Long River, Short Water: The Rio Grande Water Development Story · 2007-06-01 · flows from Colorado diversion, were less able to maintain themselves. THE RIO GRANDE PROJECT Water

T he rich alluvial valleys of the Rio Grande havesupported agriculture for nearly a millennium. In

a semiarid land, through capricious swings of droughtand flood, the soils and water of the river have nur-tured substantial civilizations and inspired cultural tra-ditions that continue to enrich modern New Mexico.

The river’s first farmers were Pueblo people. Onescholar estimates sixteenth century Pueblo populationsas great as 80,000 persons, in 100 villages, making theRio Grande Pueblo civilization the greatest concentra-tion of settled farming villages in the AmericanSouthwest. The first farmers were irrigators, thoughthey appear to have relied more upon such elegantmoisture-conserving techniques as water-retaining ter-races, cobble mulches, and self-contained “waffle gar-dens” than on intensive dam and canal systems.Because labor requirements were high, Puebloan agri-culture was a necessarily cooperative venture. Theiruse of resources was likely governed less by politicalcontrol than by traditional sacred relationships toland, sky, and river.

Though at times as many as 30,000 acres may havebeen cultivated, Pueblo impacts to the stream wouldseem modest to modern-day farmers. Europeanexplorers marveled at the quality and abundant yieldsof Pueblo plantings. Indeed, the earliest Spanishcolonists might have perished without the surpluses ofcorn and beans laid up by the first farmers.

THE SPANISH ENTRADA

In 1591 the frontier of European expansion reachedthe Rio Grande. Driven hither by the quest for wealthand Christian evangelism, Spanish conquistadorsfound both minerals and souls were hard to come by.Still, compared with the expanse of desert to thesouth, the Rio Bravo del Norte offered rich soils andabundant water. Encouraged by grants of land fromthe royal government, a stream of Spanish immigrants,mainly impoverished exiles, flowed into the regionover a 250-year period and conquered the north.

An advanced irrigation technology came with them,in the form of acequia agriculture. Headings of rockand brush and hand dug canals served to turn wateronto pastures and cultivated fields. In the few largetowns royal governments, alcaldes and ayuntamientos,

governed colonial affairs, including the division ofwater, whereas in the many small villages an indige-nous water democracy maintained a cooperative gov-ernance by acequia majordomos and comisionados.

Spanish traditions, grafted onto new world realities,suggest that water users shared the benefits and lossesof the variable supply. Priority of first use was respect-ed, though not with such exclusivity as in the modernappropriation doctrine. When water was scarce,demonstrated needs (especially for drinking water andstock watering) and concepts of fairness were oftenthe basis for an allocation decision. These traditionscontinue to carry legal weight, as formalized in the1907 Territorial Water Code.

By 1821 acequia agriculture, both Pueblo andHispanic, had grown until it involved no more than150,000 scattered acres between Taos and Tome.Possessed of pragmatic technologies for water control,acequia irrigators relied on cooperation, hard labor,and the will of God to bestow the blessings of theriver to their land.

ANGLO-AMERICAN CONQUEST

The United States’ conquest of Mexico’s northern terri-tories in 1848 signaled a profound transformation ofthe localized, cooperative traditions of water develop-ment and governance on the Rio Grande. In theirplace, industrial technologies and the U.S. doctrine of“Manifest Destiny” established a model for nationalpossession of western lands. Henceforth, a restlesshoard of speculators swept over North America’s vastwestern empire to turn minerals, timber, grass, andwater into dollars.

Agriculture was a key part of the U.S. national poli-cy of rapid immigration and development of the West’snatural resources. However, to farm successfullybeyond the one hundredth meridian required irriga-tion. From a few successful experiments with large-scale irrigated farming, ambitious water diversion proj-ects spread like prairie fire to every river valley in theWest. Quickly, possession of western rivers was grant-ed not to the owners of the land or the communitiesthrough which they flowed, but to the persons whobuilt the works that diverted them.

The growth of irrigation from the Rio Grande typifies

Long River, Short Water: The Rio Grande WaterDevelopment Story

Steve Harris, Rio Grande Restoration

WATER RESOURCES OF THE MIDDLE RIO GRANDE

T H E P H Y S I C A L A N D H I S T O R I C A L F R A M E W O R K 7

DECISION-MAKERS FIELD GUIDE 2007

the explosiveness of this process. In 1850 Rio Grandefarms from San Luis Valley of Colorado to El PasoValley, Texas, totaled less than 200,000 acres. By thetime the temporary Rio Grande Compact was signedin 1929, irrigation in the basin encompassed morethan 1,000,000 acres.

COLORADO DEVELOPMENT

The vast, fertile, high-elevation San Luis Valley wasnot settled until 1851, when Hispano settlers spillednorthward from the Taos region and were soon joinedin 1878 by westering Mormon farmers. The openingof the Denver and Rio Grande Railroad to Alamosa in1878 ignited the biggest of the Rio Grande’s “big bar-becues.” Between 1880 and 1890 British speculatorsfinanced five large canals dug by mule-drawn scrap-ers. Cumulatively these canals could divert almost5,000 cubic feet per second, virtually the entire springrunoff of the main river. Colorado attempted to securelegally the natural advantage of its location at the topof the hydraulic system, claiming an unimpeded right

to the waters that arose within the state. Though dis-abused of their “doctrine of sovereignty” by theSupreme Court in 1917, the great deeds of irrigationdevelopment were largely already done. A generalstream adjudication completed in 1891 showed thatmore than 300,000 acres had been placed underditch. The impact of this rapid and extensive develop-ment would be immediately felt by downstream water

C H A P T E R O N E8

users and would impact the course of Rio Grande history for more than a century to come.

NEW MEXICO

Meanwhile and farther south, expansion of irrigationwas also occurring, though at a more restrained pace.In the Rio Arriba (that stretch of the river north of La

Bajada), irrigation had long since reached its fulldevelopment, with perhaps 100,000 acres applyingwater from the several hundred acequias originating intributary streams. In the Rio Abajo (middle RioGrande), after the settling of Tome in 1739,Spanish–Mexican farming grew continuously until1880, when it comprised about 124,000 acres, irrigat-ed by more than 70 traditional acequias.

When the Atchison, Topeka, and Santa Fe Railroadreached Santa Fe in 1880, New Mexico also connectedto commerce with the wider nation. Industrial-scalegrazing began to make economic sense, and the northexperienced a sheep-grazing boom. Likewise, thou-sands of acres of Sangre de Cristo forests were harvest-ed and boomed down the Rio Grande to EmbudoStation for use as railroad ties. A result of this large-scale timber development, abetted by a national policyof fire suppression, was that the Rio Grande’s high-ele-vation watersheds, upon which the region’s acequiasdepended, were rapidly transformed. Snowmelt in the Rio Grande’s tributaries began to come more

Irrigated acreage in San Luis Valley, Colorado, illustrating thegrowth in upstream water diversions, which has dramaticallyimpacted New Mexico water supplies from the Rio Grande.

"The Big Barbecue”—a term applied by historian CharlesWilkinson to the rapid conversion of Western resources into wealth in the late nineteenth century. This log jam is onEmbudo Creek, circa 1915. The timber was headed to the Rio Grande, through White Rock Canyon, and ultimately byrail to Albuquerque.

quickly and in reduced volume, its hydrographs lessattenuated into the summer season. Sediments releas-ed by logged-off forests and grazed-off grasslandsaggraded river channels, which, with reduced peakflows from Colorado diversion, were less able to maintain themselves.

THE RIO GRANDE PROJECT

Water development in the Mesilla and El Paso/JuarezValleys followed the pattern of the Middle valley, up toa point. An 1858 survey portrayed acequia farming inthe region utilizing about 10,000 acres. By 1880 asmany as 25,000 acres may have been irrigated, andspeculators had their eyes on more.

But in 1890 what would become a nine-yeardrought descended on the Rio Grande. The years oflow snowpack dampened the onrush of developmentand led to substantial, if temporary, declines in irrigat-ed acreage. The San Luis Valley took whatever water itcould, and the middle valley often diverted whatremained of the river. In the El Paso/Juarez Valley,these upstream diversions compounded the droughtand caused the region’s famous vineyards to witherand die. About the same time, and far downstream,steamboat navigation of the Rio Grande below Laredoceased forever.

In 1889 the Rio Grande Dam and IrrigationCompany of Mesilla, New Mexico, was incorporatedand proposed the construction of a reservoir and canalsystem to irrigate 530,000 acres in Mesilla Valley. By1895 the company had received approval for a reservoir

right of way from the Secretary of Interior. El Paso/Juarez farmers responded to the impending, profoundwater shortage with outrage, leading Mexico to filedamage claims for $35,000,000 against the UnitedStates. Here was a serious diplomatic breach, and theInternational Boundary Commission was assigned tostudy the problem. The problem, their report conclud-ed, was that the border area “suffered from theincreased use of water in Colorado.”

After a decade of diplomatic wrangling border offi-cials in the two countries determined that the solutionwas to build a storage dam at the El Paso Narrows.Suddenly there were two conflicting reservoir propos-als on the table. El Paso/Juarez interests were utterlyopposed to the Elephant Butte Project; it was too faraway, and the speculators were geographically toowell-positioned to intercept and control the water.New Mexico strenuously opposed the El Paso dam,which provided no water storage for proposed devel-opments around Mesilla.

By 1906 an unlikely, but momentous, series ofevents had occurred, resulting in resolution of the 18-year-old problem:

• First, a territorial court rejected U.S./Mexicoarguments that the private dam would illegallyinterfere with navigation of the river. Resolvingthe litigation in the case’s third review, the U.S.Supreme Court ruled that the Rio Grande Damand Irrigation Company had waited too long tobegin construction on the Elephant Butte Project.Its patent was thereby repealed, clearing the wayfor a single federal project.

• The International Treaty of 1906 “to equitablydistribute the waters of the Rio Grande” wassigned by both nations, who assented to a three-way split of the lower Rio Grande. In the 1906treaty, Mexico settled for a 60,000 acre-feet guar-antee, delivered from the reservoir each year“except in times of extraordinary drought.”

• After more than twenty years of stormy debate onhow best to advance western irrigation, Congresspassed the landmark Reclamation Act of 1902,creating a firm policy of federal financing (andcontrol) of irrigation development. In exchangefor the territorial engineer’s granting 730,000acre-feet of water rights to the federal Rio GrandeProject, New Mexico induced its border neigh-bors to accept the Elephant Butte Reservoir site.

• In 1908 “all the unappropriated waters of the Rio

Irrigated acreage in Mesilla, Rincon, El Paso, and Juarez valleys,New Mexico, Texas, and Mexico, illustrating the sharp growthin irrigated acreage after construction of Rio Grande Projectcanals below Elephant Butte.





resolve the conflict was clear. Colorado would consoli-date the dramatic gains it had made and perhaps beallowed to build a storage reservoir. The middle valley,too, would need its own reservoir to regulate late sea-son supplies. Rio Grande Project users wanted assur-ance that the others would leave enough water to sup-ply their needs and aspirations. Thus informed, thecompact commissioners and their legal and technicaladvisers negotiated, over three years, a set of deliveryschedules and various caveats to fix their irrigationdemands to the fluctuating supply, resulting in thepresent Rio Grande Compact.

MIDDLE RIO GRANDE PROJECTS

With or without interstate accords, the Middle RioGrande valley increasingly found itself in a desperateposition: bracketed by two thirsty, fast-moving com-petitors, one of which had recently vouchsafed a claimto virtually the entire flow of the river. Its organizingprinciple, the acequia system, isolated it from thepower politics of large-scale irrigation. It had no reser-voir to regulate a diminishing river. Its economy wasalso declining in lockstep with intensifying competi-tion from the other two regions and the deterioratingcondition of its lands.

Not only were its supplies of river water diminish-ing, but in the mid-river, the Rio Grande was leakinginto the fields. By 1896 irrigated agriculture haddeclined from a high of 125,000 acres to 50,000

Grande and its tributaries” were decreed to theReclamation Service. In 1911 construction beganon the project; by 1916 a completed ElephantButte Reservoir, one of the Reclamation Service’sfirst projects, began to store the 2.6 million acre-feet for which it was designed.

RIO GRANDE COMPACTS

A federal embargo, declared by the Secretary of Interiorin 1896, prevented Colorado from constructing thedam it desired at Wagon Wheel Gap but otherwise didlittle to reduce the development of new irrigation.While Supreme Court cases, international treaties, andmajor reservoirs were being negotiated, the sovereignstate of Colorado continued expanding its exploitationof the Rio Grande. In 1924 San Luis Valley water com-missioners reported a total of 621,826 acres under irri-gation, up from 213,210 in 1896.

Fervid ambition was the only factor governing eitherthe Rio Grande Project or San Luis Valley develop-ment, as by 1929 irrigation had increased dramatical-ly, both downstream of Elephant Butte and inColorado. Even with a period of abundant snows inthe 1920s, the Rio Grande regularly disappeared atlate season in the middle valley. It was becoming clearthat only an interstate agreement could bank the firesof the Rio Grande’s big barbecue.

In 1925 the embargo was lifted, as Colorado, NewMexico, and Texas agreed to seek an interstate com-pact equitably dividing the Rio Grande among them.A temporary compact was put into effect in 1929 tofreeze the apportionment at then-current levels. Toeffect a permanent agreement, and to determine thenature of the water supply and the relationship ofeach segment’s demands, required the collection andanalysis of water-use data. Under auspices of theNational Resources Committee, national and state sci-entists conducted an exhaustive joint investigation todetermine the facts needed to equitably balance eachsection’s inflow, outflow, and demand.

Completed in 1935, the National ResourcesCommittee’s regional planning report provided thefoundation for a definitive negotiation among thestates. The report acknowledged that the Rio Grandewas at or beyond the limits of the water it could beexpected to provide: “…with the available waterresources of the Rio Grande apparently fully appropri-ated, the approval of any new projects involving addi-tional drafts upon those resources seem to pointinevitably to further conflict….”

The three sections’ bottom line for negotiations to

Irrigated acreage in Middle Rio Grande, Cochiti to San Marcial,New Mexico, illustrating precipitous declines in irrigation when sediment burdens and aggrading channels combined towaterlog more than half of the historically irrigated lands. Note also the partial reclamation resulting from MRGCD andfederal projects.

acres. The deadly combination of silt from deteriorat-ing watersheds being deposited in the channel,reduced channel-forming flows from water interceptedby upstream irrigators, and its own flood irrigationpractices clogging the Rio Grande, raised its channelabove the elevation of the surrounding floodplain andseeped into much previously productive land. A 1918state engineer inventory of middle valley conditionsrevealed nearly 60,000 acres of waterlogged and alka-li-salted former farmland. In addition, the aggradingriver flooded with increasing frequency, playing havocwith earthen irrigation works and cutting sandy chan-nels across the beleaguered farms.

Following several abortive local efforts to finance adrainage system, a joint Bureau of Reclamation–stateengineer commission proposed a solution: a compre-hensive plan for drainage, flood control, and channel

rectification, complete with a (180,000 acre-foot) stor-age dam, and a consolidated series of diversion damsand main canals to replace the primitive diversionsand ditches. Because such a project promised to beextremely costly, farmers hoped that it might befinanced through the federal Reclamation Fund.However, because New Mexico had already received asubstantial share of such funds in the lower river,another mechanism would have to be found.

An intensive lobbying effort by Albuquerque andrural leaders convinced the state legislature to approvethe Conservancy District Act of 1923. Districts createdpursuant to this act were to be organized and admin-istered by a state district court, upon petition by 100landowners. After two petitions to the district court,the Middle Rio Grande Conservancy District was suc-

cessfully organized in 1925. The Middle Rio GrandeConservancy District was to serve all lands in thefloodplain of the Rio Grande between Cochiti and SanMarcial and thus could add its assessments for floodprotection to the property taxes collected from resi-dents of Sandoval, Bernalillo, Valencia, and SocorroCounties. The 130,000-plus acres projected to receiveirrigation water from the district would be levied addi-tional assessments to construct and maintain thoseworks. Included in the conservancy district were28,500-plus acres of Pueblo Indian lands, for whichCongress appropriated more than $1.5 million tocover construction costs on Indian lands.

At the outset, some of the district’s intended benefi-ciaries opposed creation of the district, and manyremained suspicious of its subsequent arrangements.The Middle Rio Grande Conservancy District was a

new and powerful political subdivision of the state,with extensive powers to make regulations, levy taxes,condemn and own lands and water rights, salvagewater, remove or relocate structures, fill lands, retardsilt, re-engineer stream channels, construct drains,dams, levees, canals, roads, bridges, stream gages, andelectric power plants. Its water rights were to beexempt from forfeiture under state law or taking byother political subdivisions. As it condemned existingacequias, the conservancy was required to supply itsparciantes (shareholders) with the water entitlementsto which they had become accustomed.

In 1928 the Middle Rio Grande ConservancyDistrict submitted its “Official Plan for Flood Control,Drainage and Irrigation” to the district court.Construction soon began on four diversion dams,

The same bend in the Rio Grande near San Acacia in 1905 (left),showing a broad channel, flood-swept sandbar, and large wetlandin distance, and again in 1989 (right). By 1989 the river flowedonly partly in its native channel, with a levee, riverside drain,

and main canal joining the Santa Fe railroad along its neatlyengineered course. A thicket of invasive salt cedar now coversthe floodplain and confines the river channel, which has nar-rowed by 200 feet and aggraded 15 feet in the 84-year interval.





their connecting main canals, a valley-wide system ofriverside and interior drains, and El Vado Reservoir.Initial tax assessments appeared to be substantialenough to satisfy court-appointed appraisers that thedistrict could service the bonds it let to finance theestimated $12 million cost of construction.

Unintended consequences from the conservancydistrict’s project were substantial. Water supply tosome ditches was interrupted during construction,reducing their parciantes’ ability to farm for one ortwo years. Several thousand acres of farmland werecondemned for rights of way to the drains and canals.A number of irrigators failed to make their annualassessment payments, resulting in foreclosure of some34,000 acres by the state. Other ratepayers felt thatthe original glowing promises of project benefits hadbeen overstated. Certainly, the flood control works didnot prevent the devastation of the Socorro division bythe 1937 and 1941 floods. Siltation and aggradation ofthe channel continued to plague the river. Additionally,there were, and continue to be, assertions that thebroad powers of the Middle Rio Grande ConservancyDistrict inhibited the state engineer’s authority toadminister individual water rights priorities.

At least some of the project’s benefits were realized: asmany as 20,000 acres were drained and (at least tem-porarily) reclaimed for farming. The construction of ElVado Reservoir succeeded in extending late-seasonwater supply to the district’s four irrigation sections.

Further bedeviling the Middle Rio Grande valley, aninterregional conflict erupted when the district began tofill El Vado Reservoir for the first time in 1935. Texassued New Mexico and the Middle Rio GrandeConservancy District in the U.S. Supreme Court, claim-ing that the defendants were impairing the water sup-ply in Elephant Butte, in violation of the 1929 compact.This litigation was dismissed after the 1938 compactwas signed by the three states. By 1941 Congress wasauthorizing the Corps of Engineers to study the still-unmet drainage and flood control needs of the region.

Nor did the district quite succeed in rescuing itsfarmers from financial woes. In danger of defaultingon the bonds that financed the project, the districtappealed to Congress in 1948 for relief of its debtsand the rehabilitation and further improvement of itsdam, diversions, and levees. The Flood Control Act of1948 authorized a Middle Rio Grande Project throughthe U.S. Army Corps of Engineers and appropriated$15 million through the Bureau of Reclamation toprovide the district with debt relief and another roundof middle valley “improvements.” These improvementsincluded 300,000 jetty jacks to straighten and confine

the river channel. Reclamation was to hold the titles tothe water rights and capital works as security forrepayment of this crucial federal investment.

The focus of the Middle Rio Grande Project was pri-marily on the worsening siltation problems. It gavethe Bureau of Reclamation the authority to maintainan open river and funds to channelize 127 miles ofriver from Velarde to Elephant Butte, resulting in theplacement of over 300,000 jetty jacks to confine theriver channel over the next 20 years. The FloodControl Act of 1950 authorized more than $50 mil-lion to the Corps of Engineers to construct flood andsediment control reservoirs, the crux of a strategy toradically reduce sediment inputs to the valley. AbiquiuReservoir on the Rio Chama and Jemez Canyon Damwere completed for this purpose in l954. Constructionof the additionally contemplated sediment and floodcontrol reservoirs was deferred until a Rio GrandeReservoir Regulation Plan was negotiated to the satisfac-tion of Colorado and Texas. In 1965 the corps beganwork on its own Middle Rio Grande Project, complet-ing Galisteo Dam in 1970 and Cochiti Dam in 1975.

Cochiti has been a particularly significant develop-ment in the history of the Rio Grande. Its location onCochiti Pueblo lands displaced much of the floodplainfarming that was culturally and economically criticalto the tribe. During its construction, a sacred siteimportant to Cochiti and its neighboring Pueblos wascarelessly destroyed. Then, when the reservoir filled,seepage below the dam waterlogged what remained ofthe Pueblo’s farmlands. One intended benefit, sedi-ment abatement in the middle valley, resolved itself

Time series of Rio Grande channel forms near San Acacia, show-ing a dramatic narrowing and incising of the river channel.

Mexico to become compact compliant. The projectbegan diverting water from the San Juan River basininto Heron Reservoir in 1972, and started releases in1974. One of its original stated purposes was to pro-vide 42,500 acre-feet per year to “replace previous andanticipated [Rio Grande] basin depletions caused bymiscellaneous uses.” San Juan–Chama Project water hashelped ease the state’s chronic non-compliance, sendingan average of 30,000 annual acre-feet downstream,effectively offsetting water sucked from the river byground water pumping.

During the same period, Colorado’s accrued debitswelled to almost 1,000,000 acre-feet. In 1966 NewMexico joined Texas in a Supreme Court suit thatresulted in an agreement by Colorado to begin toreduce its huge deficit. Across-the-board curtailment ofSan Luis Valley irrigation forced farmers to conserveand reduce their water applications.

The federal Closed Basin Project, which began salvaging shallow ground water in 1984, was designed toreduce the state’s accrued debit. Nevertheless, it was notuntil Elephant Butte Reservoir spilled in a very wet 1985that Colorado’s 30-year-old water debt was forgiven.

Over the past 20 years, with a blessing of abundantsnowpacks, both Colorado and New Mexico havemaintained compliance with the water delivery require-ments of the Rio Grande Compact.

FINALE

Stretched thin by the dizzying pace and magnitude ofwater development, the Rio Grande/Rio Bravo basinremains enmeshed in a perpetual conundrum: there aresimply more claims to Rio Grande water than the rivercan reasonably be expected to provide. Its core ecology,the very structure, and function of the river have beenprofoundly altered, with unfortunate outcomes: It ceasesto flow at the behest of unrestrained economics; its lev-eed banks armor continuously narrowing and aggradingchannels, disrupting the conveyance of water; biodiversi-ty continues to decline. Successive engineering projectshave disrupted the productivity of its adjoining lands,now beset by invading plants and, consequently, fire.

Another chapter of the saga will be written by thepresent generation. Access to the river by growing,thirsty urban populations and the emergence of concernfor the fate of Rio Grande ecosystems have joined theperennial contenders for the limited supply. The planneduse of San Juan–Chama Project water by Albuquerque,Santa Fe, and other communities suggests that a chal-lenging new version of the intricate old balancing actlies just ahead.

poorly as sediment-starved waters began to progres-sively scour the river channel downstream. CochitiReservoir’s position athwart the mouth of White RockCanyon isolated the river below from a naturalrefugium for aquatic species, contributing to thedecline and endangered status of the Rio Grande sil-very minnow. It has, however, kept its flood controlpromise, intercepting potentially damaging floods in1979, 1985, and 1995.

RECENT TIMES

The Rio Grande Compact, with its cornerstones ofsound science and frank if difficult negotiations, hasserved to moderate the consequences of the RioGrande’s century-and-a-half-long development orgy.The compact, first administered in 1940, is the foun-dation of today’s “law of the river,” which also includesa welter of contracts between special water districts andthe Bureau of Reclamation, water rights administrationin three states, and the decisions made by thousands ofindividual water users and their districts.

Unfortunately, both Colorado and New Mexico havefound that they cannot always reliably comply withthe compact’s downstream delivery requirements. Andso, when threats of harm cannot otherwise be recon-ciled, the courts are standing by.

During the severe drought of the 1950s the NewMexico State Engineer and the Middle Rio GrandeConservancy District again found themselves beforethe U.S. Supreme Court to answer for a water debitthat had exceeded its 200,000 acre-foot limit allowedby the compact. The proximate cause of this actionwas a renewed assertion that New Mexico was storingwater in El Vado Reservoir in violation of the RioGrande Compact. The U.S. Supreme Court dismissedthis litigation, ruling that Texas had failed to name anindispensable party, the United States in its capacity astrustee for the six Middle Rio Grande Pueblos andtheir water rights.

By 1956 New Mexico’s debit had grown to more than500,000 acre-feet, before rehabilitation of waterloggedlands undertaken by the Middle Rio Grande Projectcould produce additional water flows. The Low FlowConveyance Channel, which began operating in 1953in an attempt to make New Mexico’s compact deliveriesmore reliable, provided a bit of the long soughtdrainage objective. The channel did, in fact, producewater and reduced the state’s compact debit, but todayit serves mainly to drain a Rio Grande channel percheddozens of feet above the surrounding floodplain.

The San Juan–Chama Project also helped New





One of the most startling paradoxes of the world’s dry-lands is that, although they are lands of little rain, thedetails of their surfaces are mostly the products of theaction of rivers. To understand the natural environ-ments of drylands, deserts, arid, and semiarid regionsof the earth is to understand the processes and formsof their rivers.

—William L. Graf, Fluvial Processes in Dryland Rivers

R ivers are the primary conduits by which waterand the products of continental weathering are

delivered to the oceans. Flowing water provides theenergy that moves the weathered soil and rock materi-al, along with organic matter that is important to theecosystem, downstream through the river system. Thewater is derived from precipitation that falls on theearth’s surface and continues through the hydrologiccycle in a variety of forms, including surface runoff,infiltration into the ground water system, and evapora-tion and transpiration back into the atmosphere whereit contributes to later precipitation. At many locations,much of the water that appears as river flow is derivedfrom the ground water system, which provides storagethat sustains stream flow during dry periods. In riverswith mountainous headwaters, winter precipitation istemporarily stored as snow until the seasonal tempera-tures rise above the melting point, releasing it into thelocal stream system.

The relative contribution of each of the above factorsto the flow at any particular location and time variesconsiderably. Intense local storms can cause extremechanges in stream flow over very short time periodson the order of minutes to hours, whereas seasonalchanges in temperature and precipitation result in pre-dictable patterns in stream flow over the course of theyear. Longer-term climate variability causes both cycli-cal variability and random fluctuations in the amountof stream flow over a few to many years. Global warm-ing may also affect long-term stream flow patterns.

In addition to stream flow that is mostly climatedriven, the physical characteristics of rivers are strong-ly influenced by the basin geology that controls thecharacter of the sediment in the river bed and banks,and provides structural controls on the alignment,

slope, shape, and dynamic behavior of the river. Fromthe human perspective, these characteristics are criticalbecause they affect the capacity of the river to safelycarry flood flows and its ability to continue carryinglower flows when the water is essential to ecosystemhealth and water supply needs. The dynamic (erosion-al and depositional) behavior of the river is alsoimportant because it affects the safety of infrastructure,particularly during high-flow periods.

Because water is key to nearly all physical and bio-logical processes, river corridors are a critical compo-nent of the natural ecosystem. This is nowhere moretrue than in arid regions, where the health of theriparian corridor is tied directly to the amount andtiming of flow that mostly originates from outside thelocal area. Natural stream flow patterns in arid riverstend to be more variable than in temperate- or humid-zone rivers, and the species that are found in aridrivers generally have adapted to this high variability.Human use of the water often alters the flow patternsand thus the health of the ecosystem. Understandingthese effects and managing water use to supporthuman needs and protect public safety during flood-ing, while minimizing ecosystem impacts, are keychallenges for river managers.

HYDROLOGY OF THE RIO GRANDE

All of the above factors that affect stream flow areimportant in the Rio Grande basin. The Rio Grande,originating in the mountains of southern Colorado, isthe fifth longest river in North America, with a totalcontributing drainage area of about 176,000 squaremiles, including the arid lands of New Mexico, Texas,and northern Mexico. Annual precipitation in part ofthe upper basin averages nearly 40 inches per year,whereas much of the middle and lower basin receivesless than 10 inches per year. Typical of arid rivers, theRio Grande tends to lose flow in the downstreamdirection, and did so even under native (pre-Europeansettlement) conditions, due to infiltration and evapo-transpiration. Upstream water use and to a lesserdegree increased evapotranspiration losses caused bythe proliferation of salt cedar and other non-nativevegetation now cause river flows to decrease in thedownstream direction even more so than under native

An Introduction to Rivers of the Arid Southwest

Robert A. Mussetter, Mussetter Engineering, Inc.


conditions.The annual water volume of the Rio Grande that

flows past the Otowi stream gage above CochitiReservoir averaged about 1.1 million acre-feet undernative conditions, and the current long-term averageis about the same. The volume varies significantly

from year to year. The lowest recorded volume atOtowi since 1895 was about 360,000 acre-feet (1904),about one-third of average, and the highest recordedvolume was about 2.4 million acre-feet (1940), about220 percent of average. Over the more recent six-yearperiod (water years 2000 through 2005), the average


The Rio Grande watershed. The Rio Grande drainsareas of three states and parts of northern Mexico.

Below El Paso the river forms the internationalboundary between the U.S. and Mexico.



less than the long-term average.Native flow volume estimates unfortunately are not

available for the middle and lower reaches. Since1973 when Cochiti Dam was completed, the amountof water passing Albuquerque averaged slightly lessthan 1 million acre-feet per year, and this decreases toabout 0.73 million acre-feet below Caballo Dam. Thepost-Cochiti average at Albuquerque was about 40percent greater than the 30-year average from theearly 1940s to early 1970s. The difference betweenthe two periods is due to both direct human influence(e.g., imported flows from the San Juan–ChamaProject, completed in 1971), and natural causes (i.e.,extended periods of below average precipitation in the1940s through mid-1970s and above average precipi-tation in the 1980s through mid-1990s.)

Under both native and current conditions, the bulkof the annual flow volume is derived from snowmeltfrom the upper basin; thus, the highest sustainedflows typically occur during late spring and early sum-mer. Prior to significant water development, about 60percent of the annual native volume in the upperreaches occurred during April through June, and lessthan 15 percent occurred during July throughSeptember. Seasonal patterns in the downstream

reaches were probably similarto the upstream reaches.Human-induced factors,including the water storagesystem, imported flows,urbanization and channeliza-tion, and increased evapo-transpiration as a result ofnon-native vegetation, nowaffect the seasonal flow pat-terns, with the magnitude ofthe effects increasing in thedownstream direction. In theupper and middle reaches,about half of the annual flowvolume now occurs duringApril through June, and theJuly through September per-centages have increased to15–20 percent. In contrast,only about 40 percent of theannual flow below CaballoDam now occurs during Maythrough June, but the amountduring July throughSeptember represents about40 percent of the annual vol-

ume. The reasons for this difference are complex, butthey are mostly driven by upstream, man-inducedchanges. Late summer, monsoon season runoff fromthe middle and lower basin is also a factor.

GEOMORPHOLOGY OF THE RIO GRANDE

The geomorphic characteristics of rivers represent theintegrated effects of the physical factors in the basinand drainage network that include the quantity andtiming of flow (i.e., the hydrology), drainage basingeology, riparian vegetation, and human modifica-tions. Rivers tend to adjust their gradient, plan form(their down-valley and cross-valley alignment), crosssectional shape, and the sediment size toward a stateof dynamic equilibrium (or long-term balance) withthe upstream water and sediment supply. In thisregard, it is a common premise that alluvial riverstend to adjust their cross sectional size so that thecapacity at which water will spill onto the floodplain(often referred to as the bankfull capacity) corre-sponds to the maximum discharge that occurs for afew to several days each year. Because of the highlysporadic nature of the runoff, however, arid riverssuch as the Rio Grande may be close to equilibrium

Average monthly flows on the Rio Grande, at four locations, as noted.

ing by the jetty jacks and infrastructure.Both the gradient and sediment size tend to decrease

in the downstream direction in most rivers, and theRio Grande is no exception. The Rio Grande fallsabout 12 feet per mile between Velarde and the mouthof the Rio Chama, decreasing to about 4 feet per milethrough the Albuquerque reach and to around 3 feetper mile between El Paso and Fort Quitman in thelower reach. Similarly, the typical bed materialupstream from Cochiti Reservoir is mostly gravel andcobbles, changing to mostly sand at and downstreamfrom Albuquerque. Even the sand decreases in sizemoving downstream from Albuquerque toward

Elephant Butte. The interaction of these particles withthe flow significantly affects the dynamic behavior ofthe river. In gravel and cobble bed reaches, relativelyhigh flows are required to move the material andadjust the shape and elevation of the channel bed;thus, the bed is static most of the time. In contrast,flows of nearly all magnitudes can move sand-sizedparticles; thus, the channel bed is under a state of con-stant adjustment in these reaches.

Historically, the middle and lower reaches of the RioGrande carried one of the highest sediment loads ofany river in the world, and much of this sediment

for only a very small percentage of the time; thus, thein-channel capacity can vary considerably from theannual peak flows.

High flows that occur on an annual or less frequentbasis are the most important in forming and maintain-ing the river channel. Erosion, transport, and deposi-tion of sediment during these high flows are, in turn,key to maintaining certain aspects of the ecosystem. Inriver systems like the Rio Grande, high flows resultfrom both snowmelt runoff and runoff from intenserainstorms in the lower basin. Although the storm-driven flows can be as high, or higher than, thesnowmelt-driven flows, the latter tend to occur formuch longer durations;thus, they do more work inshaping and modifying thechannel. Before constructionof Cochiti Dam, the annualmaximum snowmelt-drivendischarge near Albuquerqueexceeded 7,000 cubic feetper second (cfs) in aboutone of every two years, onaverage. With the upstreamflood-control system, thismaximum discharge is nowonly about 5,400 cfs.During the pre-dam period,peak discharges as high as42,000 cfs occurred in theAlbuquerque reach and100,000 cfs near SanMarcial, causing widespreadflooding. The highest peakflow in the Albuquerquereach since completion ofCochiti Dam was 9,500 cfsin 1984, and the maximumflows rarely exceed 8,000 cfs.

In the Albuquerque reach,flow begins to spill into the bosque at 5,000 to 6,000cfs, which is consistent with the current typical annualpeak flow. Under native conditions, the in-channelcapacity was much more variable both spatially andtemporally than today due to channel enlargementduring infrequent high flows and re-adjustment dur-ing longer duration low- to-moderate flow periods.The active channel now is more consistent in sizealong the reach and varies less in size over time thanunder native (pre-European settlement) conditionsdue to dampening of the extreme flows by the flood-control system and lateral controls on channel widen-

Average width of the active channel of the Rio Grande at different periods, from 1918 to 2001(Data from the Bureau of Reclamation).




consisted of very fine silt- and clay-sized particles.Trapping of sediment by Cochiti and other upstreamdams has significantly reduced the concentrations.Upstream from Albuquerque, concentrations are nowonly 2.5 to 3 percent of their historic values. Inflowsfrom local tributaries with fine soils in their water-sheds cause the concentrations to increase as onemoves downstream from Cochiti toward Elephant Butte.

The fine material had a profound effect on thedynamic behavior of the river under historic condi-tions, and this effect continues today. During flooding,the silts and clays were carried into the overbankswhere they settled out onto the valley floor, buildingup the floodplain and river banks with cohesive, diffi-cult-to-erode sediment. As a result, the river banks arerelatively erosion resistant, limiting the ability of theriver to erode laterally. Because the fine sedimenttends to have high water-holding capacity and is richin nutrients, the overbanks provided excellent habitatfor riparian vegetation that, historically, consisted of avariety of species including cottonwood. These char-acteristics are also very suitable for the more recentlyintroduced non-native species such as salt cedar; thus,they are both a blessing and a curse with respect tothe health of the riparian zone.

Under native conditions, the Rio Grande throughthe middle valley was relatively wide and braided withmultiple smaller channels separated by active sand-bars. In the early 1900s the active channel averagedmore than 1,200 feet wide between Bernalillo and themouth of the Rio Puerco. The river subsequently nar-rowed to only about one-third of its historic widthdue to the combined effects of the jetty-jack fields,other human encroachments, and reductions in thepeak flows. The river has continued to narrow insome locations since the 1960s, but at a much slowerrate, and it is likely approaching equilibrium with theregulated flow regime.

The river has undergone other important changesdue to human activities that include changes in size ofthe bed material and changes in the elevation of thechannel bed relative to the floodplain. The reachdownstream from Cochiti Dam, for example, hasexhibited the classic response to sediment trapping ina reservoir by coarsening the bed material and down-cutting as the river made up for the deficit in sedi-ment supply by mining material from the bed. Beforeconstruction of the dam, the bed material betweenCochiti and Angostura was mostly sand, with 20 to 30percent medium-sized gravel. The sand has now beenlargely depleted, and the bed material is primarilygravel. The bed has also downcut by an average of 2to 4 feet. The gradient restoration facilities that wereconstructed over the past several years on and nearthe Santa Ana Pueblo below Angostura are an attemptto mitigate the effects of this downcutting. The oppo-site effect has occurred at the head of Elephant ButteReservoir, where the flatter river gradient due to thereservoir has caused sediment deposition, raising theriver bed by several tens of feet. The limited hydrauliccapacity of the San Marcial Railroad Bridge is an obvious manifestation of this process.

The water development system in the Rio Grandebasin is critically important in sustaining water supplyand protecting public safety, but this system haschanged the physical characteristics and dynamicbehavior of the river. Although many of these changesprovide positive benefits to our ability to use the riveras a resource, they have also had unintended andundesirable consequences to both the health of theecosystem and our ability to protect critical infrastruc-ture. River managers face a difficult challenge in bal-ancing the costs and benefits of maintaining the sys-tem, while finding ways to mitigate the unintendedand undesirable effects.


Oblique aerial view of the Rio Grande in the early1960s, soon after completion of a jetty jack field anddredging of a pilot channel.


The surface water that we see flowing in the RioGrande, in irrigation canals, and in drains is inti-

mately linked to the underlying ground water. Thesurface water and the ground water in fact form oneintegrated system.

The ground water is not some mysterious lake orstream that’s flowing below us out of sight; instead,ground water is simply water held in the spacesbetween grains of sand, clay, or gravel. In the soil,some of the empty spaces are also filled with air; deep-er, the air is displaced, and the pores are totally water-filled. This is the ground water zone. If you dig a holealong the banks of the Rio Grande the hole willremain open and air-filled until a certain depth. Then,as you dig the hole deeper, it will fill with water tosome level below the ground surface. This water sur-face is the level of the ground water table. Below this,you’re in the ground water aquifer.

The Rio Grande valley between San Acacia andElephant Butte Reservoir can be envisioned as a sand-and gravel-filled bathtub with a shallow trench run-ning down the middle. The sides of the bathtub arethe uplands and mountains rising to the east and westof the river; the bottom of the bathtub is formed bydeep, compacted sediments. At the north end of thetub water flows into the trench (the Rio Grande)through the diversion dam at San Acacia. At the southend the Rio Grande flows into Elephant ButteReservoir and eventually “drains” from the tubthrough Elephant Butte Dam into the Mesilla Valley.

As water flows down the Rio Grande some of itseeps through the river bed and spreads out under-ground. But it can only seep so far before it hits therelatively impermeable sides and bottom of the tub. Sothe bathtub gradually fills with water, the sand andgravel become saturated from the bottom up, and thewater table rises. Once the water table reaches thebase of the trench the seepage is reduced, and more ofthe water entering the valley at San Acacia stays in theriver bed all the way to Elephant Butte.

Of course, the bathtub model is an oversimplifica-tion of the actual system; some river water can leakout the sides and bottom of the tub, and some waterenters the tub from sources (such as rising geothermalwaters) other than the river. But in general the water

that enters the valley stays in the valley until it reachesElephant Butte. One exception, however, is the waterconsumed by evapotranspiration.

Evapotranspiration is the water lost to the atmos-phere by evaporation from the soil and open water,and by transpiration of water through the leaves ofplants. Another term for evapotranspiration is “con-sumptive use”—the water that is “consumed” and nolonger available in the surface water/ground water sys-tem. To a first approximation, the only water thatleaves the valley between San Acacia and ElephantButte is that which is evapotranspired.

The pie chart on this page shows the average annualconsumptive use of water between San Acacia and SanMarcial, upstream from where the Rio Grande entersElephant Butte Reservoir. More than half of the con-sumptive use in this reach is evapotranspiration byriparian vegetation along the river, and almost a thirdis evapotranspiration from agricultural crops. In total,about 10 percent of the water flowing into the valleyat San Acacia is lost to evapotranspiration by the timethe river reaches San Marcial. Since most of the dis-solved salts remain in the water during evapotranspi-ration, the salt concentration in the river and in theground water generally increases as you move down-stream. Small inputs of high-salinity water from natu-ral sources other than the Rio Grande also make sig-nificant additions to the salt load of the system.

Until recent times the Rio Grande was the only

The Surface Water/Ground Water Connection

Robert S. Bowman, New Mexico Institute of Mining and Technology

Estimated average annual consumptive use of water betweenSan Acacia and San Marcial. Data provided by Nabil Shafike ofthe New Mexico Interstate Stream Commission.




trench running the length of the valley. However, inresponse to the drought of the early 1950s and the pre-cipitous drop in Elephant Butte Reservoir, the Low FlowConveyance Channel (LFCC) was constructed to reduceseepage losses as water moved down the valley. After1959 some or all of the water from the Rio Grande wasdiverted to the LFCC at San Acacia Dam and traveledapproximately 75 miles to the head of the reservoir.

During the wet period of the mid-1980s ElephantButte Reservoir filled to capacity; by 1987 there wasno need to divert water from the Rio Grande into the

LFCC. Elephant Butte receded during the 1990s, butsedimentation during the wet years plugged the chan-nel and left the LFCC disconnected from the reservoir.Today the LFCC no longer flows into Elephant Buttebut instead discharges into a large “delta” area at thenorth end of the reservoir, where it supports a densepopulation of salt cedar and other phreatophytes.Budgetary constraints along with the need to maintainminimum flows in the river (e.g., to support the sil-very minnow population) have slowed progress in re-engineering the connection between the LFCC andElephant Butte Reservoir.

The average bed elevation of the LFCC is below thatof the bed of the Rio Grande, and generally lies belowthe level of the ground water table—it represents anew, deeper trench in the bathtub. Since the LFCC isthe topographic low point in the valley, water tends to

flow underground toward it from the east and thewest. The LFCC thus currently acts as a surface drain.

The approximately 90 percent of the water remain-ing in the valley after evapotranspiration losses canmove interchangeably between the surface and groundwater systems. As mentioned above, water generallyseeps from the Rio Grande to the shallow aquifer as ittraverses the valley from north to south. But thisground water can appear once again as surface waterin the LFCC. This surface water-to-ground water-to-surface water interchange is shown schematically in

the illustration on this page. Just as water spilled on asloping desk runs toward the bottom of the desk, waterflows underground from a high point on the water tableto points where the water table is lower. Ground watertends to flow from below the Rio Grande west into theLFCC, and from below agricultural fields east into theLFCC. Thus, even though the Rio Grande is no longerdirectly diverted into the LFCC, much of the river’sflow still ends up as surface water flowing towardElephant Butte via this large drain.

The illustration above also points out many of theother important interactions between surface waterand ground water, with an emphasis on gains andlosses to the ground water. Water seeps from the RioGrande as it flows southward, recharging the shallowground water aquifer. Some of this water reemerges assurface water in the LFCC. Water is diverted from the

Schematic cross section of the Rio Grande valley showingthe linked surface water/shallow ground water system,

emphasizing gains and losses to the ground water. TheLFCC gain is, in fact, a loss to the ground water.



river at San Acacia Dam to supply the valley’s farmirrigation system. A portion of the irrigation waterapplied to the fields percolates through the soil torecharge the ground water, and can move under-ground to the LFCC to reappear as surface water.Some of the rain and snow that fall on the watershedcan percolate through the soil and reach the watertable. And some ground water flows into the valleyfrom the Albuquerque Basin to the north.

Simultaneously water is being lost from the system,and hence from the ground water, due to evapotran-spiration. There is direct evaporation from the openwater surfaces of the Rio Grande, the LFCC, and agri-cultural canals and drains. Close to the river, riparianvegetation transpires water to the atmosphere; fartherfrom the river, in the irrigated portion of the valley,crops also transpire water. Water pumped by wells forirrigation and for domestic purposes removes waterfrom the aquifer; much of this water is subsequentlylost from the system as evapotranspiration.

Compared to the hundreds of thousands of acre-feetthat flow down the Rio Grande between San Acacia andElephant Butte in an average year, relatively littleground water is pumped from the aquifer. During thewet period from the early 1970s to the late 1990s,when surface water was plentiful, almost no water waspumped for irrigation. Even during the dry years of thelate 1990s and early 2000s, most of the water for irriga-tion was provided by surface water diversions from theriver. Domestic well pumping in the region amounts toonly a few hundred acre-feet of water per year.

The ground water pumping situation in the San

Acacia reach is in sharp contrast to that ofAlbuquerque and Bernalillo County. In 2005 about100,000 acre-feet of water were pumped byAlbuquerque city wells, and thousands more werepumped by domestic wells.

The ground water table responds very rapidly tochanges in river flow, and flow in the river likewiseresponds quickly to changes in the elevation of thewater table. The strong connection between surfaceand ground water levels is shown in the graph on thispage, which illustrates the water table response toincreasing river flows below San Acacia Dam followinga series of heavy rainfalls in the fall of 2003. Thechanging river flow is shown by the stage (or height)of water in the river relative to sea level. The watertable elevations in wells at different distances awayfrom the river are also shown relative to sea level.

The ground water response to fluctuating river flowsis dramatic. Within a few hours after an increase inriver stage, increased seepage from the river causes thewater table at a well very near the river to rise. Anintermediate well responds later, and the well mostdistant from the river takes several days to fullyrespond. As the flow in the river decreases and theriver stage drops, the water table drops correspond-ingly. In a sense the aquifer “breathes in” water inresponse to increased flow in the river. When the riverflow decreases, the aquifer “breathes out” and returnssome of the water back to the river.

A related but contrasting situation occurs whenground water is pumped from the aquifer by irrigationor domestic wells. Pumping of wells, particularlywhen they are close to the river, causes a drop in thewater table elevation and an increase in the gradientbetween the river and the water table. Analogous tothe gradient between the Rio Grande and the LFCC,this increased gradient causes more water to seep outof the river bed, reducing the flow in the river. Theeffect of a single pumping well on river flow may bedifficult to detect, but the combined pumping ofmany wells can have a measurable effect. The large-scale pumping of the aquifer in the Albuquerque areasince the 1950s has caused the water table to drophundreds of feet in some areas, with a resultantincrease in seepage from the river. This extraction ofground water has far exceeded the ability of riverseepage to replenish it.

As shown by the above examples, the strong inter-connections between surface water and ground watermean any perturbation in one part of the systemaffects the other. Ground water pumping lowers thewater table and increases river seepage, ultimately

Changing water table levels over a one-month period in wellsin response to increasing flows in the Rio Grande below SanAcacia Dam. The nearby well is 102 feet from the river, theintermediate well is 388 feet from the river, and the distant well is 1,224 feet from the river.

22 C H A P T E R O N E

reducing flow in the river. Reducing seepage (e.g., byconcrete-lining of irrigation canals or of the river beditself) reduces recharge to the aquifer and allows thewater table to drop, potentially reducing riparianevapotranspiration to the point that undesirablespecies such as salt cedar as well as desirable speciessuch as cottonwood may be unable to survive. Thus,any alteration in river management will affect groundwater dynamics, just as increased ground water pump-ing, particularly if water is exported out of the imme-diate vicinity, will cause changes in the Rio Grandeand the riparian community it supports.



T he Middle Rio Grande is located in central NewMexico between Cochiti Dam and Elephant Butte

Dam. People have used the Middle Rio Grande and itssurrounding land for centuries for agriculture, grazing,and timber. However, widespread physical alterationsto the river did not occur until recent times. Majorchanges took place during the twentieth century aspeople became concerned about floods, accumulationof salt in agricultural soils, and delivery of waterdownstream to Texas and Mexico. Flood-controldams, levees, and diversion structures were built,

including Elephant Butte Dam (1916), drainage ditchesparallel to the river channel (1920s), levees (1950s),and Cochiti Dam (early 1970s). In addition, urbandevelopment extended closer to the river.

Before regulation the Middle Rio Grande flowedthrough a network of channels separated by islands.Channels changed position frequently over the river’ssandy foundation, and this movement maintained adiversity of habitats. Aquatic habitats included themain channel, side channels of shallow, slowly movingwater, as well as ponds and marshes. Terrestrial habitats

included patches of forests of various ages (rangingfrom recently established to mature), shrubs, herba-ceous plants, and grasslands. Islands in the river chan-nel contained a mix of semi-aquatic habitats and earlysuccessional vegetation. This diversity of habitats oversmall areas provided a mix of resources for plants, ani-mals, and microbial communities. In addition, the riverflooded predominantly in spring, as snow melted frommountains in northern New Mexico and southernColorado, and in late summer during monsoon-seasonstorms. These floods increased soil moisture, cleared

vegetation from river banks, deposited sediment fromupstream, altered channel structure, decomposed organ-ic matter, and distributed aquatic organisms and seeds.

Regulation disconnected the river from its floodplain.Installation of levees constrained the river to a singlefloodway, and side-channels, wetlands, and pondsnearly disappeared. Dams eliminated large floods,increased flows in the river during summer, trappedsediment, and produced barriers to movement ofaquatic organisms. Without floods, dense forests devel-oped along the river because scouring flows were

Ecology of the Middle Rio Grande of New Mexico

Mary J. Harner and Clifford N. Dahm, University of New Mexico

Aerial photos highlighting the channels of the Rio Grandein Albuquerque near the Rio Grande Nature Center. Photosdepict the river in 1947, prior to regulation, with overbankflooding (note colonization of young cottonwoods in vicin-ity of side channel); in 1959, following the 1950s drought

and immediately after the construction of levees (whichparallel the channel) and jetty jacks (perpendicular to thechannel); and in 1996, with a simplified river channel anda contiguous, mixed forest of cottonwood and non-nativespecies. Photos depict a 2-km stretch of river.



unavailable to remove vegetation. The dominance ofnon-native plants increased throughout the riparianareas. Wood and leaves accumulated on the forest floorbecause water was unavailable to decompose andtransport it from the forest. The combination of densevegetation, large accumulations of wood and decayingleaves, and lack of wet soil increased the size and fre-quency of wildfires. Continued growth of the humanpopulation led to more agricultural and urban flood-plain development, increased inputs of treated wastewater to the river, and increased use of river andground water.

VEGETATION

Riparian forests in low-lying regions along the MiddleRio Grande are locally known as “bosque,” the Spanishword for woodland. The native bosque contains forestsof Rio Grande cottonwoods, with shrubs and herba-ceous plants growing beneath the cottonwoods. Nativewoody shrubs along the Middle Rio Grande includeGoodding and coyote willows, New Mexico olive, bac-charis, and false indigo bush. Sedges, rushes, cattails,and yerba mansa grow in moist soils. Plants more toler-ant of drier and saltier soils, such as mesquite and saltgrass, live on higher or disconnected surfaces.

Cottonwood trees require floods to establish. Theyrelease windblown seeds in spring at the time whenthe river naturally flooded. Cottonwood seedlingsneed direct sunlight, and their roots must touch wetsoil. Floods create sites for recruitment of cotton-woods by scouring away plants that would otherwiseshade young seedlings and by elevating soil moisture.Ideal conditions for cottonwood establishmentoccurred historically once every 5 to 10 years. Theseoptimal conditions no longer exist. Rio Grande cot-tonwoods have not reestablished themselves over largeareas since the early 1940s, following the last largefloods. Therefore, most Rio Grande cottonwoods are60 years old or older. Young trees are not replacingmature cottonwoods because large floods have beeneliminated.

Non-native plants that do not require such preciseconditions for recruitment are replacing cottonwoodsalong the Middle Rio Grande. Salt cedar and Russianolive are common non-native trees in the bosque.They release seeds throughout the summer and cangrow under the shade of other vegetation. Whethersalt cedar uses more water than native plants orincreases salt in soil is a focal area of study. Measure-ments of water loss along the Middle Rio Grandeshow that forests containing a mixture of cottonwoods

and non-native plants use the most water. Forests ofonly cottonwood, only dense salt cedar, or onlyRussian olive use slightly less water than cottonwoodforests with a non-native understory. People areremoving non-native plants over large areas to possi-bly conserve water, reduce the risk of wildfires, andencourage growth of native plants.

MICROBIAL COMMUNITIES

Microorganisms, such as bacteria and fungi, play criti-cal roles in ecosystems. They decompose organic mat-ter, such as leaves and wood, and transform nutrientsinto forms that are available for uptake by algae andplants. Flooding stimulates microbial activity, andabundances of bacteria and fungi are higher at sitesthat regularly flood. Some fungi, known as mycor-rhizal fungi, live on or within the roots of plants.These fungi provide nutrients to plants in exchangefor energy (carbon) from plants. Roles of soil fungi inriparian ecosystems are a current area of researchalong the Middle Rio Grande. Some researchers seekto understand whether fungi and other microorgan-isms should be added to soil to promote plant growthduring restoration projects aimed at reestablishingnative vegetation following disturbances, such as fire.

INVERTEBRATES

Hundreds of species of invertebrates inhabit theMiddle Rio Grande. Aquatic invertebrates include var-ious species of mayflies, stoneflies, caddisflies, midges,and true flies. Aquatic invertebrates use leaf andwoody debris for habitat, and they commonly live inshallow, low-velocity channel edges and backwaters.Today, lack of habitat limits the abundance of someaquatic invertebrates in the Middle Rio Grande. Inaddition, degradation of water quality negativelyaffects some species. For example, several species ofmollusks inhabit isolated springs along the Middle RioGrande. They are sensitive to changes in their envi-ronment, and several species are now listed as endan-gered. In contrast, the introduced Asiatic clam, whichcan tolerate degraded conditions in the river, is nowfound throughout the Middle Rio Grande.

Terrestrial invertebrates also have important roles inthe bosque ecosystem. Some terrestrial invertebratesbreak down organic matter by chewing it into pieces.Two non-native terrestrial isopods (pill bugs) are dom-inant decomposers of organic matter in these forests.Other terrestrial invertebrates feed upon leaves in thecanopy of trees, and some prefer stressed cotton-



woods. Flooding has been shown to affect the compo-sition of invertebrates. Crickets, a native decomposer,tend to increase in abundance when soil moisturerises. Abundance of carabid beetles increases at sitesthat continue to flood, and their densities might serveas indictors of hydrologic connectivity between theriver and the forest.

FISH

Historically the Rio Grande in New Mexico contained17 to 27 species of fish, including big river fishes suchas longnose gar, shovelnose sturgeon, and Americaneel. Reduced river flows and increased sedimentationled to the extirpation of many fish. Elephant ButteDam stopped upstream migration of large species.Operation of Cochiti Dam reduced water tempera-tures, sediment loads, and habitat complexitythroughout the Middle Rio Grande. During recentyears, drying of the main river channel killed fish andreduced their migrations. Many native fish have beenlost from Rio Grande, and 13 to 19 non-native speciesof fish have been introduced, including the commoncarp and white sucker. Only one native minnowspecies remains, the Rio Grande silvery minnow. Itonce lived throughout the Upper and Lower RioGrande, as well as the Pecos River basin. Today thesilvery minnow lives only between Cochiti Dam andElephant Butte Reservoir.

AMPHIBIANS & REPTILES

Amphibians, which live part of their life cycle inwater, once thrived in wet meadows, marshes, andfloodplain ponds along the Rio Grande. Amphibiansliving along the Middle Rio Grande include: Couch’sspadefoot toad, Woodhouse’s toad, great plains toad,and northern leopard frog. Some native amphibians,especially the northern leopard frog, have been nega-tively affected by reductions in availability of flood-plain pools, as well as by predation by introducedbullfrogs. Common reptiles that live along the MiddleRio Grande include the eastern fence lizard, NewMexico whiptail lizard, spiny softshell turtle, andcommon garter snake.

MAMMALS

Several species of large mammals, including grizzlybears, jaguars, and gray wolves once inhabited the RioGrande valley but are no longer present. People alsodepleted populations of beavers during the nineteenth

century. Beavers were restocked from 1947 to 1958and now maintain healthy populations in riverbanks.Rock squirrels and valley pocket gophers also livealong the Middle Rio Grande. Pocket gophers burrowin floodplain soils, and their activities move deep soilto the surface, thus increasing the cycling of nutrientsand movement of soil microbes. Small mammals alongthe Middle Rio Grande include the white-footedmouse, house mouse, tawny-bellied cotton rat, west-ern harvest mouse, hispid cotton rat, white-throatedwoodrat, Ord’s kangaroo rat, and piñon mouse. Manysmall mammals prefer grassy areas of the floodplain.The white-footed mouse, which often nests in cavitiesof trees, avoids drowning during floods by climbingtrees. Three species of bats commonly roost under oldwooden bridges along the Middle Rio Grande. Thesebats are primary consumers of night-flying insects.

BIRDS

Riparian forests in the desert Southwest provide nest-ing and foraging habitats for resident and migratorybirds. Riparian areas often have high numbers of avianspecies, and this has been documented in woodlandsand marshes along several rivers. More than 270species of birds use habitats along the Rio Grande,and many breed along the river. Birds of the MiddleRio Grande include Bewick’s wren, great blue heron,black-chinned hummingbird, white-crowned sparrow,downy woodpecker, and great horned owl. Tens ofthousands of waterfowl, such as snow geese and sand-hill cranes, spend the winter along the Middle RioGrande, especially in the areas between Bernardo andthe Bosque del Apache National Wildlife Refuge.Many of these birds feed on alfalfa and corn through-out the river valley. Birds bring nutrients from adja-cent terrestrial areas to wetlands at night while theyroost in the safety of shallow water or islands.Chemical signals of corn and alfalfa, which can beidentified by stable isotopes, appear throughout thefood web, especially in fish and crayfish.

Habitat loss via reduction of wetlands and decreasesin forests with multiple layers of vegetation has con-tributed to declines in some species of birds along theRio Grande. Birds use a variety of riparian habitats fornests. Some build nests on the ground or in shrubs,others use the canopy of trees, and some nest withincavities of trees. The endangered southwestern willowflycatcher relies upon dense shrubs along waterwaysfor nests. Introduced European starlings, which nestin tree cavities throughout the bosque, compete withnative cavity nesting birds. Bird use of native versus



non-native plants along the Middle Rio Grande is anarea of current research. Some research suggests thatriparian forests with a mix of native trees and shrubsof different sizes have the greatest diversity of birds.

PEOPLE

The Rio Grande and its bosque have providedresources for humans for thousands of years. NativeAmericans living in pueblos used water from theMiddle Rio Grande for irrigation. During the seven-teenth century, Spanish settlers developed a perma-nent system of diversions known as acequias.European settlers used riparian areas for cattle grazingand timber harvesting. Today more than half of NewMexico’s population resides in the Rio Grande basin.Managers seek to balance human demands on theriver with needs of other organisms that rely upon theMiddle Rio Grande. Efforts are underway to restoreparts of the bosque to native vegetation and to reducerisks of catastrophic fires. Managed floods that matchthe timing, but only a fraction of the size, of historicfloods have been used in the Middle Rio Grande todemonstrate the importance of floods to the ecosys-tem. General knowledge of bosque ecology is reachingthe public through programs like the BosqueEcosystem Monitoring Program, a program that bringshundreds of school children to the river each year tolearn about and measure components of the ecosys-tem.

Suggested Reading

Middle Rio Grande ecosystem: Bosque Management Plan, Crawford, C. S.,Cully, A. C., Leutheuser, R., Sifuentes, M. S., White, L. H., andWilber, M. P., U.S. Fish and Wildlife Service, District 2,Albuquerque, NM, 1993.

Middle Rio Grande biological survey, Hink, V. C. and Ohmart, R. D.,Report submitted to U.S. Army Corps of Engineers, Albuquerque,NM, 1984.




The Upper Rio Grande basin stretches from theheadwaters of the Rio Grande in Colorado to Fort

Quitman, Texas, 70 miles southeast of El Paso. Surfacewater flow in the basin is highly variable and can easi-ly vary 50 percent above or below the long-term meanflow. Surface water management in the Upper RioGrande basin has evolved to address this variability sothat water users can be protected in times of bothdrought and flooding. People sought to fund andbuild water projects that would store floodwater, thusreducing flooding risk and allowing the stored flood-water to be used later in the year, when natural flowswere low, to irrigate crops. One significant example,the construction and subsequent operation of the RioGrande Project, resulted from controversy associatedwith a multi-year drought in the 1890s.

Several years of low snowmelt, in combination withincreased irrigation diversion, mostly in the San LuisValley of Colorado, resulted in water shortagesthroughout the basin with the most pronouncedeffects experienced in the lower parts of the basin.Texan, New Mexican, and Mexican farmers along theinternational border suffered significant shortages insupply. Although individual farmers complained foryears, it was not until Mexico formally complainedthat the U.S. government became actively involved. In1896, in an attempt to “freeze” development upstreamof the border, the Secretary of the Interior declared anembargo prohibiting use of federal funds or grants ofeasements across federal land for water developmentprojects. That action effectively stopped additionallarge-scale water development upstream of what isnow Elephant Butte Reservoir until the 1920s.

In 1906 the United States and Republic of Mexicoresolved their differences when they entered into theInternational Treaty of 1906. Except in times ofextraordinary drought, the treaty guarantees Mexico60,000 acre-feet of water each year at El Paso. TheU.S. Congress authorized construction of the RioGrande Project in part to assure the guarantee couldbe fulfilled.

THE RIO GRANDE PROJECT

The Rio Grande Project, located in southern NewMexico and northwest Texas, consists of two reservoirs

(Elephant Butte being one) and four river diversiondams. It extends 130 miles south from Elephant ButteReservoir past Las Cruces, New Mexico, and El Paso,Texas, to the Hudspeth County line in Texas. It wasconstructed by the U.S. Bureau of Reclamation in theearly twentieth century, in part to comply with termsof the 1906 treaty. Its primary purpose is to deliverwater to 160,000 acres of land in New Mexico andTexas for irrigation, and to provide 60,000 acre-feet ofwater annually to Mexico.

Construction and operation of the Rio GrandeProject resolved issues between the U.S. and Mexicoabout surface water delivery north of Fort Quitman,Texas, and south of Elephant Butte Reservoir. It estab-lished the infrastructure and operations necessary forthe U.S. to store Rio Grande waters in Elephant ButteReservoir and to deliver the stored water for irrigationuse in New Mexico, Texas, and Mexico. The projectalso significantly reduced flooding risk and improvedsurface water security for people living south ofElephant Butte Reservoir. Although the amount ofwater available from the Rio Grande Project has variedthrough time, the project clearly has brought muchmore certainty to landowners downstream.

Elephant Butte Reservoir was the first large storagereservoir on the Rio Grande and is the primary storagereservoir for the Rio Grande Project. It is the largestreservoir in the Upper Rio Grande basin. Its primaryauthorized purpose is to provide water for irrigation.Although Congress authorized a “recreation pool” forthe reservoir in 1974, no permanent source of waterwas reserved for the pool. (A recreation pool is storagespace within the reservoir for water that would neverbe released; it would be lost only through evaporation.The idea is to hold some water in the reservoir evenduring the driest times.) The reservoir therefore hasno defined minimum pool, and recreational waterusers do not have a water right.

The Bureau of Reclamation and two U.S. irrigationdistricts (the Elephant Butte Irrigation District in NewMexico and the El Paso Water Improvement DistrictNo. 1 in Texas) run the Rio Grande Project. TheBureau of Reclamation makes allocations of projectwater to the two U.S. districts and Mexico during eachirrigation season (March through October) throughthe U.S. International Boundary and Water Commission.

Managing Surface Waters on the Upper Rio Grande

Rolf Schmidt-Petersen, Rio Grande Bureau, New Mexico Interstate Stream Commission



Each district’s board of directors uses the Bureau ofReclamation’s allocations to develop general plans forstorage releases (of their respective allocations) duringthe irrigation season. Individual farmers are allocateda set amount of surface water by their district for useduring the irrigation season, based upon the irrigableacreage held by the farmer. An individual farmerorders surface water from the district as needed untilhis/her allotment is fully delivered. Each district and

Mexico (through the U.S. Section of the InternationalBoundary and Water Commission) make regularrequests of the Bureau of Reclamation for delivery ofwater at specific diversion dams to meet their farmers’orders. The Bureau of Reclamation then releases fromstorage the amount of water needed to provide therequested river diversions.

Rio Grande Project water is also used for municipaland industrial purposes. The City of El Paso is a

The Rio Grande basin in New Mexico north of Elephant Butte Reservoir.

landowner in El Paso WaterImprovement District No. 1,has fallowed the land it ownsor leases, requests surfacewater from the district as anydistrict farmer would, andthen uses the delivered sur-face water for municipal andindustrial purposes.

The Rio Grande Project setthe stage for battles betweendirect flow users north ofElephant Butte Reservoir andwater users south of thereservoir. Water use north ofthe reservoir continued to beconstrained by the ability ofindividuals to divert waterfrom the river, the variabilityin supply, sedimentation, theability of people upstream totake and consume water, andthe inability of most peopleto receive federal funds andpermissions to cross federalland to improve access to surface water. In bothColorado and New Mexico above Elephant ButteReservoir, water users were experiencing shortagesespecially toward the end of the irrigation season.

CONFLICT AMONG COLORADO, NEW MEXICO,AND TEXAS

By 1925 the Secretary of the Interior had lifted thefederal water development embargo of 1896 when heauthorized rights of way for construction of a reservoirin Colorado. The embargo had remained in place longafter Elephant Butte Reservoir became operational.People downstream of the reservoir were adamantabout the need for the embargo to remain in order toprotect Elephant Butte Reservoir storage. They had thefederal government as an ally, because neither theynor the federal government wanted to allow reservoirsto be built upstream, given that such reservoirs wouldreduce the amount of water making it to ElephantButte Reservoir. People upstream were adamant thatdamage from floods and drought made it difficult forthem to take water from the river and use it as theyhad before the embargo.

Upon partial lifting of the embargo, Coloradansbuilt several small reservoirs without federal funding.After lifting of the embargo, New Mexicans in the

middle valley sought to reclaim lands damaged duringthe preceding 30 years, reduce flood risk, andimprove their irrigation infrastructure. The Middle RioGrande Conservancy District was organized in 1925to do just that. It drained waterlogged lands andremoved some seventy different river diversion points,consolidating them to the four that exist today(Cochiti, Angostura, Isleta, and San Acacia) for deliv-ery of water to district farmers. Some seventy differentacequias operating in the middle valley in the early1920s were subsumed by the conservancy district. Itcentralized the irrigation delivery system and con-structed El Vado Dam and Reservoir.

The efforts of the Middle Rio Grande ConservancyDistrict resulted in reclamation of previously water-logged lands within the district, significantly improvedwater delivery to farmers, and somewhat reducedflood threat. Those efforts—and funding—were large-ly non-federal in origin, with the exception of fundsassociated with a 1928 act of Congress that providedfunding for work associated with water delivery toPueblo lands within the district. The 1928 act alsoestablished broad categories of water rights and priori-ties within the district. Most explicitly, it designatedspecific amounts of lands within the six Middle RioGrande pueblos as having a senior water right to anyother Middle Rio Grande Conservancy District lands

Unregulated native Rio Grande flow at the Otowi gage near Los Alamos. The Otowi gage pro-vides an estimate of annual water supply conditions both upstream and downstream. Ifupstream water users are experiencing flooding or drought, such events are observed in thegage record. Because 80–85 percent of the water that flows in the Upper Rio Grande basin inmost years flows past this point, the gage provides an estimate of how much surface waterwas available downstream.





The Colorado Division of Water Resources oversees sur-face water diversions north of the state line with New

Mexico to deliver water to its farmers and to the State ofNew Mexico under the 1938 compact. The Bureau ofReclamation, U.S. Army Corps of Engineers, U.S. Bureau ofIndian Affairs, New Mexico Office of the State Engineer/Interstate Stream Commission, and Middle Rio GrandeConservancy District collaborate on irrigation deliveries,maintaining U.S. Fish and Wildlife Service BiologicalOpinion river flow targets in the middle valley, and com-pact management between the state line with Colorado andElephant Butte Dam. The Bureau of Reclamation, U.S.International Boundary and Water Commission, ElephantButte Irrigation District, and El Paso Water ImprovementDistrict No. 1 collaborate on surface water management forthe Rio Grande Project.

The Colorado Division of Water Resources tracks RioGrande surface water flows at the state line with NewMexico relative to its upstream 1938 compact index flowsand associated delivery requirements. Based upon theirprojections of needed deliveries, the district engineer coor-dinates with water users to decide when the irrigation sea-son will begin and end and curtails surface water diver-sions during the irrigation season (sometimes for waterusers with rights that significantly pre-date the signing ofthe compact) in order to meet its required annual deliver-ies to New Mexico. The Corps of Engineers coordinateswith the Division of Water Resources and the Bureau ofReclamation to oversee flood operations at PlatoroReservoir on the Conejos River, when downstream riverflow conditions warrant such operations.

In the Upper and Middle Rio Grande valleys of NewMexico, surface water management can be separated intothree main categories, the winter, the irrigation season, andflood control operations. During the winter (Novemberthrough February), the Bureau of Reclamation, the ArmyCorps of Engineers, and the Interstate Stream Commissioncoordinate to manage the system’s reservoirs in compliancewith the 1938 compact to route 1938 compact deliveries toElephant Butte Reservoir and to maintain river flow on theRio Chama by moving San Juan–Chama Project water tovarious delivery points. Generally, management for endan-gered species flow targets is not necessary during the winter.

During the irrigation season (March through October),the Bureau of Reclamation, the Corps of Engineers, theBureau of Indian Affairs, the Fish and Wildlife Service, theMiddle Rio Grande Conservancy District, the InterstateStream Commission, and the Albuquerque BernalilloCounty Water Authority confer daily, as necessary, on riverflow conditions throughout the Upper and Middle RioGrande valleys. Based upon the amount of surface waterflowing naturally into the middle valley, weather condi-

tions in both the upper and middle valleys, irrigationdemand, and the needed Biological Opinion flows, deci-sions are made on whose water and how much water torelease from upstream reservoirs. The agencies coordinateto deliver water to the Middle Rio Grande ConservancyDistrict diversion dams and maintain compliance with theFish and Wildlife Service Biological Opinion. During thesnowmelt runoff they work, as needed, to provide flows forspawning and recruitment of silvery minnow. During lowflow periods they coordinate to observe and manage riverdrying throughout the Middle Rio Grande valley, salvagesilvery minnow, and thus reduce take of silvery minnow.

As warranted by snowpack, reservoir, and river condi-tions, the corps manages its reservoirs (Abiquiu, Cochiti,Jemez Canyon, and Galisteo) to store floodwater and pro-vide flood damage protection to lands and people living inthe upper and middle valleys from Abiquiu to ElephantButte Reservoir. The water management agencies coordi-nate with the State Emergency Management Office andindividual county emergency managers, as necessary, toprovide flood warnings. The Bureau of Reclamation con-ducts emergency operations as necessary to limit thepotential for flooding within the Middle Rio GrandeProject. Additionally, the Middle Rio Grande ConservancyDistrict may increase its diversions from the river in orderto reduce pressure on the downstream levee system.

In the Lower Rio Grande, the Bureau of Reclamation, theElephant Butte Irrigation District in New Mexico, and theEl Paso Water Improvement District No. 1 in Texas run theRio Grande Project. The Bureau of Reclamation makes allo-cations of project water to the two U.S. districts andMexico during each irrigation season (March throughOctober) through the U.S. Section of the InternationalBoundary and Water Commission. Each district developsgeneral plans for storage releases (of their respective alloca-tions) during the irrigation season and allocates water toindividual farmers. An individual farmer orders surfacewater from his/her district as needed until his/her allot-ment is fully delivered. The City of El Paso is a landownerwithin the El Paso Water Improvement District No. 1,receives annual allocations of water from the district, anduses its allotted surface water to provide drinking water toits citizens. Each district and Mexico (through the U.S.Section of the International Boundary and WaterCommission) make regular requests of the Bureau ofReclamation for delivery of water at specific diversion damsto meet their farmers’ orders. The Bureau of Reclamationthen releases from storage the amount of water needed toprovide the requested river diversions. The U.S. Section ofthe International Boundary and Water Commission overseesflood control operations.

How Surface Water Management Decisions are Made

and indicated that lands not previously irrigated(newly reclaimed lands) had the most junior priority.

As part of an effort to have the 1896 embargo lifted,the state legislatures of New Mexico and Coloradoeach passed statutes in 1923 authorizing designationof commissioners to pursue formulation of an inter-state compact for the Rio Grande. In 1926 Texasjoined the Compact Commission to protect its wateruser interests.

THE 1929 RIO GRANDE COMPACT—THE“STANDSTILL COMPACT”

The temporary Rio Grande Compact, signed inFebruary 1929, was designed to maintain the statusquo in the basin. Major parts of the agreement includ-ed:

• A condition that neither New Mexico norColorado could increase diversions or storage ofwater on the Rio Grande until such time as theresulting depletions were offset by drainage proj-ects. The primary assumption is that the drainageprojects would return an equal amount of waterto the river as that consumed by the reservoirstorage and associated release operations. Thisfacilitated efforts in Colorado to construct adrainage project to drain water from the “closedbasin” in the San Luis Valley of Colorado to theRio Grande and build a mainstem reservoir. InNew Mexico it facilitated efforts to improvedrainage in the middle valley and to construct ElVado Reservoir.

• A provision for creation of a CompactCommission to permanently and equitably appor-tion the river in the Upper Rio Grande basin.

• A provision that each state will maintain streamflow gaging stations and exchange records ofmeasurements.

The 1929 compact created a path for the three statesto agree upon water use limits and delivery require-ments. It also set the stage for Coloradans and NewMexicans north of Elephant Butte Reservoir to pushaggressively for funds, both private and federal, toimprove their ability to control and use water. Theperiod between 1929 and 1938 was one of coopera-tion and conflict: the Rio Grande Joint Investigation ofwater uses in the Upper Rio Grande basin was beingconducted collaboratively in the midst of a U.S.Supreme Court lawsuit filed by Texas against NewMexico and the Middle Rio Grande Conservancy

District, in part over the construction of El Vado Damand Reservoir. Not many people realize that the RioGrande Compact of 1938 settled a U.S. SupremeCourt lawsuit in addition to resolving water uses fromthe headwaters of the river in Colorado to FortQuitman, Texas; the lawsuit was dismissed upon sign-ing of the 1938 compact. And the 1938 compact wassigned shortly after agreements between the U.S. gov-ernment and water users in Elephant Butte IrrigationDistrict and El Paso Water Improvement District No. 1were signed, dividing the irrigable lands of the RioGrande Project, with 57 percent going to ElephantButte Irrigation District and 43 percent to El PasoWater Improvement District No. 1.

THE 1938 RIO GRANDE COMPACT

The 1938 compact was designed to stabilize waterdepletions in the Upper Rio Grande basin as theyexisted in 1929. It reflects the efforts of the negotiatorsto ensure that the same general quantity of flow thatmade it to Elephant Butte Reservoir before 1929 continued to do so. The 1938 Rio Grande Compactestablished surface water delivery obligations forColorado to New Mexico near the state line and NewMexico to Texas at San Marcial near the headwaters ofElephant Butte Reservoir. The compact sets annualdelivery obligations of Colorado to New Mexico and anine-month delivery obligation of New Mexico toTexas. As a result, it also established depletion amountsfor both Colorado and New Mexico of the natural flowof the river at certain points.

The 1938 compact provides for both drought andhigh-flow conditions. It does not require Colorado orNew Mexico to deliver the exact amount of waterscheduled annually each and every year. The compactallows for the accumulation of over deliveries (credit)and under deliveries (debit). It allows for significantdebits in deliveries to accrue before either Colorado orNew Mexico are in violation of the compact.Additionally, if Elephant Butte Reservoir is full andspills, the compact provides that all credits or debitsare wiped out.

What does the 1938 Rio Grande Compact mean inpractice for people living in New Mexico or Coloradonorth of Elephant Butte Reservoir? The Rio GrandeCompact establishes depletion amounts for Coloradoand New Mexico of the natural flow of the river at cer-tain points. Although it is up to each state to decidehow its water is used, any new use has to be balancedby reduction of an existing use. Alternately, new usescan be supported using imported water supplies such




as San Juan–Chama Project water. Ground water canbe used as long as the impact of that use on the RioGrande is offset. Typically, ground water withdrawnwill eventually deplete the river in the same amountas that pumped. Furthermore, the compact placesrestrictions on the operation of reservoirs in NewMexico and Colorado constructed after 1929 if thestorage supply of the Rio Grande Project drops belowa specified level, or if either state has an accrued com-pact debit.

THE 1948 RIO GRANDE COMPACT RESOLUTION

In 1948 the Rio Grande Compact Commissionapproved a resolution moving the New Mexico deliv-ery point from the headwaters of Elephant ButteReservoir to Elephant Butte Dam. The move was made


because of difficulty measuring the flow of the RioGrande at the old delivery point and to develop anannual delivery obligation of New Mexico to Texas.The commission did so in part by estimating long-term evaporation rates from Elephant Butte Reservoir,based upon historic Rio Grande Project operations.

With the above changes, the negotiators alsoremoved a clause from Article IV of the 1938 compactconcerning application of New Mexico’s deliveryschedule. The clause had required changes to thedelivery schedule should New Mexico deplete therunoff of tributaries to the Rio Grande between OtowiBridge and San Marcial during the summer months byworks constructed after 1937. The resolution removedone impediment to the federal government’s construc-tion of Middle Rio Grande Project facilities. Finally,because many of the depletions in the middle valley

Direct Flow Right—A water right, with a priority date, todivert a certain amount of water from a stream and put itto beneficial use.

Storage Right—A water right to store surface water attimes when downstream senior direct flow and storagerights are satisfied. The stored water is released upon thecall of the storage right holder for downstream beneficialuse. Once released, the storage water suffers natural lossesbetween the point of release and point of diversion.

Rio Grande Project—A U.S. Bureau of Reclamation waterproject in southern New Mexico and northwest Texas con-sisting of two reservoirs and four river diversion dams. Theproject extends 130 miles south from Elephant ButteReservoir past Las Cruces, New Mexico, and El Paso,Texas, to the Hudspeth County line in Texas. The projectwas constructed to deliver water to 160,000 acres of landin New Mexico and Texas for irrigation purposes and toprovide 60,000 acre-feet of water to Mexico annually.

Middle Rio Grande Conservancy District—An entityorganized under the New Mexico Conservancy Act of1923, as amended to plan for reclamation, flood protectionand irrigation in the Middle Rio Grande. The district islocated in the middle of the state, extending south fromCochiti Reservoir 150 miles past Albuquerque and Socorroto the northern boundary of the Bosque del ApacheNational Wildlife Refuge. The district oversees operation ofa reservoir on the Rio Chama and four river diversionstructures within the Middle Rio Grande valley.

Elephant Butte Irrigation District—An entity governed byan elected board and organized in New Mexico in 1918 tocontract with the Bureau of Reclamation for irrigationworks ultimately servicing some 90,000 acres of irrigableland within the New Mexico portion of the Rio GrandeProject. The district requests reservoir releases from theBureau of Reclamation for New Mexico farmers, divertsthat water at one of three diversion dams, and then deliv-ers the water to its constituent farmers.

Middle Rio Grande Project—A U.S. Army Corps ofEngineers and Bureau of Reclamation project authorized in1948 and 1950 to provide additional flood control, stor-age, channel rectification, restoration of irrigation works,and other efforts on the Rio Grande between Velarde, NewMexico and Elephant Butte Reservoir.

San Juan–Chama Project—A U.S. Bureau of Reclamationproject consisting of three diversion dams, three tunnels,and one reservoir. The project is used to deliver a portionof New Mexico’s Upper Colorado River Compact waterapportionment from the San Juan Basin to the Rio GrandeBasin. The Bureau of Reclamation diverts water from threetributaries of the San Juan River in southwest Coloradoand transports it under the continental divide via a seriesof tunnels to Heron Reservoir on Willow Creek just aboveits confluence with the Rio Chama. The bureau contractswith various entities in the Rio Grande basin north ofElephant Butte Reservoir for annual deliveries of SanJuan–Chama Project water from Heron Reservoir.

A Few Definitions

are natural, the resolution established a need for NewMexico to maintain the river through the middle valley to both control natural depletions and efficientlydeliver water to Elephant Butte Reservoir. To put itanother way: In order for New Mexico to increase itshuman water use in the middle valley it must reduceand control existing natural uses (evapotranspirationfrom the bosque or evaporation from open water).

THE MIDDLE RIO GRANDE PROJECT (1948 AND1950 FLOOD CONTROL ACTS)

Large floods in the early 1940s resulted in significantshort-term and long-term harm for people living inthe middle valley. The Middle Rio Grande ConservancyDistrict’s irrigation infrastructure suffered significantdamage; the district was nearly bankrupt, lands in thevalley became salinated or waterlogged, and the riverchannel ceased to exist south of the Bosque del ApacheNational Wildlife Refuge. Given the hardship, residentsonce again sought to reduce flood risk and improveconveyance of water. Additionally, New Mexico beganto accrue significant compact under deliveries.

The Middle Rio Grande Project, a joint U.S. ArmyCorps of Engineers and U.S. Bureau of Reclamationproject, was therefore advocated and supported bymany parties as a way of addressing the myriad ofmiddle valley water problems. The project includedconstruction of four large flood control reservoirs,removal of multiple miles of river channel from the val-ley, construction of the Rio Grande “floodway” and theLow Flow Conveyance Channel, and reconstruction ofparts of the Middle Rio Grande Conservancy District.

The operations of all the Corps of Engineers floodcontrol reservoirs must comply with the 1938 compact.The corps cannot store native Rio Grande water exceptfor floodwater, cannot deviate from defined operationswithout approval of the Compact Commission, andmust pass floodwater through the system at the highest“safe” rate possible. Under certain circumstances, thecorps cannot release stored floodwater after July 1 ofany year until the end of the Middle Rio GrandeConservancy District’s irrigation season.

The river realignment and water conveyance facili-ties of the Middle Rio Grande Project reduced waterconsumption and aided New Mexico in meeting itsdelivery obligations. The Middle Rio Grande Projectwas and remains a key element in New Mexico’s abili-ty to maintain compact compliance. Consequently,maintenance of the Middle Rio Grande Project is vitalfor the state.

THE SAN JUAN–CHAMA PROJECT

This project, constructed by the U.S. Bureau ofReclamation in the 1960s and early 1970s, importswater to the Rio Grande basin from the San Juan Basinfor use in New Mexico. It is accounted separately fromnative Rio Grande water and provides water to helpalleviate shortages in available native Rio Grandewater. In the future San Juan–Chama water will be aprimary source of drinking water to the citizens ofEspañola, Los Alamos, Santa Fe, and Albuquerque.

The Bureau of Reclamation operates the project,diverting a portion of New Mexico’s Upper ColoradoRiver Compact apportionment from the San JuanBasin to the Rio Grande basin. The project providesadditional surface water for New Mexico water userswith San Juan–Chama contracts. The operation of theproject requires complex accounting procedures. Inorder for the system to function properly, a partialadjudication of water rights was conducted on the RioChama by the Office of the State Engineer to protectSan Juan–Chama Project water from being consumedby the acequias once it is released from upstreamreservoirs. Additionally, the state engineer establishedbypass flow requirements through El Vado andAbiquiu Reservoirs to provide Rio Chama acequiastheir senior water rights downstream of Abiquiu Dam.

The San Juan–Chama Project brings added flexibili-ty in managing water in the Upper Rio Grande basinabove Elephant Butte Reservoir. It has been used tomaintain the pool of water in Cochiti Reservoir, therecreation pool in Elephant Butte Reservoir, for irriga-tion, and to offset the effects on the river of pumpingfor municipal and industrial uses. The water has alsobeen used to provide secondary benefits such as win-ter flows on the Rio Chama and to aid in meeting flowtargets of the Endangered Species Act between itspoint of release and point of use.

THE 1950s DROUGHT AND U.S. SUPREME COURTLITIGATION

During the 1950s drought Texas sued New Mexicoand New Mexico and Texas sued Colorado in the U.S.Supreme Court. The suits were filed to force NewMexico and Colorado to comply with the 1938 com-pact and make up under deliveries (then more than300,000 acre-feet for New Mexico and 900,000 acre-feet for Colorado). The Texas case against New Mexicowas dismissed on a technicality. Nonetheless, with onecaveat, El Vado Reservoir has since been operated incompliance with the 1938 compact. The federal gov-



34 C H A P T E R O N E

ernment, as part of its tribal trust responsibility to thesix Middle Rio Grande pueblos, stores water in ElVado Reservoir as insurance for delivery of direct flowto the Prior and Paramount Lands (lands identified ashaving a senior water right to other Middle RioGrande Conservancy District lands in the 1928 act ofCongress) of the six Middle Rio Grande pueblos. Thestored water is released for delivery when the directflow of the river drops below levels the federal gov-ernment has estimated to be needed to adequatelydeliver water to the Prior and Paramount Lands.

In 1968 the U.S. Supreme Court granted a stipula-tion for continuance of the New Mexico v. Coloradocase as long as Colorado met its annual compact obli-gation until it was once again in compliance. Coloradomet or exceeded its obligation each year from 1968through 1984 and has remained in compliance sincethen. Its remaining under-delivery to New Mexico wascancelled in 1985 when Elephant Butte Reservoirspilled. The case was subsequently dismissed. To meetits annual obligation, Colorado restricts the diversionof surface water users with rights that pre-date the1938 compact.

Suggested Reading

Upper Rio Grande Water Operations Review and Environmental ImpactStatement. U.S. Bureau of Reclamation, U.S. Army Corps ofEngineers, and New Mexico Interstate Stream Commission, 2005.

The Upper Rio Grande: A Guide to Decision-Making, Steven J. Shupe andJohn Folk-Williams, Western Network, Santa Fe, New Mexico, 1988.

The Administration of the Rio Grande Compact in Colorado, Steven E.Vandiver. CLE on the Law of the Rio Grande, 2003.


Documents

Long River, Short Water: The Rio Grande Water Development Story · 2007-06-01 · flows from Colorado diversion, were less able to maintain themselves. THE RIO GRANDE PROJECT Water