Historic Reconstruction of the Ecology of the Rio Grande/Río Bravo Channel and Floodplain in the Chihuahuan Desert <P>
<P>American Rivers has placed the Rio Grande among the 10 most endangered rivers of the United States
environmental change, the late-nineteenth and early twentieth centuries were among the most significant periods in the river?s history: agricultural development and dam construction impacted river flows and floodplain habitats directly and land-use changes elsewhere in the river?s watershed impacted run-off and sedimentation patterns.
In the thirteenth and fourteenth centuries, agricultural societies existed along several parts of the desert Rio Grande; they appear to have engaged in flood-water farming that utilized overbank flooding or diverted sheetflow after rainfall for irrigation instead of using ditch irrigation
Ditch irrigation became more widespread following the establishment of missions and settlements in the sixteenth and seventeenth centuries; although along some portions of the desert Rio Grande (such as in the Mesilla and Presidio valleys) large-scale ditch irrigation did not occur prior to the nineteenth century. Late in the nineteenth century, extensive agricultural development in southern Colorado and northern New Mexico usurped most of the river?s water, resulting in flow disruptions downstream and precipitating the construction of Elephant Butte Reservoir (completed in 1916) to store water for agriculture downstream
Above the mouth of the Río Conchos, flows peaked in late spring or early summer when snow melted in the upper reaches of the river?s watershed; below the Río Conchos, peak flows tended to occur in late summer or fall, resulting from summer precipitation in Mexico. In many years, overbank flooding occurred during periods of peak flow
Meander cut-offs and channel migration also helped to create and maintain numerous backwaters including oxbows, marshes, and sloughs. These non-river aquatic habitats may have been important refugia for aquatic organisms when portions of the river went dry; prior to extensive agricultural development, flow disruptions along limited stretches of the river occurred quite infrequently, and seem to have only occurred during extended droughts.
Flood-tolerant seep willows occurred along the river?s edge, and cottonwoods and willows ? in stands near river
Farther from the active channel, large expanses of shrubby growth dominated by screwbean and honey mesquite occurred. Salt grass often grew in association with screwbean, and elsewhere, patches of arrowweed could be found. Marshes and sloughs were typified by aquatic and emergent vegetation such as pondweed, cattails, sedges, and rushes. Nineteenth-century surveys and descriptions suggest that cottonwood stands may have occupied roughly a quarter to a third of the area in river valleys, with shrubby and more open growth covering larger expanses, and marsh and slough habitats limited to a small area (perhaps 5% or less of the river valleys).
Under natural flow regimes [the areas would have been dynamic] ? dynamic eqilibrium of this system allowed for a diversity of plant communities and stands of many different ages
vertebrate diversity is high along the river: as many as 44 native fishes, 95 amphibian and reptiles, 95 mammals, and 345 birds occur (or previously occurred) along the desert Rio Grande.
fish were abundant, and waterfowl provided a ready source of meat for travelers. Large animals including deer, turkeys, bears, wolves, and mountain sheep occurred along some reaches of the river; all but the deer have been extirpated by human activities along the river
beaver populations along the river declined severely during the 1820s due to overharvesting by fur trappers?but many followed
bears and turkeys were eliminated from the Socorro Valley sometime after 1850
conversion of natural habitats to agriculture and urban areas, the introduction of exotic species, water diversions, flow regulation, dam and levee construction, channel straightening and dredging, and numerous other changes have taken place along the desert Rio Grande and all have impacted the organisms of the riparian landscape immensely. The dynamics of the river system and the ecosystems that depend on it have been changed, reducing the natural heterogeneity of the system and severing connections between the patches within it.
restoration is necessary because centuries of intensive settlement, agriculture, and industry along the river?s shores have impacted its natural environments immensely. The environments are so degraded that American Rivers placed the Rio Grande on its list of the ten most endangered rivers of the United States in 1993 and again in 2000 (Bartlett 1995: 27, American Rivers 2000)
Weniger?s two-volume set only includes a handful of pages devoted to the Rio Grande through the Chihuahuan Desert (pp. 148-154). Although excellent historic compilations exist for the Middle Rio Grande in northern and central New Mexico (e.g.: Crawford et al. 1993, Scurlock 1998), no comprehensive document focusing on the river through the Chihuahuan Desert exists
?by 1880, every piece of irrigable land along the length of the Rio Grande [in New Mexico] was under development? (Harris 1995:
it was this reduction in flow to downstream users that necessitated the construction of Elephant Butte Dam in southern New Mexico (completed in 1916
natural flooding cycle and sedimentation patterns were disrupted, affecting not only the aquatic environment but also many terrestrial communities dependent upon these physical processes
journals of 49-ers
Geographic and Historic Overview
desert Rio Grande traverses about 1270 kilometers of the Chihuahuan Desert between the mouths of the Rio Puerco in New Mexico and the Devil?s River in Texas
alternates between a series of open, level valleys and narrows or canyons where geologic features restrict the river?s path
Through the desert, only a handful of perennial tributaries
Río Conchos, the Pecos River and the Devils River are the only major perennial rivers that reach the desert Rio Grande, and all meet the river in the lower half of its Chihuahuan Desert course. Thus, above the mouth of the Río Conchos, most of the water in the desert Rio Grande ultimately comes from snowmelt in the upper reaches of the river?s watershed in northern New Mexico and southern Colorado.
Rio Puerco has not been a perennial stream during the historic period, as its flow has ceased during dry seasons and drought periods (Scurlock 1998
The Belen and Socorro valleys [central NM] are both within the stretch of the river described as the ?Middle Rio Grande?; the desert Rio Grande accounts for the southernmost 38% of the length of the Middle Rio Grande (100 of 260 km total; Crawford et al. 1993
valleys of the Middle Rio Grande have been continuously occupied by agricultural societies since at least 1350 AD ? land clearing and water diversions for agriculture (Crawford et al. 1993
Middle Rio Grande in 1540 as a part of Coronado?s expedition, it is estimated that about 10,000 hectares of land were under cultivation in the valleys along the Middle Rio Grande (Crawford et al. 1993
first clear records of irrigation ditches watering such fields were documented by members of the Espejo expedition in 1583
Wozniak (1987: 7-15) suggests that floodwater farming (relying on overbank flows and surface run-off following rainfall) was probably the most important agricultural practice during the initial period of contact with the Spanish, and he reviews evidence suggesting that the few early irrigation ditches that were used did not tap mainstem river flows, but that they instead diverted side flows during flood events, as well as diverting waters from marshes and tributaries entering the river valley.
peak around 1880, when about 50,000 hectares were being cultivated (Crawford et al. 1993
two factors: the sudden increase in irrigated agriculture upriver, which reduced river flows coming into the Middle Rio and the lack of drainage mechanisms for excess irrigation water, which resulted in increased salinity, rising water tables, and flooding of cultivated lands in the valley (Crawford et al. 1993: 24).
introduction of huge herds of sheep to the watershed, which affected the flow of water and sediments into the river. As Horgan (1954 364 :
?by the middle of the eighteenth century there were millions of sheep grazing on the sparse slopes of the watershed. Between two hundred thousand and five hundred thousand sheep were driven every year to Mexico for sale. The grasses struggled for life in ordinary years and in dry years barely showed. The colonials looked at their hills and shook their heads. It was all very much like Spain, a condition of the natural life that seemed impossible to govern. The tilted lands were growing more and more barren, the torrents?when it did rain?swept faster and cut deeper, the earth ran into the tributaries and into the river, piling up silt on the river floor, the river spilled over its old banks and made swamps on good farm land, and a man could only bow his head and invoke patience
significant trade route, the Camino Real or Chihuahua Trail, developed along much of the desert Rio Grande downriver as far as El Paso. Thus, large numbers of people and livestock traveled along the desert Rio Grande.
Engle Valley, dam for this reservoir was built in Elephant Butte Canyon was flooded to make elephant butte
Emory (1976: 54) noted: ?The valley of the Del Norte, as we advance [southward], loses what little capacity for agriculture it possessed. ..?
Socorro valley?s relative isolation (off of the Camino Real) seems to have resulted in slower rates of settlement and agricultural development
Mesilla and El Paso Valleys
<P>in 1849, camped near Las Cruces, Eccleston (1950: 160) observed a cornfield ?3 leagues [12 kilometers] long? and surveyors crossed acequias (irrigation canals) 6 times while establishing township and range boundaries near Las Cruces and Mesilla in 1857 (Garrettson 1857).
greater El Paso area (which includes the Mesilla Valley) was probably inhabited by ?small, mobile bands subsisting predominantly on large game supplemented with the utilization of wild plant resources? between about 10,000 and 6,000 BC (Peterson et al. 1994: 55). Archaic period occupations began around 6,000 BC and by 1000 or 2000 BC, plant domestication was probably underway (Peterson et al. 1994: 58, 61)
?there is no archeological evidence of prehistoric irrigation facilities on the floodplain of the Rio Grande near El Paso? and agriculture probably depended upon the diversion of sheetwash runoff following rainstorms (Ackerly 1994: 118).
Guadalupe Mission, in modern Juarez) was established in 1659, and ditch irrigation was evidently introduced to the area at this time by the Spanish (Ackerly 1994: 118)
1934, an El Paso Herald-Post report described a four-acre tract near Ysleta ?which had been in continuous cultivation by the Roman Catholic Church since 1682? (Hammons 1942: 26).
1773, an El Paso Valley resident described a diversion dam as follows:
The upkeep of the dam is obligatory upon all. It is made of wattles, as the terrain of the river does not permit any other kind of fabrication, to say nothing of the trouble caused by its excessive floods and freshets, for it not seldom happened that after a dam had been built of stones, fagots, and stakes it was necessary to tear it down to prevent inundation of the town. (Hackett 1902: 507)
Other sources described the eighteenth-century diversion structures as large baskets woven of willow wands which were filled with small rocks and placed into specific parts of the river channel during the irrigation season (Horgan 1954: 347). Irrigated agriculture also required land-leveling
1849, [t]he whole Rio Grande Valley, from Santa Fe to El Paso, was the half-way place on the overland trip where the emigrants coming via Missouri, Arkansas and Texas expected to lay in new supplies. Most parties rested for three to four weeks to build up the animals for the balance of the journey, and consequently, there was congestion at El Paso, Santa Fe, and every little village between the two places. (Martin 1925: 300-301)
one reported to contain 800 Americans and 300 Mexicans travelling in 300 wagons with 4000 cattle and 300 mules; at least 4000 emigrants (and all of their livestock) were reported camping in the vicinity of El Paso in 1849 (Martin 1925: 301).
two International Boundary Water Commission (IBWC) projects initiated in the 1930s?the Rio Grande Canalization Project (between Caballo Dam and El Paso) and the Rio Grande Rectification Project (El Paso to Ft. Quitman)?were established to facilitate water delivery to Mexico (as required under the Treaty of 1906) and to stabilize the river?s location along the international boundary. To accomplish these goals, the river was channelized, largely confined to a floodway defined by levees, and a program of vegetation removal was initiated (IBWC 1978: 11-12). In addition, in the El Paso Valley, the river?s path was straightened, reducing the length of its path from about 250 to 138 kilometers (IBWC 1978: 5).
Forgotten Reach , ?the forgotten river? because it is remote and often overlooked , 1220 to about 1450 AD have been found through this reach; they appear to be associated with trade and wild-food gathering, with no evidence of agricultural activity (Kelley 1990: 38).
Spanish period. During a 1683 ? recorded a number of ?rancherias? belonging to members of the Suma nation; he suggested that their main food was roasted yucca roots and made no mention of agriculture along this stretch (Mendoza 1952: 321-324).
Presidio Valley 1200-1300 AD (Kelley 1992: 139); Kelley (1952: 362, 383) argues that the expansion of agricultural societies from the north may have been made possible by a shift to a slightly wetter climate. As in the El Paso area, these farmers practiced floodwater and rainwater irrigation, not ditch irrigation (Kelly 1952: 383). And as happened elsewhere in the region, many of these villages appear to have been abandoned in the 1400s (Kelley 1990: 39), a shift Kelley (1952: 362-3) again ascribes to climate change, this time to a drier climate. By the time of early contact with the Spanish in the mid- to late 1500s, a few farming villages were concentrated
1684, Mendoza (1952: 325) reported that farms belonging to the Julimes nation in the Presidio Valley cultivated corn and wheat. In 1747 and 1748, cultivated crops mentioned in Spanish documents included corn, wheat, pumpkins, and legumes (Ydoiaga 1992
settlements largely depended upon overbank flooding to irrigate their fields; early attempts at ditch irrigation from the two rivers prior to 1747 evidently failed due to the shifting location of river banks, and sandy beaches and soil (Ydoiaga 1992
?floodwater farming seems to have been the ordinary method of cultivating crops above the confluence? (Everitt 1977: 22).
mid-1970s a channel relocation project was completed along 13.4 kilometers of the Rio Grande near the mouth of the Río Conchos, including levee construction and vegetation clearing (IBWC 1978: 13).
below big bend/lower canyons, Michler?s 1856 survey of this area noted that through this area the river?s bed is narrow, and hemmed in by continuous and perfect walls of natural masonry, varying from 50 to 300 feet in height; the breadth of the river being extremely contracted, these structures, seen from our boats, look stupendous as they rise perpendicularly from the water. It is not unfrequently the case that we travel for miles without being able to find a spot on which to land. (Emory 1987, vol. 1, pt. 1: 77)
Completed in 1969, Amistad Reservoir was created by a dam constructed just below the mouth of the Devils River and flooded a 119-kilometer length of the Rio Grande; the waters of Amistad Reservoir reach 40 kilometers up the Devils River, 23 kilometers up the Pecos River, and 48 kilometers up the Rio Grande above the mouth of the Pecos (U.S. Dept of Interior 1999).
Chapter 2: Historic Ecosystem Descriptions
pre19/20th century, desert Rio Grande was likely a system in dynamic equilibrium
Bullard and Wells 1992: 28): if the energy input to a river system changes (such as changes in the volume of water or sediments entering it), then the form of the system (such as the channel gradient or shape) changes to keep the entire system in balance.
Though sediment levels likely increased dramatically in the late-nineteenth century (Sublette et al. 1990: 9, Scurlock 1998: 188)
[accounts of varying travelers over the years have described increasing muddiness and shrinking stream sizes]
upper reaches of the desert Rio Grande [aggrading sediment until an event washes some away.] ?dynamic equilibrium? (Crawford et al. 1993: 19).
for thousands of years, the Middle Rio Grande has had a ?braided, slightly sinuous channel that broadly meandered laterally within the 2-6 km (1-4 mi) wide floodplain? (Crawford et al. 1993: 27
elevation change along the river is quite gradual [5-10ft/mi] (Campbell and Dick-Peddie 1964: 492
Such a low gradient may allow the river to appear slow and calm, as noted by some early travelers. In 1582, a member of Espejo?s party noted that near the mouth of the Río Conchos, the Rio Grande ?flows so quietly that it does not make any noise in spite of being very large in some places? (Luxan 1929: 64).
Mesilla Valley , apparent widening of the channel in the Mesilla Valley during the late-nineteenth and early twentieth centuries (which also occurred in the El Paso and upper Presidio Valleys; Everitt 1993: 230, 236) may have been due, at least in part, to an increase in sediment levels and a reduction in flows, which affected the river?s ability to carry sediments
[Rio Grande meanders naturally and changes over time.] ?The ford [near El Paso] changes more or less every season? and that ?The river [at El Paso] had to be forded by daylight, in consequence of the frequent changes in the channels and bars.? Similarly, in 1846, Ruxton (1973: 168) warned that because of the river?s ??constantly shifting quicksands and bars, [the river near El Paso] is always difficult, and often dangerous to cross with loaded wagons.?
creating islands or oxbow lakes
large-scale changes in channel position also occurred. As illustrated on the Mesilla Valley map (Map 2), below Las Cruces, the 1844 channel was about 6.4 kilometers east of the channel recorded in 1903; included in this change is a channel shift which occurred between 1862 and 1865 and moved the river from the east side of Mesilla to the west. Flooding in August of 1862 left Mesilla surrounded by water on both sides (Couchman 1990: 155) and as the river tried to establish a new channel, between 1862 and 1865, ?the head of the Mesilla ditch had to be changed 13 times? (Yeo cited in Wozniak 1987: 115). In 1865, the river finally cut a new channel beginning at a point weakened by a ditch head (Baldwin 1938: 320).
[other large shifts] between 1903 and 1912, when the river shifted about 4.8 kilometers [have occurred over the years as well
?probably reflect shifts in routes of main streamflow into previously existing channels rather than erosion of a new channel? (Hall 1994: 24, 26). Historic descriptions also mention the river?s tendency to return to pre-existing channels
?They guard against the danger that the river may return to its old course by making deep ditches through which it may flow in such an event? (Hackett 1902: 508
flooding and flow
Environmental History of the Middle Rio Grande Basin, Scurlock catalogs at least 82 floods (defined as events with a flow greater than 10,000 cfs) that occurred along the river above El Paso between 1591-1942 (1998: 32
Scurlock (1998: 32) ascribes these floods to three distinct causes:
?? spring floods (April-June) resulting from heavy snowmelt,
?? widespread summer flooding resulting from extensive summer rains in years with a significant spring flood, and
?? local summer flooding (July-September) resulting from localized but heavy thunderstorms
snowmelt floods tend to build to their peak flow relatively slowly, whereas rainstorms result in flashier pulses of water that create sharp, short peaks in flow (Lagasse 1981: 29, Poff et al. 1997: 771).
For an illustration of a sudden summer rise, consider Hill?s (1901: 167-8) description of his party?s departure from a pile of rocky debris blocking the river in Santa Elena Canyon:
Having finally succeeded in crossing the obstruction early one morning, we transported our baggage to the boats preparatory to leaving. Before the boats were loaded a tremendous roaring sound like a distant thunder was heard up the canyon, and we saw what we most dreaded was happening?the river was rising. A big flood of the ordinary kind would have veneered the dangerous rocks with water and our prospects for escape would have been small. We hastily piled our baggage into the boats and sprang aboard. It was either stay and starve or go and chance it. Fortunately, this particular rise proved to be a small one, just sufficient to give the desired impetus to our craft?.
relatively low flows during the winter and early spring, and maximum flows in late spring or summer.
Bartlett (1965: 187) described how these variable flows influenced El Paso Valley agriculture in the mid-nineteenth century: ?In February and March there is always enough for the first irrigation. In April and May the quantity is much diminished; and if the rise, expected to take place the middle of May, fails, there is not enough to irrigate properly all the fields prepared for it.?
According to Kelley (1986: 119), increasing settlement, especially after the U.S. Civil War, changed the seasonal pattern of river flows associated with snowmelt:
?as settlement grew, timber was stripped from the lower elevations. The denuding of timber allowed winter snows to melt prematurely, and their waters went into winter flow, being lost to the summer season. Prior to heavy settlement, the snows remained packed until the spring thaw.
peak flood at El Paso and the Upper Presidio gage appears to have shifted from May to June, and secondly, the figures illustrate the substantial decrease in discharge during this period. Whereas the average May discharge at El Paso between 1889 and 1895 was over 420 million cubic meters, in the years that followed it was closer to 260 million
Elephant Butte Dam brought an end to the late-spring/early summer flood along the Rio Grande between the dam and the mouth of the Río Conchos
Between 1916 and 1950, total annual flows through the Presidio Valley above the Río Conchos declined gradually (except for a high water period in the early 1940s); from 1951-1969, winter flows were so low the river became ?hydrographically disjunct? when flow events recorded at the Ft. Quitman gage never reached the Upper Presidio gage (Everitt 1993: 228-9).
Elephant Butte Dam. Because of the scouring effects of the clear water released from the dam, in Las Palomas Valley, and to a lesser extent, in the Mesilla Valley, the river bed degraded, becoming about 0.6 m deeper in the first 15 years after the dam?s completion (Lagasse 1981: 32). Farther downstream, however, in the Presidio Valley, the overall reduction in flow along the river caused it to aggrade and decreased the width and depth of the channel through time, reducing the river?s capacity for carrying water and increasing flood frequency (Everitt 1998: 662). From 1 to 4 meters of sediments collected in the Presidio Valley between 1916 and 1977 (Johnson 1977: 10).
fig 2.1-2.3 discharge rates
Río Conchos [watershed dams]
?? La Boquilla Reservoir, with a capacity (about 3000 million cubic meters) slightly greater than Elephant Butte Reservoir, completed in 1913 and located 400 km upriver from the gauge;
?? Francisco I. Madero Reservoir, with a capacity of about 350 million m3, completed in 1947 and located about 310 km upriver; and
?? Luis L. Leon Reservoir, also with a capacity of about 350 million m3, completed in 1967 and located 180 km upriver (IBWC 1997: 16, 83).
fig 2.4 flow reduction
[frequent flooding over the years was noted in many journals. ]
?trees show marks of overflow off 15 to 20 inches [38-51 cm]? near the river?s edge in Las Palomas Valley (Garrettson 1857: T17S R4W, Section 31), and while working northwest of Las Cruces, they observed ?marks on cottonwood trees? approximately 0.8 km (half a mile) from the river indicating floodwaters about 0.6 m (2 feet) deep (Garrettson 1857: T22S R1E, Section 33). Even the floodplain soil held a record of flood history, as Parry observed in the El Paso Valley in 1854: ?The body of the soil is sandy, but acquires a somewhat compact texture from the deposition of river slime, and is further enriched by the decaying vegetation that luxuriates on its moist bottoms?? (Emory 1987, vol. I, pt. 2: 7).
Bartlett (1965: 217) observed when he entered Las Palomas Valley in 1851: ?From the water marks on the trees, the river rises about four feet above its banks,
16?Historic Ecosystem Descriptions
inundating the bottom lands to the base of the hills which border them, and rendering the valley impassable.?
seventeenth-century wagon trains sometimes had to wait for weeks to cross the river at El Paso
To pave the way for later travelers, Hutton considered the potential effects of flooding when building a wagon road north of Mesilla in 1857:
?.the line here follows up the valley for six (6) miles through a low, rich bottom, thickly studded with cornfields and intersected by numerous ascequias [sic] or irrigating canals. To avoid the injurious effects of rain and the frequent overflowing of the ascequias on the heavy loam portions of this route, it was found necessary for about three (3) miles to isolate the road bed by ditches on either side, and to raise the surface by the material thus excavated; also to construct bridges over five of the ascequias?.These bridges were of the simplest description, having from eight to ten feet span, and consisting of cottonwood logs (obtained along the river bank) as stringers, and similar smaller ones as cross pieces, the whole being covered with a layer of earth eight inches deep. (Hutton 1859: 81)
Bartlett (1965: 187) recorded a relatively broad river, in the early 1850s, ?[t]he river near [El Paso] varies in width from 300 to 600 feet [91.5-183 m].?
plucked a reed, ?arundo phragmites,? and without difficulty threw it across the river
[depth around el paso was 35cm-92cm, but sometimes deeper.]
Emory (1987, vol. I, pt. 1: 50) wrote, ?I was informed, on good authority, that in the summer of 1851, a man drove a gang of mules along the bed of the river from the Presidio del Norte to El Paso. The bed was dry for nearly the whole distance?.? It should be noted that such a drying of the river was a relatively extraordinary occurrence; in this case it was evidently caused by drought coupled with agricultural demands in the El Paso Valley. Bartlett (1965: 187-8) explains, ?during the summers of 1851 and ?52, there were [no spring floods]. The river not only did not swell or overflow its banks, but in the former year it became quite dry near El Paso, all the water being transferred to the acequias?.In 1851 many large tracts of land near El Paso, which were planted in the spring, and through which irrigating canals were dug at great cost, produced nothing?at San Eleazario, twenty-five miles below El Paso,?the summer of 1852 was the first one in five years when there had been sufficient [water] to irrigate all the land of that vicinity?.
1540-1542 expedition in northern New Mexico reported that a captain explored down the Rio Grande until he reached a point where ?the river sank into the earth?; Bandelier (1976: 24)
Scurlock (1998: 40) lists the years 1681-1686 as a drought period. Follett (quoted in Ackerly 1998: 27-8) describes drying of the river in southern and central New Mexico in 1860 (or 1861), 1879, and 1889; all these dates fall within drought periods lasting from two to four years (Scurlock 1998: 40). Two other observations of a dry river above El Paso catalogued by Ackerly (1998: 27, 28)?in 1752 and 1894?followed 4 and 2 years of drought, respectively (Scurlock 1998: 40).
Significant periods during which the river dried up entirely began to intensify after the agricultural development of the San Luis Valley in Colorado at the end of the nineteenth century
[in 1902] ?The valley of the Rio Grande below 18?Historic Ecosystem Descriptions Ciudad Juarez was previously rich and prosperous. For a number of years, because of a lack of water, cultivation has had to be abandoned, and the countryside presents a desolate aspect? (Granjon 1986: 105).
In the 25 years from 1889-1914, the IBWC flow gage at Presidio recorded 34 months with zero flow in the river (Everitt 1998: 660). Much of the impetus for the creation of Elephant Butte Dam
Early explorers in the region of La Junta de los Rios (the meeting of the rivers) encountered several settlements of indigenous people; as both Everitt (1977: 21) and Kelley (1992: 122) observe, almost all of these settlements were located either at the mouth of the Conchos, or downstream of its junction with the Rio Grande. Because of the location of its watershed, the Río Conchos has a more consistent flow, and Kelley suggests that these agricultural settlements may have been limited to certain areas by the rivers? flows: ?Hence, the flow of the Río Conchos is vital to irrigated farming in the area. Significantly, all the major La Junta pueblos of the historic period were located either on the Río Conchos or on the Rio Grande at and below the junction? (Kelley 1992: 122)
However, the preferred location for an agricultural settlement does not necessarily indicate that the Rio Grande was regularly dry above the Río Conchos; the timing of the flood peaks may have simply made downriver locations preferred because they facilitated crop ripening and made multiple crops possible. The agriculture in the vicinity of La Junta was evidently almost entirely based on floodwater farming instead of ditch irrigation
[exposed to very good or very bad success with] Double-cropping was practiced by La Junta farmers as early as 1747 (Ydoiaga 1992: 82, 83).
Ackerly (1998: 30) explains, ?The floodplain of the Rio Grande contained oxbow lakes that simultaneously formed refuges for many animal species during periods of low flow?.?
Still water and water table
The floodplain of the Rio Grande historically contained numerous marshes, swamps, oxbows and pools. In addition to providing evidence of channel shifting and flooding, such aquatic features also suggest a high water table within the floodplain.
1581 valley below modern El Paso, ?a valley of swamps, which extends over eight leagues [about 21 miles]? (Gallegos 1927: 23).
[next year at the same area]
?... contains an abundance of game such as ducks, geese, and cranes?
?... numerous reeds and large marshes and pools with quantities of fish close by the river?
upriver from Selden Canyon. Luxan (1929) identifies two? ?La Cienega Helada? [named ?the frozen marsh? because the water was ice-covered when they arrived on January 23
[marshes in/near present day:}
San Marcial, Caballo, New Mexico, Socorro; Fra Cristobal, Las Palomas Valley, Hatch
1849 as being ??situated on skirt of a beautiful wood, at edge of which was a lagoon of good water?; and calling it the new settlement of Santa Barbara
??We pitched our tents in a thick grove of large cotton-woods, near which passed the acequia; while on the opposite side was a pond or laguna, extending a mile or more. As this body of water was not wider than the river, and presented many sinuosities, I think it must have been formerly the channel of the Rio Grande?.The laguna is now supplied by overflows from the river. There were many wild fowl in it; but its banks were so open, we could not approach the game? (Bartlett 1965: 217-218
Ft. Selden was established there; military planners wanted to escape the malaria-carrying mosquitoes they were encountering elsewhere (Ackerly 1998: 30).
1807 salt lake just south of present-day Canutillo
pelicans in el paso valley near San Elizario
1849 several ponds near San Elizario
close to el paso ?...There are a number of small lakes or ponds scattered about, where will be found large quantities of duck & some geese?.?
1857 Fort Fillmore = water hole
Las Cruces = pond
Doniphan?s Army in 1846, describes camping in a ?fine rush bottom, the first we have seen in Mexico worthy of notice? at the northern end of the Mesilla Valley (Bieber 1936: 222). + a few days later ?at a small pond called Alamitos? (p. 237), south of the Brazito Battlefield. + ?very brackish?we could not even make coffee fit to drink.? Gibson (1935: 309-310), + La Laguna, simply a saltwater pond, half grown over with reeds + probably just south of modern Anthony: a Butterfield Overland Mail Station
various sorts of salt deposits by numerous authors; these reports are of interest because they may be indicative of a high water table, which would result in poor drainage and high salinity, but could also create and recharge non-riverine aquatic habitats.
In 1857, surveyors noted bottom land ?covered with a white incrustation of alkali? NW of modern Mesquite, New Mexico (Garrettson 1857
Because of evaporation and the deposition of sediments and organic debris, these sorts of still water features may be relatively short-lived; Ohmart et al. (1975: 250) estimated that similar sorts of backwaters which formed along the lower Colorado River in the mid- to late-nineteenth century lasted only 50 to 70 years. Hendrickson and Minckley (1984: 168) describe the marshes of desert rivers as ?transitory communities? and note that because they are relatively short-lived, less organic material collects in these backwaters, creating less anaerobic conditions than in some other types of marshes
Belen Station of the Southern Pacific Railroad (northwest of Socorro, Texas), he wrote: It is largely irrigated and under cultivation?.The Rio Grande and irrigation acequias are lined with the almondleaf willow, Fremont cottonwood, arrowwood, and Baccharis; the rich bottom land, where uncultivated, supports a luxuriant growth of mesquite and tornillo, or screw bean; and green orchards, gardens, and fields of grain and alfalfa are situated between the river and the canal. (p. 79) (Mearns 1907: 1892 and 1894
Mesilla Valley. In it, Metcalfe (1903) identified 5 distinct plant zones arrayed across the river bottom:
?? Baccharis zone occurred on impermanent sandbars in and along the river, often flooded by one to several feet of water in spring. The seepwillow (Baccharis glutinosa) shrubs were 5 to 12 feet tall, and sometimes grew so densely ?as to make walking through it rather laborious? (Metcalfe 1903: 5). Nevertheless, some annuals grew in understory.
?? Populus and Salix zone was next to the Baccharis zone and included the cottonwood Populus fremontii Wislizeni and the willow species Salix nigra and S. interior, according to Metcalfe (Metcalfe?s cottonwood and willow species probably correspond to P. deltoides ssp. wislizeni, S. gooddingii and S. exigua of current taxonomy; see Carter 1997: 322, 324, 405). In a few places the cottonwoods and willows were so dense that the cover excluded all other vegetation including salt grass (Distichlis spicta). The cottonwood-willow zone intergraded with zones on either side and was characterized by ?[a]lmost constant surface moisture?? but not severe enough flooding to disrupt first-year trees and shrubs (Metcalfe 1903: 14). Metcalfe claims that tornillo (Prosopis pubescens) could grow in this zone, but that it was probably outcompeted by the cottonwoods and willows.
?? Tornillo zone was ?[c]onfined to old river beds or deserted flood plains??(Metcalfe 1903: 7). It covered thousands of acres, and its typical growth was dominated almost entirely by tornillo (also known as screwbean) and salt grass. It grew on what was probably the best soil for agriculture, but most agriculture occurred in the Cachanillo zone. Tornillo soil was drier and less alkaline than that of the Populus-Salix zone
?? Cachanilla zone occurred 5 to 8 feet above the tornillo zone and was dominated by arrow-weed (Tessaria sericea, now known as Pluchea sericea), with Sporobolus wrightii. Scattered honey mesquite (Prosopis glandulosa), saltbush (Atriplex canescens) and cottonwoods also occurred in this zone.
?? Dondia zone was characterized by a small, salt-tolerant plant (D. sufrutescens), now known as seepweed or sea-blight (Suaeda; see Powell 1988: 136-138). The zone was a narrow, broken strip of land where the floodplain met the mesa, ?and where all of the salts leached out of these first hills or mesas seems to come to the surface? (Metcalfe 1903: 8). Again, scattered saltbush and honey mesquite occurred in this zone.
Cotton willow stands ? Oñate?s 1598 expedition described ?the pleasant shade of the wide spreading trees which grew along the river banks? (Villagra 1933: 126-7).
The cottonwoods lining the river near El Paso ?extend[ed] a few hundred yards on each side of the banks? in 1846 (Ruxton 1973: 168
1846 ?a little timber and mesquite along the banks of the river?.?
near Valverde, in 1846 ?The timber extends half a mile from the river, and the cotton-wood trees are of large size, without any undergrowth of bushes.? (Ruxton 1973: 174)
The sexual reproduction of cottonwoods and willows is closely tied to seasonal flooding, as both produce seeds which germinate and grow best in open, moist areas (Howe and Knopf 1991: 218). Fremont cottonwood (Populus fremontii) growing along the Salt River in central Arizona release their small, windborne seeds from mid-April through mid-June, with a peak in mid-May; the timing of this peak corresponds with the end of the spring snow-melt flood on the Salt River (Fenner et al. 1985: 137). Because Fremont cottonwood seeds have a short period of viability (about 5 weeks under typical natural conditions) and require a moist soil surface for germination (Fenner et al. 1984: 57-58, Stromberg et al. 1991: 225), seedlings often appear as a spring flood diminishes. However, even in years without a large spring flood, germination can occur on deposits left by floods in previous seasons if sufficient rainfall occurs during the spring (Stromberg et al. 1991: 230).
germination requirements of cottonwood and willow seeds place them in a paradox [ cottonwood more at the dry border, less flooding/scouring therewillow closer to water, more flexible, tolerant of inundation ] (Asplund and Gooch 1988: 20, Segelquist et al. 1993: 283
[sucker, asexual reproduction > seeds] document (Rood and Mahoney 1990: 456), it does appear to occur in many species, including cottonwoods. Irvine and West (1979: 342) found that root suckers from P. fremontii were the only way cottonwoods were becoming established along the lower flood terraces of the Escalante River in southern Utah because the suckers were more resistant to scouring than free-standing seedlings
Along the Middle Rio Grande, Howe and Knopf (1991: 221) documented reproduction by root suckers among Rio Grande cottonwoods; in this instance, old trees produced suckers during years of relatively low flows. In addition to producing suckers from roots, poplars can typically resprout from stem tissues that get buried by sediments; in North Dakota, such sprouting from flood-trained saplings can result in clumps containing more than a dozen trees (Everitt 1968: 424).
Scott et al. (1996) identify three different geomorphic processes that can lead to cottonwood and willow establishment: meandering, channel narrowing, and flood deposition [all different flow regimes and stand makeup]
continuous ribbon of cottonwood and willows did not occur along the river; specific mention of groves of trees or patchy distributions were common in the early accounts.
Occasionally relatively untimbered stretches were noted
(Abert 1962: 128). Also in 1846, Ruxton described the area near San Antonio, 10 or 12 miles north of Fra Cristobal, noting ?The river is but thinly timbered here, the soil being arid and sterile??
[varying density of timber/undergrowth]
1747, a scouting party reported that the cottonwoods along the river near the lower end of the El Paso Valley (or perhaps in one of the small valleys downriver) were dense enough to make it ?possible to hide a detachment of soldiers? (Ydoiaga 1992: 71). In the El Paso Valley below San Elizario, Whiting noted that ?This trail lies through a fine tract, heavily timbered?the trees are very large? (Bieber 1938: 302). In 1854 Bartlett (1965: 186) wrote: ?The Rio Grande Valley near El Paso, and generally in other places, is thickly timbered with cotton-wood. The trees sometimes grow to a large size.? Large trees were also noted near the mouth of the Río Conchos in 1849: ?Nearly opposite the Mexican town we passed a fine part of the valley where the mesquite, the willow, and cottonwood abound in size? (Bieber 1938: 289).
In some areas, travelers noted the absence of any undergrowth under the cottonwoods. Camped south of Limitar with Doniphan?s force, Gibson (1935: 291) observed that ?the river bottoms look more like an apple orchard in ?the states? than cottonwood groves, this being the only growth.?
[more open areas near/in settlements, others few miles upstream from Fra Cristobal, Ferguson wrote ?Our encampment tonight is at the edge of a large brake, or thicket, of cane grass intermixed with willow twigs and cottonwoods, so thick as to be almost impassable? (Bieber 1936: 334).
In 1846, Ruxton (1973: 171) described the vegetation in the Mesilla Valley as ?cotton-wood, dwarf oak, and mezquite, under which is a thick undergrowth of bushes.? In 1857, a chronicler of a military expedition exploring the feasibility of using camels in the desert southwest noted at a campsite near Ft. Fillmore: ?Grass indifferent; mesquite wood abundant, especially a kind of which the camels are particularly fond, the fornia [tornillo] or screwbean? (Lesley 1949: 171).
Presidio Valley, in 1901: ?In places, patches of fine white silt form bordering sand bars; about 25 feet above these there is a second bench, covered by a growth of dark-green mesquite.
mouth of the Río Conchos made in 1683. For instance, he described ?a thick-trunked cottonwood? as a landmark for reaching a watering hole at an early stopping point along the journey (Mendoza 1952: 321). In addition, at least 10 sites with good meadows and wood sources along the south side of the Rio Grande between El Paso and the Río Conchos were noted
December of 1747, Ydoiaga (1992: 66) complained about the poor quality, ?hard grasses of a wide leaf that makes a straw if tightly twisted?
spring of 1849 Whiting observed that in the southern end of the El Paso Valley, ?the grass is generally poor? and the bottom ?[is] nearly destitute of grass? (Bieber 1938: 301).
A few years later [after 1849] , Bartlett (1965: 196) noted a similar increase in grass cover above El Paso: The bottom lands [near El Paso] are not grassy, as many suppose, but are entirely bare, save in isolated spots; hence it is necessary to drive mules and cattle to these hills and valleys to feed. There are, however, some portions of the higher valley about Frontera [8 miles above the diversion dam at El Paso] where grazing is to be found. Farther up the Mesilla Valley however, Bartlett noted that ?a large
In 1857, Hutton (1859: 86) contrasted the ?coarse? grass of the floor of the Mesilla Valley with the ?excellent? grama grass of the slopes above it
Leopold (1951: 308) suggests that the coarse grass reported in some river bottoms may have been ?fine-topped sacaton [Sporobolus sp.] , which is harsh when dry.?
Rushes were also described near an 1846 campsite 13 miles south of Doña Ana, where ?a fine lot of rushes?
?a reed, ?arundo phragmites?.?
Table 2-1. Plants collected in relatively mesic microsites of the Mesilla Valley in 1930 (from Fosberg 1940: 584-593). [pdf pg 37]
1846, between La Joya and Socorro, Abert (1962: 120) observed ??mistletoe that grows so abundantly on the cottonwood, and is called ?bayote de alamo.?
Kallstroemia grandflora, Martynia violacea, M. arenia, Talinopsis frutescens, Nicolettia Edwardsii, and several special of Boerhavia. Cereus Greggii is quite common, and the delicious fruited Cereus stramineus grows in its greatest perfection? (Emory 1987, vol. II, pt. 1: 14). Near the mouth of the Pecos River, Michler noted that ?grape-vines flourish in abundance, yielding a very palatable fruit? (Emory 1987, vol. I, pt. 1: 78).
Table 2-2. Plants recorded along the Rio Grande by Emory (1976) in 1846 [pdf pg 38]
Qualitative Descriptions of Changes in Vegetation
Sublette et al. (1998: 9) argue that pueblo-dwelling cultures along the Middle Rio Grande could have negatively impacted such habitats through activities such as wood gathering and irrigated agriculture. In the Presidio Valley, native farming settlements also impacted the environment through the practice of burning their agricultural fields to clear them for planting. In December of 1747, Ydoiaga (1992: 65) asked residents to be more careful in tending such fires because they were reducing the amount of forage available for horses and mules. Madrid (1996: 19) suggests that these fires may have affected the ecology of the Presidio Valley by suppressing the growth of some shrubs, while favoring fire-adapted grasses.
Conversion of river-bottom habitats to agricultural fields obviously reduced the acreage covered by natural vegetation
1930, Fosberg (1940: 580) noted that agriculture had probably displaced the grass Sporobolus airoides as the dominant plant of the river valley
mouth of the Rio Puerco, Turner (1966: 77) noted, in 1846, that trees were a rare and valuable resource: The country along this part of the Rio Del Norte resembles very much that on the Platte and Arkansas rivers?the sand hills showing on either side of the river, and the valleys between the bluffs about the same in width. There is little or no timber on the Del Norte: perhaps less than on either of the above mentioned rivers. There is great difficulty in obtaining sufficient fuel for cooking purposes?in some the men have to go a distance of several miles to obtain a handful of wood. Occasionally see a grove of cotton wood which is preserved with great care, as it furnishes the only material with which carts for the whole country are made. These groves are always on private property.
Before reaching Socorro, Emory (1976: 51) linked the lack of settlement with an increase in trees: Yesterday and today we came across some unoccupied strips of ground. Their number yesterday was greater than to-day; for, since we passed Pulvidera, the sand-hills encroach on the river and leave the valley scarcely a mile wide. The cottonwood, however, is getting more plentiful, and we have not been obliged to use the ?bors de vache? in cooking for some days.
[Calvin (1951: 200), in annotations of Emory?s journal, identifies ?bors de vache? as the French term for dried buffalo or cow dung used for fuel].
In the limited valley reaches of the Big Bend region, cottonwood trees ?once [were] more abundant that [they are] today,? a decline that Wauer (1980: 24) blames on woodcutters working for mines and construction projects in the late-nineteenth century. In addition, some ranching and agriculture occurred along the river in this region; following the establishment of Big Bend National Park in 1944, the riparian vegetation responded to the cessation of grazing and farming by becoming more dense, including renewed growth of grasses, cane, and mesquite (Boeer and Schmidly 1977: 214).
More often the grass along the well-traveled river was in poor condition. In 1847, Magoffin (1926: 203-4) noted that at the often-used campsite near Brazito, ?the grass has been all camped off.? Eccleston?s party experienced a similar situation when they first entered the southern El Paso Valley in 1849, and they decided to move on, in part because ?the grass also was pretty well used up? (Eccleston 1950: 127
the search for forage became so desperate for Abert?s party, that at Valverde, ?We cut down some huge cotton wood trees, and turned our animals loose to browse upon the tender bark and the twigs? (Abert 1962: 131).
Table 2-3. Non-native plants collected in the Mesilla Valley in 1930 (from Fosberg 1940: 581). [pdf pg 40]
One of the most important of these exotics was tamarisk, also referred to as salt cedar
like cottonwoods and willows it is a phreatophyte that produces seeds which germinate best on open, moist soil (Horton 1977: 124).
[spread down the river over the early 1900s]
Everitt (1998:665) argues that tamarisk was not responsible for changes in river hydrology (such as channel width or flood stage), but that it simply responded to such changes that were already underway
extensive spreading of tamarisk occurred after significant flood events, in 1929 and 1942, respectively (Crawford et al. 1993: 30, Everitt 1998:665).
Everitt (1998: 658) concludes that ?[t]here is no evidence that [tamarisk] actively displaced native species? along the Rio Grande in the El Paso and Presidio valleys. However, in some areas, its ?prodigious production of airborne seeds, long germination period, and rapid growth allowed it to compete favorably with native trees in colonizing newly stabilized sandy or silty surfaces? (Graf 1994: 99).
bird species that showed habitat preferences were found most often in cottonwood-willow or screwbean-wolfberry habitats in summer and in cottonwood-willow habitats in winter; the only bird species showing a strong preference for tamarisk-dominated habitats was the white-winged dove, which nested in these areas (Engel-Wilson and Ohmart 1978: 24-30). Rodent densities were highest in honey mesquite-dominated areas and were relatively low in the salt-cedar community (Engel-Wilson and Ohmart 1978: 39-43). Skunks, raccoons, and bobcats were the only species of large mammals observed to make significant use of the salt-cedar community (Engel-Wilson and Ohmart 1978: 46-50). While reptiles were almost never observed in the salt-cedar community and were only observed in areas where its dense canopy was disrupted by fencelines or other habitat edges, Texas toads were observed relatively often in the salt-cedar, and the shading effects of its canopy may have extended their period of activity after rains (Engel-Wilson and Ohmart 1978: 51-54).
A second exotic tree, Russian olive (Elaeagnus ausutifolia) was introduced to the region early in the twentieth century (Crawford et al. 1993: 30). less dramatic than that of tamarisk
Quantifying Changes in Vegetation
1889, Follett (1901: 195) described the lower end of the Texas portion of the Mesilla Valley as about 6200 hectares of uncultivated ?swampland.? He further detailed this acreage as including the following types of vegetation:
?? 25% ?moderately well timbered?with cottonwoods,?
?? 50% ?not overflowed? and ?composed of tornillo thickets,? and
?? 25% ?overflowed land? including bare sand bars, willow thickets, and ?flat lands, which overflow every time the river rises, and yearly produce a fair crop of cockleburs? (Follett 1901: 196).
(Table 2-4). The ?No timber? category includes areas dominated by mesquite (probably both honey mesquite and screwbean) or other shrubby growth, as well as areas with little or only low-growing vegetation.
Table 2-4. Proportion of Township and Range lines crossing 7 habitat types during Garrettson?s 1857 surveys, and McBroom and Shaw?s 1877 survey.
data probably underestimate the area covered by riparian woodlands and marsh plants that have become established in drains. However, none of the ponds, sloughs or marshes documented by the surveyors currently exist, and these sorts of natural wetland habitats have been virtually eliminated in the Mesilla and Las Palomas valleys.
Table 2-5. Proportion of survey lines crossing 7 zones in 1999.
The apparent increase in floodplain area classified as river channel between the mid-nineteenth century and 1928 may be a reflection of the heavy sedimentation that followed the increase in logging and grazing in the second half of the eighteenth century. As the channel aggraded, it became shallower and wider (Crawford et al. 1993: 20), which may account for some of the increase in coverage. In addition, methodological differences may be responsible for some of the difference between the river channel proportions reported in Tables 2-4 and 2-6
Table 2-7. Classification of floodplain between Bernardo and San Marcial, 1935 and 1989. Data from Crawford et al. (1993: Map 1).
Plant Densities and Size-Class Distributions
Table 2-8. Cottonwood tree densities (trees per hectare) in the Mesilla and Las Palomas valleys, estimated from witness tree data from Garrettson (1857) and McBroom and Shaw (1877).
Table 2-9. Number of individual plants occurring on a 0.8-hectare plot mapped by Mead (1898: 28, Plate IV).
Howe and Knopf (1991:406) explain, reductions in flood frequency and decreases in the peak flow of flood events can reduce recruitment in riparian species like cottonwood. According to Scurlock (1998: 40), the years 1824-5 were drought years, and no major floods were recorded along the Middle Rio Grande between 1823 and 1828 (Scurlock 1998: 34). In 1828 however, a ?mega event? occurred, a flood with an estimated flow of 100,000 cfs (Scurlock 1998: 32), which may have scoured small saplings and seedlings from prior years. Thus it may be that the witness trees with dbh?s of greater than 30 cm represent survivors of this drought-flood cycle, whereas the trees with smaller dbh?s became established after the 1828 flood.
Garrettson?s 1857 surveys indicates that few large, old trees existed at the time. Of 165 trees measured, the maximum dbh recorded by the surveyors was 30 inches (76 cm), which would represent an individual about 43 years old using Howe and Knopf?s (1991: 220) age-to-dbh relationship. In contrast (Figure 2-4), Howe and Knopf found at least 42 individuals (of 607 measured) with a dbh>76 cm in their 1988 study of the Middle Rio Grande valley, with a maximum age of about 67 years (1991: 219). While the difference in maximum age may simply be an artifact of sample size, it also lends support to the argument that large, old cottonwood trees may have been rare under variable flow regimes because flooding and channel migration could destroy cottonwood stands (Crawford et al. 1993: 25, 28-9; Graf 1994: 99).
1849, one traveler?s journal is particularly striking because it makes note of animals generally omitted from other accounts, including rodents, reptiles, and insects. While camped in Las Palomas Valley, Powell?s party ?discovered we were in a perfect den of rattlesnakes. All set to work beating the bushes and killing the snakes. How many we killed I do not know. They were very large, had black and white rings round their tails just above the rattles?.One I helped kill had 13 rattles and his fangs were nearly an inch long!?Besides snakes, we have rats, mice, black tailed rabbits, quails and a few centipedes around us. (Powell 1981: 102-103)
A few days later, Powell reflected on the significance of such wildlife encounters:
Our lives are in constant danger. Saw centipedes today, and a dirty yellow worm about an inch and a half long, which, those who have been in this country before say, is poisonous. The grasshoppers are singularly variegated here; some black and white; others red, yellow and black, and some I have seen that have a dress of many bright colours, putting me in mind of a clown at a circus more than anything else. (p. 105)
Oñate?s 1598 journey explained that the river?s ?bountiful waters teemed with many fish and we easily caught a number? (Villagra 1933: 127).
1773 description of life in El Paso corazon and the enguila, all of more than medium size. The enguilas [eels] are found more often in the ponds formed by the overflow of the river than in its channel? (Hackett 1902: 508-509). In 1846, large fish and eels were still being reported as quite common in the river near El Paso (Ruxton 1973: 168).
northernmost reaches of the desert Rio Grande, catfish and soft-shelled turtles were featured in some mid-nineteenth-century travelers? accounts (e.g.: Johnston 1848: 570, Bieber 1936: 337, Emory 1976: 49).
[1847-49] blue catfish weighing from 5 to 20 pounds. ?Very large and heavy? catfish, as well as soft-shell turtles, were also reported from near the mouth of the Pecos River (Emory 1987, vol. I, pt. 1: 78).
late 1890s, near El Paso, Mearns (1907: 80-81) observed that ?[o]wing to the lowness of the water in this portion of the Rio Grande at certain seasons, the fish fauna is limited and variable?.? Negative impacts have continued into more recent years, and flow reductions are not the only problem; for instance the fish fauna below the mouth of the Río Conchos declined noticably between 1977 and 1988, perhaps due to pollution from sewage (Bestgen and Platania 1988).
The loss of backwater habitats such as sloughs and ponds probably had a negative impact on species preferring pools and still water, such as eels and sunfish. Small, short-lived fish largely restricted to the mainstream flow of the Rio Grande?including the endangered Rio Grande silvery minnow (Hybognathus amarus), and 3 types of shiners (Notropis simus simus, N. orca, N. jemezanus) and one chub (Hybopsis aestivalis) that have disappeared from the river in New Mexico?were negatively impacted by reductions in flows which reduced and fragmented available habitat (Bestgen and Platania 1990: 7). Increasing levels of fine sediments could negatively impact species like the blue sucker (Cycleptus elongatus), which prefers hard-bottomed or gravelly substrates (Propst 1999: 52-53); because its eggs attach to bedrock and cobbles (Propst 1999: 52), its spawning habitat could also be affected by reductions in flow, which can allow fine sediments to clog the spaces between cobbles (Beschta and Jackson cited in Poff et al. 1997: 775).
decline preceded flow reduction
Weber (1982: 130) explains, after the opening of the Santa Fe trade in the early 1820s, American fur-trappers were quick to exploit the beaver populations in New Mexico [trapped out by 1823]
In 1826, an official complained that ?The taking of peltries of beaver is a branch of trade profitable only for the Anglo-Americans, who make up hunting parties and also establishments for them which last several months; as a result the specie will soon be destroyed.? (Madrid quoted in Marshall 1916: 259). To avoid such officials, in 1827, American trappers used rafts to float down the Rio Grande to El Paso; after trapping beavers along the river, they dismantled the rafts, sold the valuable wood, and snuck out of Mexican territory with their furs (Weber 1971: 157). In 1838, a similar raft expedition was planned to include the Rio Grande all the way to the mouth of the Pecos (Weber 1971: 224). By this time, official concern about the survival of beaver populations had reached El Paso, where authorities wrote to Chihuahuan officials about the imminent extinction of both beavers and otters along the Rio Grande (Weber 1971: 224).
[beaver populations survived and] To this day, beavers maintain a tenuous hold along some reaches of the desert Rio Grande
[otters survived but are less frequently seen]
Deer were mentioned in numerous nineteenth-century accounts [1807-1849]
Large predators?bears, mountain lions, and wolves?were mentioned in mid-nineteenth-century accounts, but only occasionally
Johnston (1848: 575) provides one of the most exhaustive lists of mammal and bird sign for a single stretch along the river, in the Engle Valley: ?In passing along the river, I saw the tracks of the otter, the catamount, the wildcat, the bear, the raccoon, the polecat, the crane, the duck, the plover, the deer, and the California quail.?
?We have got among wild animals?turkeys, deer, and bears.? (Bieber 1938: 93)
Similarly, near Fra Cristobal, Emory (1976: 55) observed ?here we saw for the first time in New Mexico, any considerable ?signs? of game in the tracks of bear, the deer, and the beaver.? Contrast that statement with one Emory made just 3 days earlier (and farther north), when he observed that ?Game in New Mexico, is almost entirely extinct, if it ever existed to any extent? (p. 54). The relative lack of game upriver may have been the result of centuries of settlement along the upstream reaches of the Rio Grande (Calvin 1951: 201).
[game] cranes, ducks, geese, turkeys, and quail) were prominent in many accounts from 1582-1902
1849, Whiting described seeing ?a flock of twenty-five huge white pelicans? in a pond near the southern end of the El Paso Valley (Bieber 1938: 300).
Table 2-10. Birds listed in the accounts of Abert (1962) and Emory (1976) from their travels along the desert Rio Grande above San Diego crossing in 1846.
Although relatively few bird species have been completely extirpated from the desert Rio Grande, many species have been impacted by human activities along the river (see Chapter 6). Finch et al. (1995: 149)
6 human-related factors that have influenced bird populations along the Middle Rio Grande, all of which also occurred along other reaches of the desert Rio Grande:
1. replacement of native vegetation with exotic woody plants like tamarisk,
2. reduction in marsh habitats,
3. reduced cottonwood regeneration,
4. loss of native habitats to agriculture and urbanization,
5. cowbird parasitism, and
6. nest disruption by humans or domesticated animals.
supported not only by the different age-classes of riparian forest stands created by natural geomorphic processes, but also by the diversity of non-forest habitats spread across the floodplain and found within the river. The dynamic and patchy nature of the riparian landscape helped support the regional biodiversity catalogued in Chapters 3-6, and thus, even the terrestrial organisms of the desert Rio Grande should be viewed in light of the hydrologic and geomorphic processes that ultimately determine the riparian landscape.
Chapter 3: Fishes of the Desert Rio Grande
Between the Rio Puerco, NM and the Devils River, TX, the Rio Grande?s fish fauna includes (Table 3-1): ?? 7 native fishes completely extirpated from the Rio Grande (including 3 extinct endemics),
?? 8 native fishes extirpated from at least part of their historic range along the Rio Grande,
?? 23 native fishes that still may occur throughout their historic range along the Rio Grande,
?? 6 species that are native to part of the desert reach and introduced to other parts,
?? 22 introduced species with established populations, and
?? 2 non-native species whose introductions failed.
In addition, there are 12 native fishes (two extinct) that occur (or occurred) in waters adjacent to the desert Rio Grande (including the Devils River)
Fourteen of the desert Rio Grande?s fish species are listed as threatened or endangered
Table 3-1. Fishes of the Rio Grande [pdf pg 60-80)
Chapter 4: Herpetofauna of the Desert Rio Grande
as many as 100 species of amphibians and reptiles may occur (Table 4-1). Of these,
?? 7 are obligate river residents,
?? 31 are facultative floodplain residents, and
?? 62 are incidental floodplain users.
?Obligate? residents include river-dwelling turtles and water snakes that require aquatic habitats. ?Facultative? residents, such as the tiger salamander, are generally associated with mesic or forested habitats, but they are not restricted to the river?s floodplain because they can also be found in other areas where appropriate conditions can be found. ?Incidental? users are species that are generally either desert-adapted species or widespread species whose occurrence along the river does not seem to be related to the presence of the river itself
Relatively few extirpations of amphibians and reptiles along the river seem to have been documented, perhaps because of both the lack of research
the New Mexico garter snake and western painted turtle appear to have been lost in the El Paso area because of habitat degredation (Garrett and Barker 1994, Tennant 1986), and Woodhouse?s Toad may have been lost from the river?s floodplain in Big Bend National Park because of flow regulation upstream (Sam Droege, pers. comm.). None of the species along the desert Rio Grande is currently listed as threatened or endangered in the United States, but 13 are listed as threatened in Mexico
Table 4-1. Amphibians and Reptiles of the Rio Grande [pdf pg 82]
Chapter 5: Mammals of the Desert Rio Grande
As many as 102 species of mammals may occur (or have occurred) along the desert Rio Grande (Table 5-1). Of these:
?? 6 are species introduced from other continents,
?? 2 are species whose populations along the Rio Grande may have been introduced from elsewhere in North America,
?? 12 appear to have been extirpated from all or part of their range along the river, and
?? 14 are federally listed as threatened or endangered in the United States or Mexico
For some species, the river, particularly the canyon-dominated stretch through and below Big Bend, has functioned as a dispersal barrier, limiting the distribution of entire species or subspecies. According to Schmidly (1977a) and Baker (1956) that stretch of the river has determined the edge of the following species? ranges: Black-tailed Prairie Dog, Southern Plains Woodrat, Mexican Ground Squirrel and Nelson?s Kangaroo Rat. They also note that for several species?Botta?s Pocket Gopher, Yellow-faced Pocket Gopher, Ord?s Kangaroo Rat, and the Eastern Cottontail?the river provides enough of a barrier to create distinct subspecies on each side.
approximately 70% of the desert?s native mammalian fauna can be found along the Rio Grande
Table 5-1. Mammals occurring along the Rio Grande [pdf pg 106-124]
Chapter 6: Birds of the Desert Rio Grande
As many as 351 species of birds regularly occur along the Rio Grande through the Chihuahuan Desert, and of these, 155 species (44%) breed in areas near the river (in either riparian, desert, or cliff habitats), while the remainder (196 species, 56%) use such areas during migration or during the non-breeding season
23 are listed as endangered or threatened
134 obligate species (38% of the total) that are restricted to aquatic or riparian habitats: these habitats include open water, shorelines, and marshes; as well as cottonwoods, willows and other phreatophytic vegetation; and urban/agricultural areas within the floodplain. Another 165 species (47% of the total) are facultative users of valley habitats; these species are generally more abundant in relatively mesic or wooded habitats (for some of these birds, habitat preferences change with the seasons)
Fifty-one species (15% of the total) are incidental users of river valley habitats, more commonly found in desert, grassland, or rocky habitats
only breeding birds, the percentages remain similar: 37% restricted to riparian habitats, 41% facultative users of riparian habitats, and 21% incidental users
Johnson et al.?s (1977) habitat use designations differ
bottom line is still clear: riparian habitats are used by the majority of bird species in arid regions
Relatively few complete extirpations appear to have occurred among the avifauna, although a number of species have undergone significant population declines or increases
Wild Turkeys were extirpated from the northernmost reaches of the desert Rio Grande and have been reintroduced to the Bosque del Apache
birds that feed or nest in backwaters or marshes have likely been affected by changes in flow regimes and channelization; Chris Rustay (pers. comm.) also notes that obligate cottonwood bosque nesters, such as the Summer Tanager and Yellow-billed Cuckoo, have necessarily declined anywhere that the bosque has been destroyed. In contrast, a number of other species (including many of the doves, grackles, and cowbirds) have evidently benefited from human modifications to the environment and expanded their ranges. These factors also interact with one another: the combination of habitat destruction and expanding populations of the Brown-headed Cowbird, and to a lesser extent, the Bronzed Cowbird, may have made some passerine species more susceptible to nest-parasitism and local breeding declines.
Tables 6-1and 6-2 were compiled mostly from bird lists available at three federal sites along the river?Bosque del Apache National Wildlife Refuge (U.S. Fish and Wildlife Service 1999), Big Bend National Park (Selleck 1994), and Amistad National Recreation Area (Sorola, n.d.)?and an El Paso area checklist (including all of El Paso County, the southern third of Doña Ana County and the western half of Hudspeth County) published by the El-Paso/Trans-Pecos Audubon Society (Donaldson et al. 1987, Zimmer 1999).
Species Accounts [pdf pg 132-137]
There is no table 6-1 or 6-2 ?
Appendix B: Threatened and Endangered Vertebrates of the Desert Rio Grande
A total of 75 native vertebrates that occur along the desert Rio Grande are listed as endangered or threatened by the federal government of Mexico (72 species) or the United States (16 species; see Table B-1). Eleven of these species are birds that have only been recorded occassionally, as vagrants, along the river (indicated with an asterisk in Table B-1; see also Chapter 6, Table 6-2). In the United States, an additional 50 species (including 9 birds recorded only as vagrants) have been listed as threatened or endangered by the states of New Mexico or Texas, although they are not listed by the U.S. federal government
Table B-1. State and Federally listed vertebrates native to the desert Rio Grande