Introduction

Signed in November 2012 by the Mexican and U.S. sections of the International Border and Water Commission, Minute 319 has been the object of positive comments regarding future possibilities in the international basin of the Colorado River (^{Minute No. 319 of the International Boundary and Water Commission, 2012}). In official documents and at some environmental agencies that support this draft, the minute has even been described as a re-negotiation of the International Waters Treaty of 1944. An issue mentioned by the press is the benefits of base flow and pulse flow for the ecosystem itself, flows that have improved some zones of the Mexican section of the low Delta (^{Haro, 2017}). This conclusion is supported by diverse studies of the Delta carried out as early as the 1990s (^{Glenn, Felger, Burquez & Turner, 1992}; ^{Glenn, Garcia-Hernández, Congdon & Luecke, 1999}; ^{Glenn, Lee, Felger & Zengel, 1996}; ^{Glenn et al., 1998}; ^{Luecke et al., 1999}; ^{Valdés-Casillas et al., 1998}).

As a counterpart, there have also been actions against and criticism of the signing of Minute 139. Organizations of Mexicali Valley farmers, who were left without the means to farm, have questioned this Minute (^{“Forman Frente Común”, 2013}). In 2013, 778 farmers disclosed to the media that they had not received water since 2010. Baja California's state legislature even urged the Senate of the Republic to establish a working group on the matter (^{Dictamen de acuerdo del congreso [Congressional approval decision], 2013}). The Senate's response was that it was not its responsibility to review the Minute (^{Boletín 1398, 2013}). An order was issued to Enrique Martínez y Martínez, past secretary of Agriculture, Livestock, Rural Development, Fisheries and Food, and also to David Korenfeld, past general director of the National Water Commission, to provide the farmers with water. In August 2015, Max Correa, representing 250 farmers of the Central Campesina Cardenista, indicated that he would present a lawsuit to the Latin American Water Tribunal denouncing the International Boundary and Water Commission (IBWC) for “acts of corruption in water management” (^{Muñoz, 2015}).

One of the most important points of Minute 319 refers to proactive measures regarding the possible effects of climate change. It is affirmed, mostly by academics, environmental institutions and organizations, that the flow rate estimate for the decade 1910-1920 registered higher levels of humidity than in following decades (^{Robison et al. 2014, p. 18}). In Minute 319, it is agreed that both countries recognize the value of an interim period in which the Colorado River is operated in a proactive way “in light of historical variability and the possible increase of this in the future, due to climate change”. Based upon this, proposals for the modification of deliveries to Mexico were drafted, as this was a long debated matter.

Repeatedly over the last 60 years, it has been observed that the Draft does not define the term “extreme drought” (^{Hundley, 1963, p. 167}). As this definition was supposedly not provided, in light of the prospect of a dry 21st century (^{Cook, Ault & Smerdon, 2015}), the matter became an important issue in bi-national cooperation. Given this situation, a large number of documents have been produced to compel both countries to take alternative measures to respond to a natural condition that would appear to be irremediable: climate change and its consequences for productive processes, urban supply, electrical energy production, fishing and environmental conservation (^{Christensen & Lettenmaier, 2006}).

In different media outlets, an important relationship has been noted: that between the flow variability of the Colorado River and climate change. In recent years, many authors and studies of the Reclamation Bureau have published the results of forward projections of water availability over the next several decades (^{Barnett et al., 2004}).

In this context, the purpose of this article is to demonstrate that the variability of the Colorado River has been used since the beginning of the 20th century to justify 1) support of the construction of dams with multiple uses, 2) impeding the 1944 treaty with Mexico, 3) promotion of the diversion of water from other currents, such as the Columbia River, 4) the attempt to change the conditions of the 1922 Santa Fe pact¹ or 5) negotiation of reductions based on possible conditions generated by global warming, as in the case of Minute 319 (^{Overpeck et al., 2013}; ^{Squillace, 2012}).

The process through which the 1906 and 1944 treaties were constructed was based upon an idea of the future framed by the conservationism that triumphed over preservationism in the U.S. (^{Samaniego, 2006}, ²⁰¹⁵). In this context, the objective was that the water resource not be emptied into the ocean: this was a waste that should be avoided at all costs. Instead, the resource should be used to generate electrical energy (^{Blanchard, 1924}; ^{Gray, 1913}). This notion of a relationship with nature impacted the decision that all of the water -in the future and in perpetuity- of the Rio Grande and Colorado River be allocated to states having basins with projections of use over large distances and in close relation to the production of electrical energy. The “extreme drought” referred to in both treaties would occur within a period of one year. The basis of negotiations and projections was created using acre-feet, an English measure of the amount of water at a height of 1 foot during one year. This measurement was considered to be most suitable, given the types of crops present. This measure was the focal point of all negotiations, and it is a concept that is strongly linked to that of “extreme drought”, that is, the concept is not defined in the treaties because it is directly linked to an annual measure.

In this article, we highlight that in Minute 319, as well as in press releases commenting on it, current variability is linked to climate change. Nonetheless, the current has varied since its flow-rate began to be recorded at the end of the 19th century (^{Birdsong, 2011, p. 118}; ^{Bureau of Reclamation, 1946}; ^{La Rue, 1916}). Flow-rate variation was used to argue for the construction of dams, to argue for efforts to stop the treaty with Mexico, to justify the supposed ambiguity regarding the quality of the Colorado River's water, and to defend Minute 319. For this reason, although we only comment on it briefly, we consider Minutes 317, 318 and 319 to be aspects of the treaties that Mexico negotiated while pressured by the effects of the earthquake of April 4, 2010. This same earthquake affected the irrigation canals in the Mexicali Valley, and U.S. aid was required to conserve the resource in the neighboring country's hydraulic system. It should be noted that there are other points of utmost importance with regard to Minute 319. However, for the purposes of this article, we address only the aforementioned aspects.

In the present article, we address the subject in chronological fashion. First, we present bibliographical references on the river's variability, dating to the first years of the 20th century. Then, we show that the river’s variability was used to justify projects such as the Hoover dam and the All-American Canal. During arduous negotiations in the U.S. Senate, the same argument was used to prevent these projects. In the 1960s, the river’s variability was used to justify a project that would transfer water from the Columbia River to the Colorado River. In recent years, the river’s variable current has been linked to climate change. In other words, current variability has been used in diverse ways to change the ways in which the water resource is distributed.

Brief Reflection on the Colorado River

The Colorado River is an international current shared by Mexico and the United States. In the United States, it runs from northeast to southeast, passing through the states of Wyoming, Utah, Colorado, Nevada, California, New Mexico and Arizona, stretching some 2 300 kilometers. The total area of the river basin is 484 330 square kilometers, 3 840 of which are in Mexican territory, the rest being in the United States. In percentage terms, 99.28% of the river basin is in the US, and only .72% of it is in Mexico. Thus, practically all of the river basin is located in the US (^{Colorado River Development, 1928, p. 150-213}).

Currently, and mostly due to the September 11th attack on New York City, this body of water is seen as a matter of national security in both Mexico and the United States (^{Cortez, Whiteford & Chávez, 2005}). Approximately 40 million people depend on the water resource for diverse social uses. The Colorado River is the source of water for urban uses in such cities as Los Angeles, California; Denver, Colorado; and Phoenix, Arizona, as well as for agricultural valleys in Mexico and the United States. Among the largest of these valleys are the Imperial Valley in California and the Mexicali Valley in Baja California, Mexico. Additionally, the current is used to generate electrical energy for fishing and tourism. In California alone, 17 million inhabitants use the water resource in one way or another. A city such as Tijuana in Mexico, with a little over 1 500 000 inhabitants, obtains 98% of its water from the Colorado River current, which is located 180 kilometers from the city.

One of the natural conditions in the lower part of the international Colorado River basin is its aridity. The existence of large desert areas contrasts with the presence of a current that for many years was a natural barrier to the movement of people and goods. In the 20th century, with advances in hydraulic technology, the current could be controlled and managed. In the words of Secretary of the Interior Hubert Work, the intention was to turn the great river from a “natural threat” into a national resource (^{Work, 1925}). This generated a great transformation in the ways in which the current was used and how large areas were populated and developed, for example, in the lower Delta section (despite summer temperatures that reached 50 degrees centigrade), a highly productive region with close ties to international markets.

Variability as a Constant: An Argument for Different Ends

Many authors affirm that the 1922 Santa Fe pact was signed with optimism and that the amount of water in the current was over-estimated (^{Adler, 2008, p. 30}; ^{Birdsong, 2011, p. 118}). However, a study by E. C. La Rue, which provided the basis for the analysis of the current's potential, and the declarations of representatives of entities such as the state of Arizona do not support these claims. In the United States, some promoted the idea that there was sufficient water for all. This was true of Reclamation Service director Arthur P. Davis and Secretary of Commerce Herbert Hoover. However, Arizona representative, W. S. Norviel, upon hearing such comments, asked: If there is sufficient water for all, why were restrictions established regarding flow quantity (^{Colorado River Commission, 1922}).

The engineer E. C. La Rue noted in his document that his results were an approximation and that demand could exceed the actual amount of water available, i.e., that there may not be sufficient water for all of the land that could be cultivated. In the table published by La Rue, the river current was estimated to range from 7 960 000 acre-feet to a maximum of 25 300 000 acre-feet in 1909. The average for the years 1895 to 1914 was 16 200 000 acre-feet (^{La Rue, 1916, p. 192}). The author stated that construction of derivations far from the river course, as had in fact occurred, would prevent a flow return, reducing the river basin’s size. Similarly, the author estimated water loss through evaporation in potential dams to be approximately 1 000 000 acre-feet per year (^{La Rue, 1916, p. 167-169}). La Rue's conclusions were a constant reference for subsequent plans (^{Colorado River Development, 1928, p. 150-213}) but were also among the most criticized by those who argued that he had miscalculated. In another report published in 1925, La Rue affirmed that the river's current would be changed when the projects that were planned -13 at that time- were carried out (^{La Rue, 1925, p. 9}).

F. E. Weymouth, the Reclamation Service's head engineer and among the authors most cited in hearings on the proposed dams, conducted a study of the construction of the Boulder dam. In a report presented to the United States Senate, Weymouth wrote in one of its sections:

The Colorado River is subject to periods of great floods and great droughts. It has been known to reach a maximum discharge of more than 200 000 cubic feet of water per second and a low flow at the headworks of the Imperial system of 1,250 cubic feet of water per second (^{Colorado River Development, 1928, p. 75}).

Thus, the United States Senate's Irrigation and Reclamation of Arid Lands Studies Committee learned that the amount of water in the Colorado River current changed from year to year, averaging 17 000 000 acre-feet. In accordance with the conservationist vision, it affirmed that it was necessary to avoid wasting water and thus prevent water from emptying into the ocean (^{Boulder Canyon Reclamation Project, 1926}).

Current variability was among the arguments presented by those who supported the construction of the aforementioned dam. An example was Clyde L. ^{Seavey, who in 1928} published an article titled “What the Boulder Dam Project Means to California and to the Nation”, supporting the dam. Seavey argued that, due to the changes that occurred each year in the river basin, the dam was necessary to regulate the current, produce electrical energy and prevent water from being emptied into the sea and wasted (^{Seavey, 1928}).

Many texts have been written about the 1922 Santa Fe pact (^{Hundley, 1975}; ^{Olson, 1926}). This document contains tentative estimates of the Colorado River current: 17,000,000 acre-feet and 18 000 000 acre-feet with surpluses. It should be noted that La Rue never mentioned the second figure. In the pact, 7 500 000 acre-feet were allocated to entities upstream from the basin, and the same amount was allocated to entities downstream, California, Arizona and Nevada. In the case of surpluses, these last three entities could be allotted 1 000 000 more acre-feet. Similarly, how water would be delivered to Mexico in the event that a treaty was implemented was considered. The upper and lower states (upper and lower basins, respectively) should divide equally what was considered to be the projected amount: 1 500 000 acre-feet for Mexico.

The pressures brought by the basin states were diverse and conflicting and created interrelated possibilities. It is impossible to enumerate all of them in this article. However, it is important to note that the figure was annualized for the purpose of establishing a parameter for negotiating what everyone wanted: the possibility of future development. That is, the aforementioned figures were drawn up on paper, and several years later, amounts that exceeded the stipulated amount arrived in Mexico. The building projects that we mention in the following pages systematically reduced this amount. What pressures existed in the years before the Santa Fe pact? Below, we mention only a few points that contextualize the negotiating process.

Under this framework, which is even more complex, mainly due to a dispute among private companies that wanted to take advantage of electrical energy production, the figure of 15 000 000 acre-feet was established among the basin states, along with the greater estimate that was already mentioned, 16 200 000 and 18 000 000. However, as we have noted, this was the result of arguments and dealings among a group of actors who were interested in obtaining the resource at a future time rather than dependable information based on confirmed data. La Rue had been clear. Extractions that would not allow return flows would reduce the flow rate. The lakes formed by the dams would also decrease in size due to evaporation (^{La Rue, 1916}).

In the Senate's irrigation committee report in 1926, it was noted that an aqueduct for Los Angeles and other neighboring counties was necessary. The committee indicated that on the Pacific coast, there were periods of drought that lasted between 10 and 12 years. The bonds to pay for the project, which amounted to two million dollars, had already been authorized (^{Boulder Canyon Reclamation Project, 1926, p. 21}). A similar situation existed in Denver, Colorado in the first decade of the 20th century. Such extractions, which became common in the following years and do not allow for return flow, are exactly what La Rue described.

In December 1928, construction of the Boulder dam was approved. The declared objective was to control the current, produce electrical energy and deliver water for farming uses. The main priority was to protect the Imperial Valley in California from floods and correct, to the extent possible, the tremendous salinity problem caused by the current's passage through the Grand Canyon. Similarly, under the principle of “first appropriation”, California was guaranteed greater use of the river's water. Undoubtedly, Mexico benefitted greatly from this building project, due to the reduction of flooding and sediments and to the opening of land that allowed it to claim the amount of water it required for 200 000 projected hectares. On the other hand, with the dam and the construction of the All-American Canal, Mexico lost its advantage in negotiations on international currents, as the Imperial Valley would no longer depend upon Mexican territory (^{Samaniego, 2006}).

The Boulder dam signified a great triumph for the engineering methods that it sought to replicate. Thus, while this project was developed, research was conducted, and the Colorado-Big Thompson Project in Colorado was approved. This project required that electrical energy be produced, so that the water would rise and then, through the pull of gravity, reach the eastern side of the mountains. With this, water from the Colorado River withdrew from return flow processes, as has occurred in many projects since that time.

A pair of droughts in 1932 and 1934 as well as the reference to benefits in Mexico led representatives of California and Arizona to oppose the amount of water that Mexico intended to claim. During these years Mexico modified its position and claimed 1 500 000 acre-feet annually. California sought approval for 750 000 acre-feet, while Arizona claimed nothing. Two court cases arose between California and Arizona, both of which were won by California. For this reason, during the process of treaty ratification, Arizona preferred to oppose California's wishes. The upper basin entities, which included Colorado, supported the treaty with Mexico. As we will see below, Colorado's attitude changed substantially during this century.

The Earthquake of April 2010 and Minutes 317, 318 & 319. A New Form of Distributing a Resource? An Unrecognized Re-negotiation of the 1944 Treaty?

California and the Baja California Peninsula are located on the San Andreas Fault (^{Cruz-Castillo, 2002}; ^{Shor, Russell & McGowan, 1976}), a seismic zone that includes San Francisco, California and whose southern limits include the entire Baja California Peninsula.

On April 4^th, 2010, an earthquake with a magnitude of 7.2 on the Richter scale affected California, Baja California and Baja California Sur. There were various effects in the Mexicali and Imperial Valleys, where earth tremors occur with greater intensity. One of these effects was damage to the irrigation canals that supply water for farming in Mexicali. In light of this damage, if monthly deliveries made in coordination with the International Boundary and Water Commission (IBWC) were sustained in accordance with established timetables, the result would be a loss of vital liquid and possible flooding in certain areas.

It should be noted that since the 19th century, the International Boundary Commission -International Boundary and Water Commission since 1944- has signed minutes to reach agreement regarding diverse aspects. Regarding the issue that concerns us here, we note that following the earthquake, the conditions of the irrigation canals placed Mexico in a situation where it had to negotiate. Several issues were up for debate, but mostly those raised by conservationists who wanted water to be allocated to the restoration of zones in the lower Delta (^{Conway, Nadeau & Piest, 2010}). In 2000, Act 306 was signed. This act responded to proposals of this sort, but a decade later, no progress had been made in its implementation.

Minute 319

Minute 318 was signed under the following assumption: the irrigations canals in the Mexicali Valley would be ready by December 2013. Nonetheless, by mid-2012, it was clear that this would not be possible. Therefore, on November 20, 2012, Minute 319 was signed, ten days after the change of Mexico's Federal Administration. It is helpful to note that for the first time, a different method was established in reference to a portion of the water that Mexico will receive through December 31, 2017. That is, it is a minute that includes a five-year plan, one that could be extended through 2026 if it works well for both countries.

This consideration has occasioned commentary and criticism, given its future importance in the Colorado River's international basin. In this sense, as we have already noted, this minute's broad scope and the fact that it could mean de facto renegotiation of the International Waters Treaty of 1944 have been mentioned in formal documents and by environmental associations that support it. Evidently, the mere mention of this point has generated various tensions among farmers of the Mexicali Valley (^{Arellano, 2013}).

In Minute 319, other important issues were touched upon that we will not address in this article, as we limit ourselves to water deliveries to our country. Therefore, we highlight that in the minute, the scenario of the amount that corresponds to the 1944 treaty is modified, as the height of Lake Mead, which was formed by the aforementioned Hoover Dam, is taken as a reference. Both countries recognized the value of an interim period of cooperation, which would enable the body of water to be pro-actively managed “in light of historic variability and its potential future increase due to climate change. Thus, this proactive management would maximize the resource's use during conditions of storages variability, for the benefit of both countries” (^{Minute No. 319 Interim International Cooperative Measures in the Colorado River Basin, 2012, p. 3}). As we have noted, this is not the first time that the variability of the river's flow rate has been used in arguments relevant to decision-making. This argument was the basis for dam construction, discussions about how to distribute the resource, opposition to the treaty with Mexico and the attempt to transfer water from the Columbia River. On this occasion, it was related to an issue on the global agenda: climate change.

In this way, in Minute 319, it was established that in the event that a decrease is registered, the negative effects would be proportional in all of the lower basin states, that is, California, Arizona and Nevada in the United States and Baja California in Mexico. The entities considered upper riparian states. Wyoming, Colorado, Utah and New Mexico in the United States, are not involved in this agreement, as the modifications correspond to waters below the Hoover Dam. In other words, as a result of the undeniable current variability, of which there was knowledge as early as the beginning of the 20th century and which led to prognoses of water cutbacks in the 1940s and 1960s, it was possible to change deliveries in direct relation to the effects of the 2010 earthquake. As we have noted, in principle, this is a five-year plan, even though the possibility of its extension to 2026 is being considered. It is noted that a complete minute will be generated. We can only assume that this has been a source of debate within the CILA-IBWC.

In accordance with the aforementioned, the changes would consist of the following; Mexico can receive surpluses in accordance with the increases in Hoover Dam storage (see Table 1).

Table 1: Annual increase for Mexico, according to the height of Lake Mead  

Source: ^{Minute No. 319 Interim International Cooperative Measures in the Colorado River Basin (2012)}.

In the opposite direction, reductions are stipulated in cases of supply shortages. This issue causes concern mainly among farmers of the Mexicali Valley. Mentioned in the minute are the amounts of water that are no longer delivered to U.S. entities of the lower basin, and similarly, the reductions for Mexico are established. The figures are based on January 1st of each year, and reductions can be observed in Table 2.

Table 2: Annual reduction for Mexico, according to the elevation of Lake Mead  

Source: ^{Minute No. 319 Interim International Cooperative Measures in the Colorado River Basin (2012)}.

In this way, and without the formal approval of both countries' senates, the table of deliveries, defined in direct relation to the Hoover Dam storage, was modified. While the temporary nature of Minute 319 is repeatedly mentioned in its contents, this is not sufficient for farming sectors that see it as a violation of Mexico's rights by the United States. As we already observed, the 1944 treaty foresaw reductions due to consumption, not flow rate. It is among the points that have sparked controversy, given the way in which “extreme drought” is interpreted, as noted at the beginning of this article.

On the other hand, in the agreement signed in Minute 319, the concept of Intentionally Created Mexican Allocation (ICMA) appears. This concept allows the country to have a reserve in Lake Mead in the event that it should need it before 2018. In the event that Mexico does not need this reserve, it can sell it in the U.S. water transfer market, which has been in operation for several decades. California in particular is perhaps the entity that, since the 1970s, has generated the largest number of agreements for carrying out transfers (^{Bauer, 2010}).

Preventing a Dry Future. Conceptual Change Regarding the Meaning of Historical Variability

In this article, we have proposed that the issue of flow rate variability has persisted since the beginning of the 20th century. Is current variability an effect of climate change or the result of evaporation in dams and extractions that eliminate return flows? As noted above, 68 years ago, ^{Parsons (1950)} observed the over-allocation of the Colorado River. The discourse on climate change has permeated many decisions and has generated a large number of proposals and changes in future scenarios.

In March of 2014, Sally Jewell, Secretary of the U.S. Department of the Interior, and Juan José Guerra, Mexico's Secretary of the Environment, issued declarations regarding the virtues of Minute 319. In these declarations, they related the changes in Mexico's water allocation described here to the benefits of the document with regard to the flow that will allow for the restoration of areas of the Colorado River's lower Delta (^{Jewell & Guerra, 2014}). These authorities indicated that “…the Mexican and United States governments have learned about the limits and flow variability of the Colorado River. For this reason, we are convinced that cooperation is the best way to confront these challenges” (^{Jewell & Guerra, 2014, par. 7}).

In the note, it would appear that the issue of variability is something that we learned about recently, whereas in reality -the main argument of this article- knowledge of it has existed for more than a century, and at different times, it has been the central point in debates concerning possibilities of the current's future use. Furthermore, while some authorities indicated that the 1944 treaty did change, others affirmed the opposite: there had been no change (^{Bono, 2012}). Regarding the issue of concern here, we note that this point remains vague and that, in the minute, the 1944 treaty is cited as a principle reference. Additionally, it is explicitly noted that the minute is not a precedent for future agreements. In other words, with the exception of the minute's extension, the international treaty is what will be fulfilled, this being in the interest of both parties. In this sense, in our reading of the minute, there is the possibility of maintaining the agreement regarding the height of the Hoover Dam until 2026.

To Conclude

With the above analysis, it is not our intention to add to arguments that negate climate change (^{McCright & Dunlap, 2000}). Our position is not one of calling into question this socially constructed global phenomenon. We only indicate that the variability of the Colorado River is a factor that has been studied and recognized since it became the focus of research at the beginning of the 20th century.

Approximately 60 million people depend upon the resource of two international rivers between Mexico and the United States, the Colorado River and the Rio Grande (others exist that we do not mention here). Here, we only address the case of the Colorado River, which involves approximately 40 million inhabitants. While other issues might be higher on the bilateral agenda, such as immigration or drug trafficking, the issue of possible conflicts over water persists, and the question remains: At what point could the river be substantially modified by climate change? For our part, we indicate that variability has been extensive in certain periods and not only as an effect of climate change.

Minute 319 is a preventative measure within a framework in which measures to address climate change is the discourse with the greatest impact. Therefore, it appears that the arguments used are a solution to a new problem or condition: the variability of the Colorado River. Nonetheless, this condition is a long-term issue that has appeared repeatedly.