125
							e22_2	126	162.85	80	158
25	Nogales–Rio Santa Cruz aq./Upper Santa Cruz Basin	Middle Proterozoic	PUB	4921, 110?m	7442	3	e25_1	2619	119.10	1075.00	244
							e25_2	235	54.30	136.00	173
							e25_3	214	41.02	124.00	173
							e25_4	122	109.53	54.00	226
27	Rio San Pedro Aq./Upper San Pedro Basin	Late Pliocene to Early Pleistocene	PUB	2014, 105?m	5350	2	e27_1	609	105.74	271.00	225
							e27_2	166	83.54	91.00	182
							e27_3	108	88.76	73.00	148
29	Rio Agua Prieta Aq./Douglas Basin	Paleozoic	AGR	1504, 118?m	2064	2	e29_1	970	119.83	529.00	183
							e29_2	77	78.33	60.00	128
GROUP 3 ETAAs between New Mexico, U.S., and Chihuahua, MX.
33	Animas Basin	Quaternary	AGR	1674, 115?m	4871	1	e33_1	416	101.91	155.52	267
							e33_2	304	115.58	154.77	196
							e33_3	144	102.82	55.43	260
							e33_4	75	189.67	50.25	149
34	Janos Aq./Playas Basin	Quaternary	AGR	1594, 46?m	4012	1	e34_1	577	25.96	273.26	211
							e34_2	282	50.40	157.17	179
36	Ascension Aq./Hachita Moscos Basin	Quaternary	PUB	2325, 75?m	4560	1	e36_1	862	93.60	222.98	387
							e36_2	327	73.36	148.53	220
							e36_3	298	–	137.33	217
							e36_4	165	40.27	81.75	202
38	Las Palmas Aq./Mimbres Basin	Quaternary	PUB	7432, 89?m	11,215	1	e38_1	4751	80.45	1127.75	421
							e38_2	390	119.18	243.87	160
							e38_3	660	130.63	167.41	394
40	Conejos Menados Aq./Mesilla Bolson	Quaternary	PUB	15097, 58?m	9016	3	e40_1	14246	56.93	1172.13	1215
							e40_2	159	23.95	74.63	213
GROUP 4 ETAAs between Texas, U.S. and Chihuahua, Coahuila, Nuevo Leon, Tamaulipas, MX.
41	Valle de Juarez Bolson/Hueco-Tularosa Bolson	Pleistocene-Holocene	AGR	1546, 102	11,852	3	e41_1	1090	87	897	122
46	Presidio Bolson	Pleistocene-Holocene	AGR	352, 36	2469	1	e46_1	33	12	31	107
54	Edwards Aq.	Albian	AGR	11,446, 190	98,742	3	e54_1	49	119	30	161
60	Allende-Piedras Negras Aq.	Pleistocene-Holocene	AGR	775, 37	9109	2	e60_1	297	41	294	101
							e60_2	50	11	30	169
							e60_3	21	21	17	121
							e60_4	20	30	13	156
67	Bigford Fm	Eocene	AGR	627, 150	4660	1	e67_1	54	136	47	114
68	El Pico Clay Fm	Eocene	AGR	593, 288	7395	3	e68_1	121	181	73	167
72	BRB/Gulf Coast Aq.	Pleistocene-Holocene	PUB	6370, 86	60,652	3	e72_1	178	48	128	139
							e72_2	141	38	109	129
							e72_3	58	47	53	109
							e72_4	23	51	22	103
							e72_5	12	209	11	113
							e72_6	11	85	10	108

a

HGU = Hydrogeological Unit.

b

Geologic ages were adopted from Sanchez and Rodriguez (2021)

c

Water uses: AGR = Agriculture (livestock and irrigation uses); PUB = Public (municipal and domestic uses); IND = Industrial; MIN = Mining; POW = Power; OTHER = Other uses; UNKNOWN = Use not defined

d

Criteria for population classification (adapted from Sanchez et al., 2020). 3: > 400,000; 2: 100,000 – 400,000; 1: <100,000.

e

ETAA = Effective Transboundary Aquifer Area with well densities above 99 wells per 100?km².

1. Download: Download high-res image (451KB)
2. Download: Download full-size image

Fig. 8. Major and minor ETAAs between Mexico and California, Arizona, and New Mexico in the U.S. Properties for major ETAAs are described in Table 2. Additional information on ETAAs, including minor ETAAs is included in the supplementary Table S-1.

1. Download: Download high-res image (528KB)
2. Download: Download full-size image

Fig. 9. Major and minor ETAAs between Mexico and the State of Texas, U.S. Properties for major ETAAs are described in Table 2. Additional information on ETAAs, including minor ETAAs is included in the supplementary Table S-1.

3.1. ETAAs between California, U.S., and Baja California, Mexico (Group 1)

Fig. 2, Fig. 3 show the HGUs between California and Baja California (G1). From Fig. 2, a couple of ETAAs have been identified. A major ETAA can be seen in the Tijuana-San Diego transboundary aquifer (HGU 1) which encompasses more than half of the complete boundary of the HGU (Fig. 2). Therefore, there is not a significant ETAA that would make any differences in the prioritization or identification of hot spots in this aquifer apart from the current physical boundaries of the system. However, it highlights the priority of this aquifer over the rest of the HGUs in G1. It concentrates a population of around five million people between the San Diego metroplex and Tijuana areas, and it is currently under pressure from surface water reductions imposed by the Drought Contingency Plan of the Colorado River and its corresponding Minute 323 of the 1944 Water Treaty (IBWC, 2017; Sheikh and Stern, 2019). Fig. 3 shows the water depth contours of the ETAA with an average depth of 100?m (m) with some areas at 50?m close to the border. This can be an indicator of potential shallow and local groundwater movement, that in fact was already reported by Chavez-Guillen & Klein (2017). The expectations of more groundwater reliance in a region with an already vulnerable groundwater system will likely increase. Therefore, the potential impacts associated with over-drafting and contamination at the transboundary scale will also increase.

The second major ETAA in G1 is located in the Laguna Salada Aq./Coyote Wells Valley (HGU 4) (Fig. 2, Fig. 3). This HGU is also considered a transboundary aquifer, but according to previous assessments (Sanchez & Rodriguez, 2022), it might be just partially connected to the Mexico side only through its Quaternary deposits, which is precisely where one major ETAA is located. This ETAA is close to the border in the U.S., encompassing only 3?% of the total area of the complete HGU. It reports an average well depth range of 50–100?m (Fig. 3), which can be a condition of potential shallow interactions.

The third and last major ETAA identified in G1 is located in the Valle de Mexicali-San Luis Rio Colorado Aq./Yuma-Imperial Valley transboundary aquifer (HGU 5). This ETAA is located across the boundaries of four states: California, Arizona, Baja California, and Sonora where one of the most important agricultural regions of both countries is located. It covers 6?% of the total HGU area, concentrating over 2000 wells with an average well depth of 62?m mostly located adjacent to the path of the Colorado River on its way to the Sea of Cortez (see Table 2 for details). Previous studies have demonstrated the strong relationship of surface and groundwater linkages of this aquifer and the river flows, as well as local transboundary groundwater flows at a very shallow level. Over-extraction for irrigated agriculture on both sides of the border, as well as salinity concerns, have been reported over the last 15 years (CONAGUA, 2009, Hathaway, 2011, Mumme, 2020, Sanchez et al., 2016).

Within Group 1, the most important ETAAs are located at the most important urban centers across the border of Tijuana-San Diego, and around Imperial Valley and Mexicali Valley. Even though these urban centers still rely heavily on surface water, the current surface water shortages and federal regulatory systems that establish reductions on water allocations from the Colorado river to both sides of the border will likely increase groundwater dependency for agriculture production.

3.2. ETAAs between Arizona, U.S., and Sonora, Mexico (Group 2)

In Group 2 (G2), between Arizona and Sonora, major ETAAs have been identified in four HGUs along and across the border (Fig. 4, Fig. 5, Table 2). The first one is the Nogales-Rio Santa Cruz Aq./Upper Santa Cruz Basin (HGU 25), which shows a clear and extended pattern with major ETAAs located from the north part around the city of Tucson to the southern part of the unit covering over 18?% of the total area of the HGU. The map shows a variety of major well densities, the one north of Tucson being the densest ETAA (Fig. 4). Followed by two more ETAAs located in the south closer to the border and potentially expanding into Mexico around Nogales sister cities and in alignment with the natural flow of the river. This pattern also seems to extend to the northwest towards the Arroyo Seco aquifer (HGU 22) where it connects to other two major ETAAs identified in the north part of this HGU, following the course of the Santa Cruz River. To the East, going around the Cienega Creek Basin (HGU 26), the pattern continues towards the upper part of the Rio San Pedro Aq./Upper San Pedro Basin (HGU 27) which also reports three major ETAAs that cover around 10?% of the aquifer area. The first ETAA is in the north part of HGU 27 and reports the highest density (225 wells per 100?km²) and two more ETAAs very close to the border with similar geographical extensions and well densities close to the City of Naco, Sonora (see Fig. 4 and Table 2). From Fig. 4, though major ETAAs have been clearly spotted, it is also clear that the concentration patterns across the HGUs represent more than half of the geographical extensions of HGUs 25 and 27. On these HGUs, the pumping densities seem to be concentrated in the Upper parts of both aquifers but clearly following a southern pattern across the border where it intensifies again. Well-depth of the identified ETAAs in the Nogales-Rio Santa Cruz Aq. /Upper Santa Cruz Basin ranges from 50 to 120?m, with an average of 80?m, and slightly shallower well depth close to be border around the sister cities of Nogales (Table 2). Groundwater flows along and across the aquifer seem to be controlled by the early layers of Quaternary Alluvium and Conglomerate deposits which signal local flows. The ETAAs located in the San Pedro Aq./Upper San Pedro Basin report a well depth average of 92?m with an important concentration of wells across the border near the cities of Cananea and Naco (Fig. 5).

The transboundary nature of these two aquifers as well as their high level of vulnerability considering their almost complete dependency on groundwater for all uses in the region has been reported before by Callegary et al. (2018), Carter et al. (2017), and Tapia-Villaseñor & Megdal (2021).

Another two major contiguous ETAAs are located in the Rio Agua Prieta Aq./Douglas Basin (HGU 29) (Fig. 4, Fig. 5). An intensified density pattern can be seen in the northern part of the aquifer, showing a clear trend along the Arroyo Pinto/Whitewater River and across the border to the Mexico side, where it seems to gain stronger density again around the cities of Douglas, Arizona and Agua Prieta, Sonora. The well depth average is 78?m across the border, and greater depth (119?m) towards the northern section of the ETAA (Fig. 5). The shallower depths of wells governing the density pattern in these ETAAs that practically cover over 35?% of the aquifer area, presume the possibility of significant groundwater flows or groundwater fluxes in the upper layers at a local scale and therefore can potentially have an impact on groundwater levels and/or contamination from either side of the border. The vulnerability of the system also relates to the high dependency of the population that relies on groundwater from this aquifer. It is the most important source of water in this region providing 75?% water for agriculture on the U.S. side and 81?% for public water supply on the Mexico side (Sanchez et al., 2016).

3.3. ETAAs between New Mexico and Chihuahua (Group 3)

The highest well-density areas on the border between Mexico and the U.S. are reported across the HGUs between New Mexico and Chihuahua (Fig. 6, Fig. 7). For example, four major ETAAs are identified in the northern portion of the Animas basin aquifer (HGU 33) around the cities of Lordsburg, Cotton City, and Animas, with a well density that ranges from 196 wells per 100?km² in an area of 154?km², to 260 wells per 100?km² in an area of 55?km² (Table 2 and Fig. 6, Fig. 7). This implies that a very localized and small geographic area is concentrating a significant number of wells as compared to the rest of the HGU or even to the rest of major ETAAs identified in Groups 1 and 2 (except for HGU 25 which reported a major ETAA of 226 wells per 100?km² in an area of 54?km²).

Towards the East, the next transboundary aquifer that reports major ETAAs is the Janos Aq./Playas Basin (HGU 34) which shows two ETAAs in the southern portion of the aquifer on the Mexico side with a well density ranging from around 200 wells per 100?km² and an average well depth of around 35?m (Fig. 7). The rather shallow depth of the wells estimates some potential local groundwater flow in this part of the HGU. As groundwater development increases around the cities of Janos and El Valle, mostly for agriculture development, groundwater flow could potentially have an impact at a transboundary level in the future. Over-drafting has been already documented on the Mexican side since 2002 due to the intensification of agricultural activity (Sanchez et al., 2016).

Other significant ETAAs with high density but comparable small geographical extensions are identified in the Ascencion Aq./Hachita Moscos Basin transboundary aquifer (HGU 36). Four major ETAAs are located predominantly on the Mexico side of the aquifer around the cities of Ascension and Las Palmas. Groundwater is used primarily for irrigated agriculture and some for rangeland. Of the four different high-dense ETAAs identified (Fig. 6, Fig. 7), two are located very close to the border with an average of 210 wells per 100?km² over a geographical extent of 115?km² in average. The average well depth ranges from 73 to 40?m (Fig. 7 and Table 2). The largest ETAA is located towards the south which reports a well density of 387 wells in an area of 222?km² (Table 2). According to the literature, this aquifer is a closed transboundary system where groundwater is only recharged by groundwater flow coming from the New Mexico side that discharges into the Moscos Laguna on the Mexico side where it evaporates rapidly (CONAGUA, 2009). The connectivity of the groundwater system across the border and its vulnerability as the main source of water for this region highlights the need for attention to future shared groundwater development (Sanchez et al., 2018).

The third transboundary aquifer that shows three important ETAAs is Las Palmas Aq./Mimbres Basin (HGU 38). One important hot spot is localized around the city of Deming in the center part of the aquifer on the New Mexico side, and another one on the extreme north part of the aquifer around Silver City. Another ETAA is identified across the border around Columbus, New Mexico, and Puerto Palomas, Chihuahua, with a density of 160 wells per 100?km² at a well depth of 119?m covering an area of 243?km² (Table 2). Previous studies have already reported cones of depression around this transboundary area as well as a reversal of groundwater flow towards New Mexico due to over-pumping. Additionally, water quality issues have been reported on both sides of the border (salinity and Total Dissolved Solids mostly) due to mining activities in the region (Hawley et al., 2000, Sanchez et al., 2016).

The last two major ETAAs identified in G3 are the Conejos-Medanos Aq./Mesilla Bolson (HGU 40). Though an ETAA in the Mesilla region was reported already by Sanchez et al. (2020), this study reassesses the ETAAs approach using our current methodology which refined the visualization and the robustness of our conclusions. We can now say that there is an ETAA with 1215 wells per 100?km² with an average depth of 56?m. This is the highest density ETAA reported so far in the border with a rather shallow depth of 56?m covering an area of 1172?km² located around Las Cruces all the way to the border with Mexico (Fig. 7 and Table 2). The transboundary nature of this aquifer, as well as the strong connectivity of groundwater flows contribution to river baseflow and the contamination challenges related to increasing salinity levels from the intense irrigated agriculture mainly on the U.S. side, are well known in the literature (Robertson et al., 2022). This ETAA covers a localized area of 13?% of the total extension of the HGU on the New Mexico side mainly following the flow of the Rio Grande. So far, it represents the most dense and sensitive ETAA in the border.

3.4. ETAAs between Texas, U.S., and Mexico (Group 4)

Even though an ETAA analysis between Texas and Mexico has been already reported by Sanchez et al. (2020), we considered it useful to reapply this refined methodology in HGUs covering the state of Texas in the U.S. and the states of Eastern Chihuahua, Coahuila, Nuevo Leon, and Tamaulipas in Mexico for two main reasons: to compare previous and current results and to refine previous assessments. Fig. 9 shows both minor and major ETAAs in this Group (4) for visualization purposes. It also shows a consistent pattern of well density across the HGUs controlled mainly by minor ETAAs (10–99 wells per 100?km²). This pattern is very similar to what Sanchez et al. reported back in 2020 which validates the approach previously used. The former publication, however, did not differentiate between minor and major density concentrations. Fig. 9 shows a variety of major ETAAs (16 total) with rather small geographical extensions and with an average density of 127 wells per 100?km². Except for the major ETAA reported in HGU 41 Valle de Juarez Bolson/Hueco-Tularosa Bolson which has an area of 897?km², and another ETAA in the Allende-Piedras Negras aquifer (HGU 60) with an area of 294?km², the rest of the major ETAAs report an average area covering barely 40?km^2, with some even reporting 10?km² (Table S-1). Additionally, apart from HGUs 41 and 60, the rest of the HGUs show the smallest area coverage of major ETAAs in the complete border region. HGUs 46 and 54 report one ETAA covering an area of around 30?km² each. Similarly, HGUs 67 and 68 report one ETAA each covering an average area of 60?km². Contrastingly and significantly, Group 4 shows the largest geographical extensions of minor ETAAs on the border (see Table S-1 for details on Minor ETAAs). Another important finding comparable to its antecessor publication is that Bigford Fm. (HGU 67) and El Pico Clay Fm. (HGU 68) are not reported as transboundary aquifers before due to their geohydrological and water quality conditions (Sanchez et al., 2018). However, they both report major ETAAs in the northern part (Texas side) of the HGUs precisely where Sanchez et al. (2018) identified small areas of good aquifer potential. Lastly, in terms of depth of wells, apart from major ETAAs identified in HGUs 67, 68, 54 and one in HGU 72 that report well depth above 100?m, the remaining major ETAAs report an average shallow depth of 42?m (Table 2). Shallow and local groundwater fluxes close to the Rio Grande with potentially local impacts at transboundary scale, have been already reported in the case of the Edwards Aq (HGU 54), Allende-Piedras Negras Aq. (HGU 60), Valle de Juarez Bolson/Hueco-Tularosa Bolson (HGU 41) and BRB/Gulf Coast Aq. (HGU 72) (Sanchez et al., 2016, Sanchez et al., 2020, Talchabhadel et al., 2021).

4. Discussion

From the total of 28 transboundary aquifers previously reported by Sánchez and Rodriguez (2022), 17 aquifers report major ETAAs within their boundaries. Between Baja California and California, three major ETAAs were identified: one on the Tijuana-San Diego Aq. (HGU 1) which covers basically all the aquifer; another one in Laguna Salada Aq./Coyote Wells Valley (HGU 4) very close to the border in the U.S., and a third one localized precisely within the political boundaries of the States of California, and Arizona, Baja California, and Sonora (Valle de Mexicali-San Luis Rio Colorado Aq./Yuma Imperial Valley – HGU 5), with an overwhelming well density pattern governed by the agricultural sector of the Imperial Valley. Groundwater pumping in this area is expected to increase as reductions to surface water allocations, according to the Drought Contingency Plan (Spivak, 2021), have started to take place at the transboundary level since 2021 (IBWC, 2017). In the case of aquifers shared between Sonora and Arizona, a variety of major ETAAs can be seen along and across three transboundary aquifers. First, the Nogales-Rio Santa Cruz Aq./Upper Santa Cruz Basin (HGU 25) shows a consistent pattern of major ETAAs both on the U.S. side as well as across the border into Mexico. Second, the Rio San Pedro Aq./Upper San Pedro Basin (HGU 27) likewise shows important major ETAAs in its northern portion and towards the south into Sonora, Mexico. The third aquifer is Rio Agua Prieta Aq./Douglas Basin (HGU 29) which shows a very intense and predominant major ETAA along and across the aquifer into the Mexico side. It must be noted that only HGUs 25 and 27 are the only two aquifers recognized as priority aquifers under TAAP (Alley, 2013); however, HGU 29 shows a rather significant vulnerability, groundwater dependency, and potentially higher transboundary impacts than those shared with HGUs 25 and 27. Therefore, this research provides additional support to consider the Rio Agua-Prieta Aq./Douglas Basin transboundary aquifer as a priority aquifer under TAAP due to its vulnerability and increasing dependability of the border population to groundwater resources (Tapia-Villaseñor and Megdal, 2021).

Between Chihuahua and New Mexico, nine major ETAAs are shown. Though their geographical extent is generally smaller as compared to those shown in Groups 1 and 2, their density patterns are the highest in the border region. The Conejos Medanos Aq./Mesilla Bolson shows the highest well density in the border region as well as the shallowest well depths. The vulnerability of surface-groundwater interactions in this region is likewise, highly sensitive and therefore its potential transboundary impacts have shown to be significant (CONAGUA, 2015, Talchabhadel et al., 2021). As surface water in the Rio Grande becomes scarcer, groundwater use in the middle Rio Grande region has become the main source of water for irrigated agriculture on both sides of the border with significant impacts both on over-extraction and salinization levels (Meyer et al., 2012).

Finally, between Texas and Mexico, apart from Valle de Juarez Bolson/Hueco-Tularosa Bolson (HGU 41) which shows a significant ETAA along and across the aquifer at transboundary level, the rest of the major ETAAs show the smallest geographical areas, most of them following the course of the Rio Grande. Contrastingly, this region also shows the largest extensions of minor ETAAs particularly in the Edwards Aq. (HGU 54), Allende Piedras Negras Aq. (HGU 60) and the BRB/Gulf Coast Aq. (HGU 72). A consistent pattern of well-density is identified across the HGUs between Texas and Mexico, even at smaller densities, highlighting the need to address more seriously the consistently shared dependency on groundwater resources systems in the border region. This pattern shows how groundwater dependency is shaping current and future demand for transboundary aquifers in the U.S.-Mexico border, and this trend is expected to increase as surface water becomes scarce and drier conditions persist.

Fig. 8, Fig. 9 show the distribution of ETAAs along the U.S.-Mexico border. The regions of attention and prioritization of areas are clear, as well as the potential for transboundary impacts on either side of the border. Further research will be needed to measure physical and socio-economic impacts at a local and transboundary scale.

The limitation of this approach relies on data availability, mostly from the Mexico side, and the criteria used to categorize water well information across states and among both countries. Therefore, well depth of available neighboring wells was extrapolated to develop the contouring lines. Though this technique can offer some initial insights into shallow versus deeper groundwater layers and potential movement across the unit, it could be overestimating groundwater flows as gradient and vertical analyses, among other hydrochemistry analyses are not considered. There is also the limitation that an important number of wells are not reported or registered to the National Water Commission (CONAGUA) in Mexico, therefore some ETAAs on the Mexico side might been underestimated in extent and magnitude. An additional limitation is that we are not considering socio-economic variables as a potential parameter to define an ETAA. Though this is the ultimate objective of any impact assessment in terms of management and governability, that will be outside of the scope of this research that pertains to identifying first and only the area of attention and potential physical impacts of the complete border. The next step for future research development will be to include socio-economic variables in the measurement of those transboundary impacts within the ETAAs on a more regional or local in-depth analysis.

This research does not pertain to be conclusive, but rather an indicator of potential areas for prioritization purposes at a transboundary scale for future in-depth research.

5. Conclusions

This paper highlights the need to prioritize areas (ETAAs) along and across the U.S. – Mexico border where well density and well depth provide a spatial extent of intensive and extensive groundwater production. Not surprisingly, the identified ETAAs coincide with those aquifers that have already witnessed over-extraction, water quality degradation, and vulnerability of border communities. However, ETAAs provide a clear visualization of the potential geographical extent of where the problem might have originated and its potential hydrogeological and socio-economic impacts. Results indicate a trend between higher well densities at shallow well depths, generally but not exclusively, near international rivers. Therefore, there seems to be a strong potential for local transboundary flows with a faster transboundary impact as opposed to regional flows.

As climate imposes additional challenges to the current limited legal framework on transboundary groundwater resources along the U.S-Mexico border, future research will be needed to measure those potential transboundary impacts at both physical and socio-economical levels, considering current and future scenarios of groundwater dependency. Likewise, it will be important to monitor how the current consistency of ETAAs distribution evolves over time with an increasing demand for groundwater along the U.S.-Mexico border.

These results will not only serve to identify vulnerable regions along the border but also prioritize regions for future water policy development and management schemes at domestic and transboundary scales. The challenges of this and future research mainly lie in its reliance on the availability and reliability of well data. Likewise, another challenge is that for evaluating the extent of potential contamination at transboundary scale for example, the boundaries of the ETAAs might not reflect the accurate extent of vulnerable areas, because the probability of contamination will also depend on specific local conditions of the aquifer/area. Field work to confirm the connectivity of the system at a local scale as well as modelling groundwater flows will be necessary in almost all the transboundary aquifers along the border to confirm our initial groundwater flows insights. In the meantime, ETAAs help as a proxy for prioritization areas that will allow redefining the criteria traditionally used to assess transboundary aquifers using only physical boundaries or administrative boundaries. ETAAs provide an alternative approach to assess the “effective” area of interest.

Under current water scarcity conditions as well as limited data and resources to assess transboundary aquifers in the U.S-Mexico border region, an approach that concentrates on identifying specific and local regions at potential risk according to its level of vulnerability and groundwater dependability, is not only necessary but could offer an efficient way to start designing the route for assessing shared groundwater conditions at a border-wide scale with a local focus.

Supplementary Materials

There is one supplementary document.

Funding

This research was funded by the United States Geological Survey (USGS) through the Transboundary Aquifer Assessment Program Act.

CRediT authorship contribution statement

Rosario Sanchez: Writing – review & editing, Writing – original draft, Visualization, Supervision, Resources, Methodology, Investigation, Formal analysis, Conceptualization. Luna Yang: Writing – original draft, Visualization, Validation, Software, Methodology, Data curation. Duncan Kikoyo: Writing – review & editing, Visualization, Validation, Software, Methodology, Data curation, Conceptualization.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors are grateful for the data provided by Josselin Portugal at the Universidad Autonoma de Chihuahua.

Appendix A. Supplementary material

Download: Download Word document (1MB)

Supplementary material

Data Availability

Data will be made available on request.