ChapterPDF Available

Rivers of Mexico


Abstract and Figures

This chapter discusses the physical and biological features of five major Mexican rivers-the Río Pánuco, Usumacinta-Grijalva rivers, Río Candelaria, the Yaqui, and the Río Conchos. Five additional rivers-the Chihuahuan Desert's Río Salado; the Río Tamesí, which joins the Río Pánuco near its mouth; the Río Fuerte, which flows through some of the continent's largest canyons in the Sierra Madre Occidental to the Gulf of California south of the Yaqui and Mayo rivers; the Ayuquila-Armería river system, which empties into the Pacific Ocean; and the Río Lacanjá, a small mountainous tributary of the Usumacinta-are also briefly reviewed. The history of human impacts on Mexico's rivers includes many groups of prehistoric inhabitants. Mexico's major rivers are highly exploited. Construction of dams, primarily for crop irrigation in otherwise desert environments, is one of the major factors. Water pollution from discharge of domestic wastes, high salinity, and nutrients from irrigation returns, mining, and industrial wastes is widespread throughout Mexico. With the increase in population and associated land-use change and generally limited resources available for conservation of natural resources the scenario of the rivers are becoming much worse.
Content may be subject to copyright.
Ch23.qxd 3/24/05 2:17 PM Page 1030
sumption. This is especially true in the central
portion of the country, where nearly 25% of the total
population lives in approximately 1% of the
country’s total surface area in the states of Mexico
and Distrito Federal (
Mexico.html). With a population growth rate of
about 2.4% per year (Population Reference Bureau,, pressure to use water from Mexico’s
rivers is rapidly increasing. Unfortunately, the fact
that the hydrology, geomorphology, biodiversity,
and ecology of Mexico’s rivers are relatively poorly
studied is a serious shortcoming in understanding the
environmental impacts of current and future water
The history of human impacts on Mexico’s rivers
includes many groups of prehistoric inhabitants.
Various accounts suggest the earliest inhabitants
probably arrived more than 20,000 years ago and
several major Mesoamerican culture regions devel-
oped large populations with sophisticated and
complex cultures. Following the Olmec civilization
that developed along the southeastern Mexican Gulf
Coastal Plain around 4000 years ago, other major
culture regions (Maya, Toltec, Huastec, and Aztec)
developed within Mexico during the Classic period
Mexico, with an area of 1.97 million km2, has
approximately 150 rivers and appears to have an
abundance of water resources. However, the distri-
bution of these water resources is far from homo-
geneous, with northern Mexico extremely dry and
southern Mexico among the wettest areas of North
America. Located between 15°N and 33°N latitudes,
Mexico is the warmest part of North America (Fig.
23.2) but has tremendous variety in climate and
topography. The combination of its mountainous
topography producing strong orographic influences
on precipitation and the fact that it straddles the tem-
perate–tropical divide produces a diversity of runoff
patterns and river environments.
With 105 million people (Population Reference
Bureau,, or an average population
density of 53 people/km2, tremendous pressure is
placed on Mexico’s rivers for hydropower, irrigation,
waste disposal, and domestic and industrial con-
FIGURE 23.1 Rio Candelaria at bend in river, with surrounding
floodplain and karstic hills in state of Campeche. Further
investigations of basin management, pre-Hispanic floodplain
manipulation, and canalization are discussed in Siemens and
Soler-Graham (2003) (Photo by A. H. Siemens).
© 2005, Elsevier Inc. All rights reserved.
Ch23.qxd 3/24/05 2:17 PM Page 1031
(A.D. 150 to 650) (Coe and Koontz 2002). These
complex civilizations were heavily dependent on
Mexico’s rivers and water resources until the Spanish
conquest. It is now well accepted that some prehis-
toric societies had impacts on watershed processes of
various basins that extended over several millennia.
For example, O’Hara et al. (1993) found that in the
Mexican Volcanic Belt, conversion of upland forests
to traditional slash and burn agricultural land use
by prehistoric civilizations resulted in accelerated soil
erosion and associated sedimentation, with conse-
quential impacts to downstream aquatic environ-
ments, and that the sediment produced by their
actions continues to be stored within the basin. Beach
(1998) found that slash and burn agriculture and
resultant soil erosion in the Maya heartland
(Yucatán) produced distinctive soil horizons. Along
the Mexican Gulf Coastal Plain, prehistoric Maya
and Toltec practiced floodplain irrigation and
wetland manipulation for intensifying agricultural
production (Siemens 1980, 1983; Sluyter 1994;
Whitmore and Turner 2002). The Olmec were pri-
marily located within the states of Veracruz and
Tabasco, from the Tuxla Mountains to the mouths
of the Usumacinta, Grijalva, Coatzacoalcos, and
Papaloapan rivers, and developed the first conduit
drainage system in the Americas. Far to the north,
other complex cultures developed, albeit to a some-
what lesser extent and in somewhat later periods.
Doolittle (1985, 1988, 2000) provides information
on the early development of agriculture, irrigation,
and related erosion and other hydrologic control
23 Rivers of Mexico
FIGURE 23.2 The rivers of Mexico described in this chapter.
Ch23.qxd 3/24/05 2:17 PM Page 1032
systems in northwestern Mexico (Sonora and
Chihuahua) in areas where the Papago, Pima, Yaqui,
and Tarahumara (Rarámuri) peoples, among others,
still reside.
The Mayan civilization flourished from about
A.D. 600 to 900, reaching a population of approxi-
mately 5 million around 700. They inhabited the
entire Yucatán peninsula as well as the states of
Chiapas and Tabasco. Covering most of the Usumac-
inta River basin, the Maya used the river and its trib-
utaries extensively for travel and trade. To the north,
the Huasteca region, located along the east-central
Gulf Coastal Plain states of Veracruz, Tamaulipas,
and San Luis Potosí, is the northernmost major
culture region in Mesoamerica, reaching its apex in
the Post-Classic period, around 1000. West of the
Mayan region, the Aztecs dominated Central Mexico
from about 900 to 1521, from the state of Veracruz
on the Gulf of Mexico to the west coast states of
Guerrero and Oaxaca. The Aztec population had
reached an estimated 25 million when conquered by
the Spaniard Hernán Cortés in 1521. Cortés’s con-
quest resulted in Spanish rule for about 300 years
until Mexican independence was declared in 1810.
Thus, like many other regions of North America,
prehistoric humans throughout Mexico depended on
rivers for travel and food, but the complex societies
and higher densities within Mexico likely impacted
watershed processes and riverine resources more than
in the United States and Canada (O’Hara et al. 1993,
Doolittle 2000, Whitmore and Turner 2002).
In this chapter the physical and biological features
of five major Mexican rivers are described in detail
(see Fig. 23.2). The Rio Pánuco drains arid and high-
rainfall mountainous areas in east-central Mexico
before flowing into the Gulf of Mexico. The com-
bined Usumacinta–Grijalva rivers drain lush tropical
rain forest before flowing together into the southern
Gulf of Mexico. The Rio Candelaria drains the
western karstic Yucatán Peninsula. The Yaqui of
northwestern Mexico is a desert river draining into
the Gulf of California. The Rio Conchos is an im-
portant tributary of the Rio Bravo del Norte (known
in the United States as the Rio Grande, covered in
Chapter 5). Five additional rivers described in one-
page summaries include the Chihuahuan Desert’s Rio
Salado, another tributary of the Rio Bravo del Norte;
the Rio Tamesí, which joins the Rio Pánuco near its
mouth; the Rio Fuerte, which flows through some of
the continent’s largest canyons in the Sierra Madre
Occidental to the Gulf of California south of the
Yaqui and Mayo rivers; the Ayuquila–Armería river
system, which empties into the Pacific Ocean; and the
Rio Lacanjá, a small mountainous tributary of the
Physiography and Climate
Exclusive of Baja California, which has no major
rivers and is not further considered, Mexico’s phys-
iography, and thus its rivers, is dominated by its
mountain ranges. Most rivers drain eastward to the
Gulf of Mexico (Atlantic) and Carribean or west-
ward to the Pacific (including the Gulf of California).
Two major mountain systems, the Sierra Madre
Oriental and Sierra Madre Occidental, serve as the
major drainage divides for the Pacific and Atlantic.
However, between them lies the extensive Mexican
Altiplano (or Mesa del Norte), which in some
instances is characterized by having large areas with
rivers draining to closed basins that prehistorically
sometimes held large inland lakes, such as the
Lagunas Palomas and Mayrán (Smith and Miller
1986), which undoubtedly had some degree of past
interconnections. The Mexican Altiplano is bounded
to the south by the Trans-Mexican Volcanic Belt, and
further south is the Sierra Madre del Sur.
The physiography of Mexico’s Pacific slope
includes the Buried Ranges province in northwestern
Mexico (Arbingast et al. 1975). This arid and hilly
region, which includes part of the Sonoran Desert to
the north, slopes gradually to the sea as it parallels
the mountainous Sierra Madre Occidental province
along a northwest–southeast shoreline. Inland, the
Sierra Madre Occidental is the major physiographic
province of Mexico, at 1200km long, 200km wide,
and averaging 2000masl. It includes some of the
world’s largest canyons, such as the “Grand Canyon
of Mexico” or Barranca del Cobre (Copper Canyon)
of the Rio Fuerte basin. These mountains parallel
and form the western edge of the Basin and Range
province that extends from the southwestern United
States into north central Mexico and includes the
Chihuahuan Desert and Desert Grasslands (northern
portion of the Altiplano). Southward, across the
Tropic of Cancer in the southern portion of the
Altiplano, the Basin and Range rises and the Sierra
Madre Occidental declines to form the northern and
western edges of a vast tropical–subtropical plateau
that makes up the forests and steppes of the Central
Mesa province (also called the Central Plateau) and
is the southern portion of the Altiplano. Further
south, the Central Mesa ends at the Trans-Mexican
Volcanic Belt (also called the Neovolcanic Plateau
province), a chain of west-to-east-trending volcanic
mountains 900km long but only 100km wide. These
Ch23.qxd 3/24/05 2:17 PM Page 1033
mountains have numerous active volcanoes, includ-
ing Mount Orizaba (5747m), the third-highest peak
in North America. Significantly, much of the drainage
within this extensive volcanic system is interior drain-
ing, feeding many large freshwater lakes. Although
the rivers, such as the Rio Lerma, draining the
Mexico City metropolitan area and flowing into
these lakes are not further considered, it should be
noted that these lakes are distinctive within North
America and include many endemic, and often now
endangered, aquatic species. To the south of the
Volcanic Belt lies the Sierra Madre del Sur province,
which extends to Mexico’s southern border.
The physiography and drainage patterns of
Mexico’s Atlantic slope are controlled by three
mountain systems: the Sierra Madre Oriental, the
Trans-Mexican Volcanic belt, and the Sierra de San
Cristobal. The largest, the Sierra Madre Oriental, is
a major physiographic province aligned north–south
along Mexico’s Gulf Coast. These mountains consist
predominately of folded Cretaceous limestone, with
several ridges exceeding 3500masl. Between the
Sierra Madre Oriental and the Gulf of Mexico lies
the Mexican Gulf Coastal Plain province, a continu-
ation of the U.S. Coastal Plain with seaward-dipping
Tertiary and Quaternary strata of marine and fluvial
origin (Grubb and Carillo 1988) that extends to the
Yucatán peninsula. Compared to the U.S. Coastal
Plain, Mexico’s Coastal Plain has greater relief
because of the structural control imposed by the adja-
cent mountain systems, and it is much narrower,
being widest (150km) near the Rio Bravo (Rio
Grande) valley and narrowing to the south to pinch
out in central Veracruz as the Volcanic Belt intersects
the southern edge of the Sierra Madre Oriental (de
Cserna 1989). Isolated outcrops of Tertiary and
Quaternary volcanics disrupt Coastal Plain drainage
patterns and localized topography in southern
Tamaulipas and northern Veracruz. Just north of
Ciudad Veracruz a small but topographically signifi-
cant volcanic range, Sierra Punta del Morro, appears
to have been a biogeographical barrier to temperate
and tropical fauna (Contreras-Balderas et al. 1996,
Hulsey et al. 2004) as it extends to the Gulf of
Mexico. South of it the Coastal Plain broadens as it
arcs around southern Veracruz and Tabasco, fronting
the Bahia de Campeche (part of the Gulf of Mexico).
Here the Coastal Plain borders the enormous car-
bonate platform of the Yucatán Peninsula. In com-
parison to the mountainous and Coastal Plain
environments of most of the rest of the country, this
region represents a significant change in geomor-
phology and hydrology. The Yucatán is a classic
example of a karstic landscape, with low relief char-
acterized by predominantly subsurface drainage that
has produced numerous caves, cenotes, springs, and
solution depressions, and thus few large surface
rivers. To the west and south the Yucatán Peninsula
borders the Sierra de San Cristobal, a small moun-
tain system of folded Cretaceous limestone (West
and Augelli 1989) that is part of a larger system
(Chiapas–Guatemala Highlands province) that
extends into Guatemala and Belize, but which is dis-
tinct from the granitic Sierra de Chiapas of southern
Approximately half of Mexico is south of the
Tropic of Cancer (23.5°N) and within a tropical
climatic regime. The significance of this is that the
hydrological mechanisms of Mexico are distinct from
those found in the temperate and northern latitudes
of North America, and in particular are associated
with the strengthening of easterly trade winds in
summer months rather than westerly migrating
midlatitude cyclones during winter months. Located
at the interface between the midlatitudes and the
tropics, the climate of Mexico becomes increasingly
warm and humid toward the south. Mean annual
temperatures exceed 20°C in low elevations of both
the east and west coasts, and in the far south, mean
annual lowland temperatures exceed 25°C. Temper-
atures are considerably cooler in mountainous areas,
with mean annual values of 10°C to 15°C in the
Sierra Madre Occidental and the Volcanic Belt. Sea-
sonal differences in temperature are greater in the
north and at high elevations; for example, January
mean daily temperatures in the northern montane
Sonoran and Chihuahuan deserts are about 5°C,
whereas mean daily summer temperatures exceed
30°C. In southern Chiapas, however, mean January
temperatures are about 25°C, with mean summer
temperatures only a few degrees higher.
Precipitation is highly seasonal throughout
Mexico, occurring mainly from summer through
early fall with the strengthening of easterly trade
winds associated with a northerly shift in the
Intertropical Convergence Zone (ITCZ), and the
occurrence of tropical cyclones between August and
October (Metcalfe 1987). Mountain systems exert
a significant orographic influence on precipitation,
focusing Gulf moisture along the eastern mountain
flanks and Coastal Plain and resulting in a steep
west–east, low–high precipitation gradient. Mean
annual precipitation increases toward the south. The
driest regions are in the northern interior, where
mean annual precipitation is less than 50cm in
the Sonoran and Chihuahua deserts. East of the
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:17 PM Page 1034
Chihuahuan Desert near the coast, the lower Rio
Grande receives 80cm, and the greatest rainfall
exceeds 300cm in portions of southeastern Veracruz
(INEGI 1981c). Midlatitude cyclones (nortes) repre-
sent only a minor source of precipitation during
winter months, occasionally penetrating as far south
as Tabasco and Yucatán (West et al. 1969, Metcalfe
1987). Galindo (1995) suggests that ENSO (El Niño)
has an influence on seasonal precipitation patterns
in eastern Mexico. Specifically, the warm phase of
ENSO is associated with increased winter precipita-
tion along eastern Mexico due to a strengthened
meridianal circulation that results in a higher fre-
quency of midlatitude cyclones penetrating into
Mexico (Diaz and Kiladis 1992). In contrast, La Niña
years appear related to an increase in summer pre-
cipitation because of a higher frequency of tropical
cyclones (Jauregui 1995). Thus, La Niña has more
significant effects on streamflow and sediment trans-
port than El Niño (Hudson 2003).
Basin Landscape and Land Use
The variability of the Mexican climate allows this
part of North America to support diverse vegeta-
tion, including desert, tropical, subtropical, tem-
perate, and montane communities. More than 40
terrestrial ecoregions are recognized based on the
recent analysis by the Comisión Nacional para el
Conocimiento y Uso de la Biodiversidad (CONABIO)
(Ricketts et al. 1999, National Geographic, www.
Among the largest ecoregions are the Sonoran Desert
(northwest) and Chihuahuan Desert (north-central),
both characterized by creosote bush, tarbush,
mesquites, acacias, yuccas, and diverse cacti. Sur-
rounded by these deserts to the east and west, the
diverse Sierra Madre Occidental Pine–Oak Forest
ecoregion includes 27 species of conifers and 21
species of oaks. Pine–oak forest ecoregions also char-
acterize other mountainous areas, which roughly
correspond to physiographic provinces, such as the
Sierra Madre Oriental, Trans-Mexican Volcanic Belt,
and Sierra Madre del Sur. In the northeast, along
the lower Rio Bravo del Norte, is the Tamaulipan
mezquital ecoregion, with acacia, desert hackberry,
javelina bush, cenizo, common bee-brush or white
brush, Texas prickly pear, and tasajillo or desert
Christmas cactus. South of the Tamaulipan mezquital
is the Veracruz Moist Forests, the northernmost
“tropical rain forests,” characterized by many broad-
leafed species and extremely high diversity of terres-
trial plants and animals. Further south and east are
the Petén–Veracruz Moist Forests, a lowland tropical
forest dominated by Mayan breadnut, sapodilla,
rosadillo, and gumbo limbo. Various types of dry
forest ecoregions are found primarily in western
Mexico. In the middle of the continent are arid high
plateaus, such as the Meseta Central Mattoral and the
Central Mexican Mattoral.
According to a 1993 analysis, land use in Mexico
is 12% arable land, 1% permanent crops, 39%
permanent pasture, 26% forest, and 22% other
(probably mostly arid lands) (
02–3/countryp.html). The major agricultural prod-
ucts are corn, wheat, soybeans, rice, beans, cotton,
coffee, fruit, tomatoes, beef, poultry, dairy products,
and wood products. Approximately half of the
forested area is the coniferous and broad-leaved
forests of the more mountainous areas that account
for 90% of Mexico’s forest production (www. The other half
of Mexico’s forests is the tropical and subtropical
forests of southern Mexico that account for only
10% of forest production.
The Rivers
Abell et al. (2000) defined 25 highly diverse fresh-
water ecoregions on the basis of the distribution and
characteristics of Mexican river basins. The rivers
selected for discussion in this chapter cover only a
fraction of these ecoregions, but were selected to
illustrate the diversity of the country’s rivers for
which there is a reasonable amount of information.
Although the majority of Mexico’s 150 rivers drain
into the Pacific Ocean, the majority of the streamflow
is discharged from the Gulf Coastal Plain river
systems, and the latter receive more attention in this
The rivers described for northern Mexico (Yaqui,
Fuerte, Conchos, and Salado) are arid systems. The
Conchos and Salado rivers are both tributaries of the
Rio Bravo del Norte (Rio Grande). The Conchos (Rio
Conchos ecoregion), the primary drainage of the
north-central state of Chihuahua, has historically
provided the majority of the flow in the Rio Bravo
del Norte after most of the latter’s water is extracted
for agriculture and domestic uses in New Mexico.
The Salado (Rio Salado ecoregion) drains eastward
across the states of Coahuila, Nuevo León, and
Tamaulipas before emptying into Falcon Reservoir
on the Rio Bravo del Norte, northeast of the large
industrial city of Monterrey. The Yaqui (Sonoran
ecoregion) and Fuerte (Sinaloan Coastal ecoregion)
rivers both drain into the Gulf of California (Pacific
Ch23.qxd 3/24/05 2:17 PM Page 1035
drainage) after arising in the mountains of the Sierra
Madre Occidental and flowing through the Sonoran
Desert. The Yaqui drains the states of Sonora and
Chihuahua, as well as the extreme southeast corner
of Arizona, and the Fuerte drains a large part of
Chihuahua and small areas of both Durango and
Sonora, with its lower reaches passing through
Sinaloa. Further south, the Armería–Ayuquila river
system (Manantlán–Ameca ecoregion) flows directly
to the Pacific Ocean from the states of Jalisco and
Colima, south of the large city of Guadalajara.
Most rivers draining into the Gulf of Mexico
are from more mesic basins, with five of these, the
Grijalva, Usumacinta, Pánuco, Papaloapán, and
Coatzacoalcos, accounting for over 50% of Mexico’s
average annual flow discharged into ocean basins.
Here we describe the Usumacinta–Grijalva and
Pánuco in some detail. The Rio Pánuco is the only
major Mexican river located outside of southeastern
Mexico. It and its lowermost tributary, the Tamesí,
drain the Tamaulipas–Veracruz ecoregion and join
near the port city of Tampico. Their headwaters
begin in the arid Mexican Altiplano and the mesic
Sierra Madre Oriental and flow across the Coastal
Plain, draining parts of the states of Tamaulipas,
San Luis Potosí, Hidalgo, Querétaro, Mexico,
Guanajuato, and Veracruz. The largest river system
of Mexico, the Usumacinta–Grijalva (Grijalva–
Usumacinta ecoregion), drains into the southern Gulf
of Mexico near Villahermosa, Tabasco, from tropi-
cal rain forests of Chiapas and Tabasco, although
headwaters of both rivers are in Guatemala. The
much smaller Candelaria (also Grijalva–Usumacinta
ecoregion) to its northeast is one of the larger rivers
draining the highly karstic Yucatán Peninsula.
In general, Mexico’s major rivers are highly
exploited. Northern and central Mexico, which have
over 45% of the land area and roughly 60% of
Mexico’s population, have fewer than 10% of the
country’s water resources (
Mexico-geography.htm). Since arid and semiarid
regions are highly sensitive to hydrological change, it
is not surprising that easily exploitable water sources
have already been developed (Tortajada and Biswas
1997) and rivers in these more northerly regions have
become highly degraded. Many dams have been
built, primarily for crop irrigation in otherwise desert
environments (Contreras-Balderas and Lozano
1994). Water management that favors irrigation,
such as in the Rios Conchos, Yaqui, and Fuerte, often
results in a complete elimination of flow, ignoring
aquatic biological and ecological concerns. Water
pollution from discharge of domestic wastes, high
salinity, and nutrients from irrigation returns,
mining, and industrial wastes is widespread through-
out Mexico. Such problems are particularly acute in
the Pánuco and Conchos, as well as many other rivers
not described in this book, such as the Lerma, San
Juan, and Balsas. The most pristine waters, contain-
ing the greatest biodiversity and the most natural
ecosystem functioning, are usually found in the more
mountainous regions and in the tropical rain forest
of southern Mexico. Even the tropical rain forest
rivers are facing extensive exploitation, with plans
for multiple hydropower dams that would flood large
natural ecosystems as well as important archeologi-
cal sites. In particular, biosphere reserve sites in the
Usumacinta basin are in considerable peril, both
from impoundment and forest exploitation. Thus,
although many of Mexico’s rivers are already under
considerable stress, the situation appears to be
getting substantially worse, with an increasing pop-
ulation and associated land-use change and generally
limited resources available for conservation of
natural resources.
Although Mexico’s rivers are under great stress
and are poorly studied, the evidence suggests they
contain unique collections of aquatic communities
(Minckley et al. 1986, Smith and Miller 1986, Miller
and Smith 1986, Obregón-Barboza et al. 1994, Abell
et al. 2000, Arriaga-Cabrera et al. 2000). The fishes
are probably the best known group, and it is clear
that there are many species in Mexico found nowhere
else in North America. Nelson et al. (2004) list 1277
North American freshwater fishes, with 521 occur-
ring in Mexico (348 endemic), whereas 912 are
found in the United States (544 endemic) and 212 in
Canada (8 endemic). Mexican endemics thus com-
prise nearly 28% of the North American fish fauna,
and its fauna is more highly endemic (67%) than are
those of either the United States (60%) or Canada
(4%) (Miller et al. in press,
fish/data/Mexico.htm). If these numbers are scaled to
account for the differences in area among these coun-
tries, the number of Mexican fish species per km2is
found to exceed that of the United States by a factor
of 2.7 and that of Canada by a factor of 12.4.
Numbers of endemic species per unit area in Mexico
exceed that of the United States by a factor of 3 and
that of Canada by a factor of 220. Such statements
about the diversity and uniqueness of Mexico’s fish
fauna may also be true of its significantly less studied
freshwater invertebrates. For example, among the
aquatic insects, odonates (dragonflies and dam-
selflies) show similar diversity patterns, with 326
species recorded from the United States, 160 from
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:17 PM Page 1036
Canada (Bick and Mauffray 2004,
~iori), and 342 from Mexico (Paulson and Gonzalez
Soriano 2004,; the hellgrammite
genus Corydalus has only two species found north of
Mexico, but five within Mexico (Contreras-Ramos
The Pánuco is the second largest (98,227km2) of the
Mexican watersheds draining to the Gulf of Mexico
(Hudson 2000). Three major subbasins supply the
majority of the system’s runoff and sediment: the
Moctezuma, Tamuin, and Tamesí (Fig. 23.11).
Because the Rio Tamesí (19,127 km2) joins the
Pánuco near its mouth, it is a distinct system and is
considered separately. The Pánuco basin, without the
Tamesí subbasin, drains 79,100 km2. The Rio Pánuco
is formed by the confluence of the Rios Tamuin
(33,260km2) and Moctezuma (42,726km2) after
they cross the Sierra Madre Oriental. The Rio Pánuco
then meanders 185km before being joined by the
southeasterly flowing Rio Tamesí (see Fig. 23.23) at
Tampico, Tamaulipas, several kilometers before
discharging into the Gulf of Mexico. The small Rio
Topila (3114 km2) flows north within the Coastal
Plain, joining the Rio Pánuco 25km upstream of
The major prehistoric culture region associated
with the Pánuco system is the Huasteca, located in
east-central Mexico along the Coastal Plain and
eastern ranges of the Sierra Madre Oriental (Hudson
2004). Hunters and gatherers occupied the region
since at least the mid-Holocene, but there is no evi-
dence that these primitive peoples significantly dis-
turbed the Pánuco system. The Teenek are the major
Indian group associated with the Huasteca, but that
group is not considered to be homogeneous. The
Teenek migrated north along the Coastal Plain before
the Maya civilization reached its zenith. Indeed,
although the Teenek speak a primitive Mayan lan-
guage, their material culture is distinct (Ekholm
1944). Although the Huastec civilization did not con-
struct extensive irrigation projects, they settled along
the floodplains of the Gulf Coastal Plain to take
advantage of the resources, such as shellfish, and to
develop extensive floodplain agriculture. Thus, the
Huastec would have greatly modified the riparian
environments of the lower Pánuco system; however,
rather than constructing large population centers,
such as the Aztec in the Mexican Altiplano, the
Huastec settled in small dispersed villages with the
Pánuco valley serving as the major axis within the
Huastec culture region (Hudson 2004). The Huaste-
cans were in decline by the time of Spanish arrival,
and were quickly subjugated by the conquistadors,
with many fleeing to the adjacent Sierra Madre
Orientals, where subsistence agriculture continues to
be practiced. The Spanish also contributed to exten-
sive land-use change within the lower Pánuco system,
which they found particularly useful for cattle and
horse raising, and this area became the focal point of
cattle raising in the Americas.
Physiography, Climate, and Land Use
The Pánuco basin drains four major physiographic
provinces: the Central Mesa (CM), the north–south
Sierra Madre Oriental (SO), the east–west Trans-
Mexican Neovolcanic Belt (or Neovolcanic Plateau,
NP), and the Mexican Gulf Coastal Plain (CP). The
regional physiography of the basin has a strong influ-
ence on local climate within the basin, especially
spatial variability in precipitation.
The basin’s precipitation exhibits a steep
west–east gradient, becoming increasingly humid
toward the Gulf of Mexico (INEGI 1984a, 1984b).
Major mechanisms responsible for generating pre-
cipitation include trade winds and tropical cyclones
during summer months and early fall. Midlatitude
cyclones (nortes) deliver brief episodes of rainfall
during winter months, but collectively are not as
significant as summer precipitation mechanisms.
Indeed, precipitation for this portion of Mexico aver-
ages >12cm/mo from June through October (Fig.
23.12). Likewise, temperature is highly variable,
depending on altitude, but an approximate monthly
range for the entire basin is only from about 15°C in
January to 24°C in June.
The eastern Central Mesa (or Central Plateau) is
semiarid to arid, with average annual precipitation
ranging from 30 to 40cm and occurring mainly from
May through October (INEGI 1984a, 1984b). Tem-
peratures on the plateau are mild, but killing frosts
are not uncommon during the winter along higher
ridges and slopes when nortes sweep southward from
the United States (West and Augelli 1989). The ele-
vation of the Central Mesa increases south towards
Mexico City and the increase in volcanics creates a
rugged landscape with greater relief than seen in
the northern Altiplano. Near Guanajuato elevations
reach 3000masl. The major terrestrial ecoregions in
this area are the Mexican Central Mattoral and
the Meseta Central Mattoral (Ricketts et al. 1999).
Vegetation varies within this region; it is largely
Río Pánuco
Ch23.qxd 3/24/05 2:17 PM Page 1037
limited by the lack of precipitation and consists
mainly of desert scrublands, cacti, and short grass-
lands, with low forests of pine and oak in higher ele-
vations (INEGI 1984c).
The central portion of the Pánuco basin drains
the mountains of the Sierra Madre Oriental physio-
graphic province. Within its folded western escarp-
ments, headwater tributaries converge to form the
Rio Tampaón (draining from the west) and the Rio
Moctezuma (draining from the southwest), resulting
in deeply incised valleys. The mountains here are
mainly comprised of Cretaceous limestone, folded
into a series of parallel ridges and valleys. In addi-
tion to having the greatest relief, this mountainous
portion of the basin has the greatest climatic diver-
sity. Average annual precipitation increases from
35cm in the western mountains to 240cm along the
easternmost escarpments of the Sierra Madre Orien-
tal (INEGI 1981b). Temperatures are moderate at
higher altitudes, although light snows are common
during winter months along higher ridges, which may
exceed 3500masl. Desert scrub predominates along
the western ranges of the Sierra Madre Oriental.
Toward the interior of the mountains to the east
and at higher elevations a high diversity of pines and
oaks are present within the Sierra Madre Oriental
Pine–Oak Forests terrestrial ecoregion (Ricketts et al.
1999, INEGI 1984c), although there may be dra-
matic differences in vegetation along a single ridge
because of east- and west-facing slopes. With increas-
ing precipitation to the east one finds the Veracruz
Moist Forests ecoregion, the northern extent of trop-
ical broadleaf evergreen forests (West and Augelli
1989). Traditional farming practices occur through-
out the mountains, even on steep slopes, whereas
mechanized agriculture and ranching are practiced in
the larger river valleys of the Coastal Plain.
The Veracruz Moist Forests ecoregion continues
across the Sierra del Abra, the easternmost ridge of
the Sierra Madre Orientals, into the Mexican Gulf
Coastal Plain physiographic province. The Coastal
Plain extends another 90km to the Gulf Coast at
Tampico. Due to significant structural controls
imposed by the adjacent Sierra Madre Orientals, the
Mexican Gulf Coastal Plain is more complex than the
U.S. Gulf Coastal Plain. A south-plunging anticline
results in the Rio Tamesí and Rio Topila being
diverted toward the east (Trager 1926, Muir 1936)
to eventually join the Rio Pánuco just upstream of
Tampico. Most of the Tertiary deposits consist of
weakly consolidated shale, with thin beds of friable
sandstone that are highly erodible and contribute to
a more diverse Coastal Plain landscape than is seen
in the U.S. Gulf Coastal Plain. Just south of Tampico,
Tertiary deposits extend to the coast and reach a
height of 90m (INEGI 1984d, 1984e).
Land use and land cover exhibit considerable
spatial variability across the Pánuco basin due to the
west–east precipitation gradient and the regional
environmental history. There is very little large-scale
agriculture practiced within the arid western portions
of the Sierra Madre Oriental and Altiplano, where
small-scale subsistence agriculture confined to the
river valleys predominates. One exception is the
large-scale citrus production around the city of Rio
Verde in the central Sierra Madre Oriental. Toward
the eastern and increasingly humid ranges of the
Sierra Madre Oriental, extensive slash and burn
(swidden) agriculture is practiced in river valleys and
on toe and steep mountain slopes. Common crops
include corn, beans, bananas, coffee, and citrus
(Alcorn 1981). Although the forest has the appear-
ance of being natural, in many instances it is sec-
ondary or tertiary growth and is heavily managed. In
addition to corn and beans grown on small plots,
some coffee and citrus are grown within the under-
story of the forest canopy. The natural vegetation
of the Coastal Plain originally included thornbrush
savanna on terraces and lush tropical forests within
the river valleys; however, satellite imagery reveals
very little forest remaining in the Coastal Plain
(Crews-Meyer et al. 2004). Most of the Coastal Plain
is now used for cattle ranching and farming, a land-
use legacy that began in the early 1500s and con-
tinues today. Within the river valleys of the western
portions of the Coastal Plain, sugar cane and citrus
(mixed with some banana and papaya) are the major
agricultural products, and this agriculture spread
down valley during the twentieth century to displace
some former cattle ranching. In part, displacement of
ranching in favor of farming within these valleys is
due to the Mexican government and the World Bank
initiating an enormous irrigation project, designed as
one of the largest irrigation systems in Latin America.
Due to poor planning, however, the project is largely
inoperable and was never completed (Aguilar-
Robledo 1999, Hudson et al. in review).
River Geomorphology, Hydrology,
and Chemistry
The Pánuco drainage system undergoes tremendous
changes in river morphology and hydrology from its
headwaters to the coast (Fig. 23.3). From the Central
Mesa through the mountains the tributary rivers have
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:17 PM Page 1038
wide and shallow channels adjusted for the trans-
portation of bed load. Within the mountains, rivers
have narrow valleys and channels tend to braid with
laterally active channel margins. The floodplain con-
sists of narrow ribbons of deposits comprised of
coarse sediments, and lack the complexity found in
the lower portions of the basin. Because of the high-
energy setting within the mountains, floodplain
deposits are likely rapidly reworked during extended
periods of episodic flooding.
Larger rivers within the Central Mesa have flashy
discharge regimes characterized by great differences
in base flow and storm flow. Smaller streams are
intermittent or ephemeral, transporting water sea-
sonally or only after precipitation events. The Rio
Tula, the main river in the upper Moctezuma basin,
flows north from the edge of Mexico City (see Fig.
23.11). The Rio Santa Maria and Rio Verde are the
major headwater streams for the Rio Tampaón
(Tamuin) basin and form to the east of the capital
city of San Luis Potosí. Here the river drains mainly
Tertiary and Quaternary lacustrine and alluvial
deposits of the Central Mesa about 2000masl, with
isolated Tertiary volcanics having peaks of about
Within the mountains, river gradients are highly
variable but can exceed 5m/km in some locations.
Spectacular waterfalls occur where tributaries join
incised valleys. For example, the Cascada de Tamul is
a 102m waterfall at the confluence of the Rios
Gallinas and Santa Maria, forming the Rio Tampaón.
The Tampaón and Moctezuma, the two largest tribu-
taries of the Pánuco, exit the Sierra Madre Orientals
as formidable rivers. Along the eastern Sierra Madre
Oriental, before the Tampaón exits the mountains,
it encounters numerous karstic features, including
sinkholes, natural bridges, and waterfalls. Surface
drainage is often disrupted by solution cavities, man-
ifested in dry valleys and disappearing streams. This
water eventually returns as enormous springs along
Río Pánuco
FIGURE 23.3 Lower Rio Pánuco at average flow, 20km downstream of Ciudad Pánuco. Note the small boat
for scale (Photo by P. H. Hudson).
Ch23.qxd 3/24/05 2:18 PM Page 1039
the lower flanks of the eastern Sierra Madre Oriental.
For example, the Rio Coy forms from one of the
largest springs in the world (Fish 1977) and joins the
Rio Tampaón upstream of Ciudad Tamuin to become
the Rio Tamuin (Fig. 23.4).
There are fewer karstic features within the lower
Moctezuma basin. Tertiary shale units in the Gulf
Coastal Plain produce much larger sediment loads.
The Rio Tempoal, receiving runoff and sediment
from this lithology, enters the Moctezuma in the
Coastal Plain at El Higo, resulting in an immediate
increase in the sediment load of the Moctezuma and
producing café-colored waters that contrast sharply
with the clearer spring-fed streamflow of the Tamuin.
The Moctezuma transports an annual sediment load
of 4623 ¥103tons, whereas the Tamuin transports
2030 ¥103tons annually. Much of the suspended
sediment is transported in early summer and is
exhausted before the arrival of larger flood events in
September (Hudson 2003a). The two rivers meander
through wide alluvial valleys before joining to form
the Rio Pánuco in the western Coastal Plain, 185km
upstream of the Gulf of Mexico. Valley gradients
are predictably low, decreasing from the western
edge of the Coastal Plain to the Gulf of Mexico.
Between Tamazunchale and El Higo, for example,
the Rio Moctezuma has a gradient of 1.2m/km,
whereas downstream of Ciudad Pánuco the gradient
is 0.04m/km. This extremely low gradient of the
lower Pánuco valley enables saltwater to intrude as
far upstream as Ciudad Pánuco during low flows and
results in a daily tidal exchange between the large
lagoons downstream of Ciudad Pánuco (Hudson
23 Rivers of Mexico
FIGURE 23.4 Lower Rio Tampaón at low stage. The Rio Valles enters from the right. Further downstream the
Tampaón becomes the Tamuin, and after joining the Moctezuma it becomes the Pánuco. Note the small boat
for scale (Photo by P. H. Hudson).
Ch23.qxd 3/24/05 2:18 PM Page 1040
Upon entering the Coastal Plain, the Moctezuma
and Tamuin valleys widen considerably, and the rivers
develop meanders. Holocene floodplain deposits in-
crease in complexity and include deposits represen-
tative of lateral and vertical floodplain construction,
such as point bar, natural levee, backswamp, crevasse
splay, and infilled paleochannels. In several reaches the
river is in contact with Tertiary outcrop, the contrast-
ing resistance of which disrupts the planform geome-
try from a symmetrical meandering pattern. The
sinuosity (ratio of channel length to valley length) of
the Rio Pánuco averages 1.85 over its 185km channel
but shows considerable variability where the river is
reworking more resistant Tertiary deposits (Hudson
2002). The lower 30km of the Rio Pánuco is incised
into Holocene deltaic deposits with sinuosity of 1.15,
essentially a straight channel. The fine-grained clayey
bank sediments are sufficiently cohesive to resist
erosion, and the lack of sandy bed material does not
permit point bar construction. Moreover, at the lower
limits of the delta, shortly before discharging into the
Gulf of Mexico, the Rio Pánuco is incised into a ridge
of Tertiary sandstone that prevents lateral channel
migration (INEGI 1984e).
Lakes and wetlands are largely absent in the
mountains and Central Mesa but become more
common in the lower reaches of the basin, where they
are related to floodplain geomorphology (Hudson
2002). Abandoned channel courses and meander
neck cutoffs serve as oxbow lakes and arcuate
swamps, depending on the degree of infilling. Such
lakes and wetlands are common in the lower
Moctezuma and Tamuin valleys and upper Pánuco
valleys. Backswamps, evidence of long-term flood-
plain stability, are common where valley width is suf-
ficient. In the lower Pánuco and lower Tamesí valleys,
large lakes and wetlands form between meander belts
and resistant Tertiary deposits that serve as topo-
graphic drainage barriers. Although these features
were naturally created, the river water is used by the
city of Tampico and so is heavily regulated by a
system of weirs and control structures. The lagoons
and wetlands in the lower Tamesí are tidal and
important to the regional fishery, and provide vital
habitat for an array of North American migratory
Average discharge of the Rio Pánuco at Pánuco,
upstream of the Tamesí and Topila, is 473m3/s, and
the annual discharge regime strongly reflects the
regional precipitation pattern for eastern Mexico
(Hudson 2000). Discharge and runoff are uniformly
low from December through early May but increase
rapidly with the onset of summer trade winds in late
May (see Fig. 23.12). Peak discharge events usually
occur in September, often associated with an increase
in tropical cyclone activity (Hudson and Colditz
In spite of the strong seasonal pattern of runoff,
only the largest floods generated by tropical cyclones
inundate the entire valley. Mechanisms for inundat-
ing the floodplain include a rise in the water table
of the alluvial aquifer; conduits, such as crevasse
channels and paleochannels connected to the active
channel; and local precipitation (Hudson and Colditz
2003). The timing of flood events is similar for the
Rio Tamuin and Rio Moctezuma, although the more
humid Moctezuma basin supplies the majority of
flow. In addition, the hydrograph for the Tamuin is
less flashy than the Moctezuma, probably because
more of its discharge derives from groundwater and
spring-fed sources (Hudson 2003b). In comparison
to the Usumacinta–Grijalva system to the south,
the Pánuco is not as flood prone because the flood-
plain is approximately 10m above average stage.
For example, at Ciudad Pánuco, ~30km west of
Tampico, the floodplain is elevated above the water
surface by 12m where the valley crosses the axis of
an anticline. Indeed, the abundance of prehistoric
sherds along the channel banks attests to this site
long being considered less likely to flood (Ekholm
1944, Hudson 2004).
Although information on water chemistry of the
Pánuco system is not easily accessible, it is clear that
the river and its tributaries are seriously threatened
by pollution from agrochemicals, municipal dis-
charge, and salinization (Abell et al. 2000).
River Biodiversity and Ecology
The Rio Pánuco is the major river within the
Tamaulipas–Veracruz freshwater ecoregion (Abell et
al. 2000), where CONABIO has identified several
priority sites for conservation. Although information
is available on fishes of the river, other aspects of bio-
diversity and ecology are less well known.
Information on aquatic plants and major riparian
vegetation is not easily accessible for the Pánuco
system. However, nonnative hydrilla is found in some
of the oxbows and sloughs of the major valleys in the
Coastal Plain.
Information on invertebrates also is sparse,
although some information is available on mollusks,
Río Pánuco
Ch23.qxd 3/24/05 2:18 PM Page 1041
crustaceans, and hellgrammites. Crustaceans found
in the Pánuco system include the freshwater shrimp
Palaemonetes mexicanus, as well as the cave-
dwelling shrimps Troglomexicanus perezfafantae,T.
huastecae, and T. tamaulipenses. The crayfish include
Procambarus (Ortmannicus) ortmanii,Procambarus
(Ortmannicus) acutus cuevachicae, and Procambarus
(Scapullicambarus) strenghi, and mollusks include
Hydrobia tampicoensis,Littoridina crosseana, and
Lithasiiopsis hinkleyi (A. Contreras-Ramos, personal
communication). Among the aquatic insects are the
hellgrammites Chloronia mexicana and Corydalus
magnus (Contreras-Ramos 1998). Abell et al. (2000)
state that the Tamaulipas–Veracruz freshwater ecore-
gion (of which the Pánuco is a major part) has 17
endemic species of crayfish, although the specific
species are not mentioned.
The Rio Pánuco (not including the Rio Tamesí)
harbors at least 80 described native fish species, with
more than one-third of them endemic (Miller and
Smith 1986, Rauchenberger et al. 1990). Much of the
basin is remote and underexplored by ichthyologists,
and many more species (notably cichlids, catfishes,
and goodeids) surely will be discovered and described
in the near future. The much smaller and more
thoroughly sampled Rio Tamesí has at least 93 fish
species. Notable Pánuco endemics among the cichlids
include the Media Luna cichlid, blackcheek cichlid,
chairel cichlid, and slender cichlid. There are also
several endemic goodeids, including bluetail splitfin,
dusky goodea, relict splitfin, and jeweled splitfin, all
but the last endemic. Among pupfishes, the endemic,
endangered, monotypic genus Cualac is noteworthy.
Swordtails (genus Xiphophorus of the family Poe-
ciliidae) include many endemic species and are
popular in the aquarium trade and important as
research organisms (Ryan and Rosenthal 2001).
They are diverse and well studied in this basin,
including sheepshead swordtail, short-sword platy-
fish, delicate swordtail, highland swordtail,
Moctezuma swordtail, barred swordtail, mountain
swordtail, Pánuco swordtail, pygmy swordtail, and
variable platyfish (Rauchenberger et al. 1990). The
minnow family, nearing its southern distributional
limit, also is represented by numerous endemics,
especially in the genus Dionda (Pánuco minnow,
bicolor minnow, chubsucker minnow, lantern
minnow, flatjaw minnow, blackstripe minnow),
which includes several sympatric species pairs
(Mayden et al. 1992). An extremely rare and endan-
gered blind cave catfish, the phantom blindcat,
recently discovered in the Rio Guayalejo in south-
ernmost Tamaulipas (Walsh and Gilbert 1995,
Hendrickson et al. 2001) presents an interesting
evolutionary enigma (Willcox et al. 2004). Many
caves of this region harbor the blind form of Mexican
tetra, which may well be one of the most studied
nongame fishes of North American (e.g., Mitchell
et al. 1977, Langecker et al. 1995, Borowsky
1996, Jeffrey 2001, Dowling et al. 2002). Several
more fascinating cave fishes are likely to be discov-
ered in this large karstic river basin. Larger fishes
include the endemic large native fleshylip buffalo and
the endemic Rio Verde catfish.
Abell et al. (2000) mention that nonnative tilapias
(Oreochormis spp.) are a problem in this basin,
having irreversible negative impacts on the native
fauna as seen throughout much of Mexico, and we
concur (D. A. Hendrickson, personal observations).
Fish culturists have, unfortunately, also introduced
herbivorous cyprinids (grass and silver carps),
channel and blue catfish, largemouth bass, and other
centrarchids, to mention only some nonnatives
(Garcia de León et al. in press). These will undoubt-
edly impact native faunas via hybridization (catfishes
especially) and competition or habitat alteration.
A high diversity of other aquatic vertebrates is
present in the Pánuco basin, but we failed to find
basinwide compilations of aquatic herpetofauna,
birds, or mammals. Abell et al. (2000), however, state
that the Tamaulipas–Veracruz freshwater ecoregion
contains at least 16 endemic species of aquatic
Ecosystem Processes
Although there have been no ecosystem studies
done in the Rio Pánuco, this system offers an inter-
esting contrast with the River Continuum Concept
(Vannote et al. 1980). Instead of headwater streams
beginning in high-altitude forests, they begin in the
arid Central Mesa, then flow through pine–oak
mountain forests and finally through tropical rain
forests at lower elevations. The upper arid streams,
both spring fed and not, appear to be supported by
autochthonous production (with several herbivorous
fishes, such as Dionda species, and abundant grazing
snails in some streams). There is, however, likely to
be a major shift to increasing allochthonous inputs
as the major tributaries flow first through the moun-
tains and then into the Coastal Plain as turbidities
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1042
Human Impacts and Special Features
Before colonization by major civilizations, the Rio
Pánuco system must have exhibited some of the
greatest physical and ecological diversity of the
region. Draining from arid plateaus in its headwaters,
its tributaries cut through the rugged pine–oak
forests of the Sierra Madre Oriental, entered the
Veracruz Moist Forests in the eastern ranges, and
then spilled into the low-gradient tropical forests of
the Coastal Plain. Among the scenic features are spec-
tacular waterfalls, natural bridges, tremendous sink-
holes, enormous springs, and some of the world’s
most extensive and deep cave systems, dropping from
high pine and cloud forests precipitously to foothills.
The transition in physical features along this river
continuum was undoubtedly reflected in major tran-
sitions of its biological communities. Although many
of the river’s scenic physical features still exist, it is
likely that remnants of its natural aquatic communi-
ties will be found primarily in tributaries that have
been spared the exploitation and pollution found
throughout much of the system.
Because of the physical diversity and history of
land use, the Pánuco system has been impacted by
humans in many ways. It is difficult to establish
causal relationships on the current status of the
system. However, it is worthwhile to provide a brief
characterization of these different activities.
Upper arid portions of the basin are impacted by
the legacy of mining. In smaller valleys, check dams
(trincheras) are emplaced. These small structures
result in infilling of sediment and increase the arable
land area. Perhaps the most significant human
impacts in the upper reaches involve pollution.
Although Mexico City is not within the Pánuco
basin, a network of canals delivers raw sewage from
that metropolitan area to the Rio Tula, a tributary of
the Rio Moctezuma. An even greater threat to the
system, however, may be that headwaters of the basin
drain the mining district of Mexico. Old colonial
cities such as San Luis Potosí, San Miguel de Allende,
Guanajuato, and Queretero were established cen-
turies ago by Spain and extensive silver, copper,
heavy-metal, and semiprecious stone mining opera-
tions remain. Unfortunately, mismanagement of these
operations has resulted in extensive environmental
damage that is contaminating many of the regional
drainage systems, and these contaminants are likely
adsorbed to fluvial sediments transported by the
Pánuco system. Another form of mining, ground-
water pumping for agriculture, is having profound
negative impacts on some important endemic fish
habitats of the arid altiplano that are drained by this
river system.
Although dams are not as extensive as in some
other river basins of Mexico, they interact negatively
with other human impacts, and future dams are in
the planning or construction phases. In the moun-
tainous middle reaches there is a large dam on the
Rio Moctezuma, upstream of the major inputs of
streamflow and sediment on the arid western side of
the Sierra Madre Oriental. A new hydroelectric dam
is under construction in the lower Rio Tamuin (at
Ciudad Tamuin), approximately 75 km upstream of
the confluence with the Moctezuma. How this will
influence river morphology and ecology remains to
be seen, although at a minimum it will contribute to
the hydrologic fragmentation and reduce down-
stream sediment loads. In the lower reaches of the
basin there are few engineering modifications influ-
encing the drainage basin hydrology, although artifi-
cial flood-control levees were constructed along
selected reaches of the lower Pánuco after the devas-
tating flood of 1955. Here the river is a freely mean-
dering channel, as there have been no artificial cutoffs
or bank protection works constructed to reduce
lateral migration. Although the prehistoric legacy did
not leave a conspicuous imprint, agriculture during
the twentieth century extensively altered composition
of the riparian corridor. Because of the significant
fluctuation in the water table of the alluvial aquifer,
most agriculture requires irrigation. In some
instances the oxbow lakes are used for irrigation, but
primarily the water is locally pumped directly from
the channel. Chemical fertilizers and pesticides used
for sugarcane farming throughout the lower Pánuco
basin are of concern and have resulted in extensive
pollution of the lower Tampaón (Tamuin) system,
particularly downstream of Valles. More significant
impacts, however, are likely to occur due to contam-
ination of river sediments associated with the exten-
sive petroleum activities along the lower river.
Petroleum was discovered in the lower Pánuco basin
in 1901, and this is the oldest region of petroleum
development in Mexico.
Draining 112,550km2of southern Mexico, the
Usumacinta–Grijalva drainage system is the largest
in Mexico. The basin has two distinct regions,
the mountainous uplands and the Coastal Plain in
Ríos Usumacinta–Grijalva
Ch23.qxd 3/24/05 2:18 PM Page 1043
Tabasco, both receiving ample amounts of rainfall
(Fig. 23.13). The main rivers within this system are
the Usumacinta and Grijalva, both with headwaters
in Guatemala (35% of entire basin) but with 45%
of their basin in Chiapas, Mexico’s southernmost
state. The Usumacinta drains western Chiapas, serves
as the international border between Mexico and
Guatemala, and is joined by the San Pedro down-
stream of Tenosique (Fig. 23.5). The Grijalva begins
as the Rio San Miguel in Guatemala becomes the
Grijalva in Mexico and is joined by several large
tributaries, including the Rio Suchiapa, which drains
the Sierra Madre de Chiapas of western Chiapas. As
it flows toward the Coastal Plain the Grijalva also
receives drainage from the eastern margins of the
states of Oaxaca and Veracruz. The Grijalva and
Usumacinta join, but only partially, near Frontera,
Tabasco, about 15 km above their mouth in the Gulf
of Mexico (Rodiles-Hernández 2004).
The basin has a rich legacy of prehistoric human
history, and prior to the Spanish arrival supported a
dense population. Uplands in Chiapas were probably
densely populated by the Maya during that culture’s
Classic period from A.D. 150 to 650 (West et al.
1969), and the area underwent land-use change as a
consequence of slash and burn agriculture. The lower
23 Rivers of Mexico
FIGURE 23.5 Upper portion of Rio Usumacinta (Rio Lacantún) (Photo by R. Rodiles-Hernández).
Ch23.qxd 3/24/05 2:18 PM Page 1044
reaches of the basin were colonized by several dif-
ferent Indian groups and served as an important
trading link between the Aztec to the west and the
Maya in the Yucatán. Indeed, lowland Tabasco long
formed the western edge of Mayan civilization during
the Classic period, and the eastern fringe of the
Olmec heartland, the most ancient of the great
Mexican culture groups (Coe and Koontz 2002).
Most prehistoric settlement in lower portions of the
basin occurred along broad natural levees of main-
stem channels and distributaries, with higher grounds
of the levees providing safety from frequent lowlands
flooding. Prehistoric peoples residing in lowlands
practiced slash and burn agriculture of maize, beans,
squash, and various types of tubers. The wet climate
enabled cacao cultivation and that product was used
extensively throughout the region for trading (West
et al. 1969). Although there are few reliable popula-
tion estimates from the time of Spanish conquest,
archaeological evidence suggests that indigenous
populations had already declined but were decimated
by the Spanish after contact and did not recover until
the mid-nineteenth century (West et al. 1969).
Physiography, Climate, and Land Use
Three physiographic provinces are represented in
the Usumacinta–Grijalva basin (see Fig. 23.13). The
largest is the Chiapas–Guatemala Highlands (CG),
with its mountainous uplands, primarily in Chiapas.
The other major province is the Mexican Gulf
Coastal Plain (CP) to the north, mostly in Tabasco.
A small part of the basin (upper Rio San Pedro
drainage) is in the Yucatán (YU) province to the east.
Fundamental differences in lithology of the head-
waters of the Usumacinta and Grijalva result in dif-
ferences in surface erosion and drainage patterns.
The Usumacinta drains the extensive folded block of
Cretaceous limestone of the Sierra de San Cristobal,
which approaches 3000masl. This range is part of
a more extensive mountain system that reaches
3500masl in northern Guatemala, where it is known
as the Sierra Alto Cuchumatanes (West and Augelli
1989). The folded terrain of the upper Usumacinta
produces a trellis drainage pattern. As is common in
limestone regions with abundant precipitation, there
is considerable subsurface dissolution that creates
numerous karst landforms, particularly in the eastern
portions of the basin. Common topographic features
include cenotes (sinkholes), disappearing streams,
and major spring systems. In contrast to the lime-
stone of the Sierra de San Cristobal, the upper
Grijalva basin drains the granitic Sierra Madre de
Chiapas, part of a much larger mountain system
that extends throughout Central America. Along its
eastern fringes the lithology includes deeply weath-
ered Tertiary shale and sandstone that produce much
higher rates of surface erosion than are seen in the
Usumacinta system.
In contrast to the Pánuco system, which has its
headwaters in the arid and semiarid Central Mesa,
the upper reaches of the Usumacinta–Grijalva system
receive abundant precipitation (West et al. 1969).
Annual precipitation ranges from more than 400cm
along the eastern side of the Sierra de San Cristobal
to 80cm in the Sierra Madre de Chiapas highlands
of the western Grijalva basin (INEGI 1981c). Most
of the basin is influenced by a tropical monsoon pre-
cipitation regime set up by seasonal northeasterly
trade winds and tropical cyclones that persist from
early summer through early fall, although less sea-
sonality characterizes the swampy lowlands near
Villahermosa, Tabasco. The highlands also are trop-
ical, but drier in winter and less humid. Temperatures
are hot in the lowlands but decrease with altitude. At
lower elevations average daily high temperatures
range from 29.4°C to 32.2°C. At 2000masl temper-
atures are moderate for most of the year, with
average daily highs from 23.9°C to 26.7°C, but
killing frosts occur several times per year (West and
Augelli 1989). There can be wide variation in tem-
perature between mountains and lowlands, but mean
monthly temperatures from four climate stations
throughout the basin (Villahermosa, Ville Flores,
Pichucalco, Huehuetenango) give an annual mean
temperature of 23°C. When viewed on a basinwide
scale, mean monthly temperatures fall to only about
20°C in January and are consistently about 25°C
from April through August (Fig. 23.14). On the other
hand, mean monthly precipitation for the basin is
strongly seasonal, being less than 8cm/mo from
February through April but exceeding 20cm/mo
from June through October (see Fig. 23.14). Annual
precipitation for the entire basin is estimated to be
199cm based on the four climate stations.
Much of the basin was originally covered in trop-
ical rain forest, part of a larger zone of tropical
broadleaf evergreen forest that extends from the
southeastern portions of the Moctezuma (Pánuco)
basin throughout eastern Central America to repre-
sent Mexico’s most extensive forests. Large portions
of this forest have been removed for coffee planta-
tions and cattle ranching, whereas smaller plots are
used for slash and burn (swidden) agriculture. What
may appear to be pristine forest is now more likely
secondary or tertiary growth (West and Augelli
Ríos Usumacinta–Grijalva
Ch23.qxd 3/24/05 2:18 PM Page 1045
1989). In the cooler environment above 3000masl,
large tracts of evergreen coniferous trees are
common, as well as deciduous broadleaf vegetation
more commonly associated with higher latitudes. The
upper Grijalva basin drains the Chiapas Depression
Dry Forests terrestrial ecoregion, consisting largely
of pine–oak forests (Ricketts et al. 1999, www.
A larger high-elevation ecoregion to the north, west,
and east of the dry forests is the Central American
Pine–Oak Forests, including a high diversity of
endemic plant species. To the north of these pine–oak
forests is the Chiapas Montane Forests ecoregion, a
relatively narrow strip of extremely high precipita-
tion on the steep northeastern slopes of the Chiapas
The Tabascan lowlands include lush tropical rain
forests and extensive wetlands, but much of the trop-
ical rain forest in the Usumacinta–Grijalva basin is
part of the larger Petén–Veracruz Moist Forests
ecoregion, which extends to the Pánuco basin to the
northwest. Wetlands of this basin include saltwater
and freshwater marshes, many within the Pantanos
de Centla ecoregion, a seasonally flooded moist forest
with associated bogs and swamps. Closer to the
mouths of both rivers is the Usumacinta Mangroves
ecoregion, a complex system of marshes and bogs,
considered to be one of the most important wetlands
in Mexico, with mangroves that reach 30m in height.
Revenga et al. (1998) reported that land cover in
the Usumacinta portion of the two-river drainage is
59% forest, 30% cropland, 7% grassland, and 3%
developed, with a mean population density of 25
people/km2. These same authors also reported a 37%
loss of original forest. Historically, cattle ranching
and tropical agriculture (cacao, bananas, sugar cane,
henequen) were the most important industries in the
lowlands of both rivers, but they are now second to
the petroleum industry, the major economic force
within Tabasco. As with the lowlands, subsistence
agriculture is still practiced by small communities
in the mountainous portion of the basin, although
coffee production is also important.
River Geomorphology, Hydrology,
and Chemistry
The Usumacinta and Grijalva rivers are Mexico’s
longest rivers, with a combined total of 1521km
of main-stream channel in Mexico (CNA 2004, Although the Rio Usumacinta
remains unimpounded, the Rio Grijalva has two
large dams (Presa Nezahualcóyotl or Malpaso and
Presa de la Angostura or Belisario Domínguez) con-
structed in the 1960s and 1970s, and so now serves
as a major source for hydroelectric power generation.
Both rivers flow through spectacular gorges in the
upper reaches of the basin but become increasingly
sinuous as the valleys widen in the Coastal Plain.
Coastal Plain geology is characterized by a sequence
of seaward-dipping Pleistocene terraces and, closer
to the coast, Holocene deltaic deposits that onlap
Pleistocene terraces. It is here that the rivers form a
complex system of interconnected channels (West et
al. 1969), with multiple channel bifurcations that
result in no single channel carrying the entire dis-
charge of the basin. As several channels may trans-
port streamflow only during flood events, the result
is a mosaic of wetlands interconnected by arcuate
swamps and sloughs. Saltwater intrusion during the
dry season creates favorable habitat for extensive
mangrove swamps. Other types of wetlands include
freshwater backswamps, estuarine, and abandoned
channels and sloughs.
In the Tabascan lowlands, geomorphology and
flood regime of the rivers are intricately related.
Floods are of long duration, with the Grijalva the
most prone to flooding. Flooding is important
because it transports sediment overbank, forming
broad flanking natural levees that are an essential
component of the landscape. Coarse sediments (silty
sand) and higher slope of the natural levees permits
rapid drainage. For this reason natural levees do not
remain inundated long after peak flood events, and
they have therefore been favored sites for human set-
tlement over the past few millennia and continue to
be important for agriculture. Levees, however, also
increase the severity and duration of flood events by
preventing floodwaters from draining back into the
river, and because of the extensive size of the delta
plain, flooding can occur from local precipitation
collecting in the basins.
Before joining upstream of Frontera, both the
Grijalva and Usumacinta have split much of their
flows into distributaries. The Grijalva, also known as
the Rio Mezcalapa in the lower mountains, splits into
several channels after it enters the Coastal Plain. One
of the channels is the Rio Samaria, which becomes
the Rio Cañas before flowing into an extensive marsh
and wetland system, never reentering the main-stem
channel. Another distributary, the Rio Carrizal,
transports approximately one-third of the Mezcalapa
discharge and subsequently becomes the Grijalva
upstream of Villahermosa (West et al. 1969). To
make the system even more complicated, two smaller
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1046
basins, the Chilapa (7000km2) and Sierra (5180km2)
systems, flow into the Grijalva shortly before it joins
the Usumacinta (West et al. 1969). Similarly, the
Usumacinta has at least two major distributaries (Rio
Palizada, Rio San Pedro) that split from the main
channel before it joins the Grijalva. Thus, although
the Usumacinta and Grijalva rivers partially join to
form a single channel near the coast, much of the dis-
charge from these two basins never flows within a
single channel, finding multiple pathways to the Gulf
of Mexico.
The discharge and runoff of the entire Usumac-
inta–Grijalva system is thus complicated and some-
what difficult to quantify. Daily discharge of the Rio
Usumacinta before it splits into distributaries was
measured at the Boca del Cerro gauging station (near
Tenosique) from 1949 to 1983 (UNESCO website
by I. A. Shiklomanov,
water/ihp/db/shiklomanov/index.shtml), and average
discharge over that period was estimated to be
1857m3/s. In contrast to the long-term record of the
Usumacinta, only a single estimate of 821m3/s is
available from 1960 for the Grijalva (Rio Mezcalapa)
before it branches into distributaries within the
Coastal Plain (West et al. 1969). Thus, an approxi-
mate total for the combined drainage is 2678m3/s,
even though the partially combined rivers never carry
nearly this much water in a single channel on
average. A long-term estimate (most years from 1947
to 1981) of 532m3/s exists for the Rio Samaria, a
distributary of the Mezcalapa that never joins the
Usumacinta. Thus, only about one-third of the
Grijalva discharge (Rio Carrizal) joins with only a
portion of the Usumacinta discharge north of
Villahermosa before they form a single channel to the
Gulf of Mexico.
Estimates of monthly runoff are presented only
for the single site on the Usumacinta at Boca del
Cerro (UNESCO,
ihp/db/shiklomanov/index.shtml), which represents
approximately 43% of the entire two-river basin (see
Fig. 23.14). For this portion of the basin, annual
runoff is extremely high at 123cm, and monthly
runoff exceeds 10cm/mo from July through Decem-
ber (see Fig. 23.14). Although Fig. 23.14 suggests
that runoff is 62% of precipitation (estimated from
four weather stations as 199cm), this percentage is
unrealistically high. Runoff for the entire Usumac-
inta–Grijalva basin is only about 74cm because the
Grijalva basin has less precipitation and contributes
much less discharge than the Usumacinta. Thus, dis-
charge to the Gulf of Mexico is roughly 38% of pre-
cipitation, rather than 62%. This is still a relatively
high-percentage runoff for a basin with high temper-
atures and high evapotranspiration.
The Rio Grijalva has a much larger sediment load
than the Usumacinta due to high uplands erosion
rates. The Rio Usumacinta transports an annual load
of 6257 ¥103tons of sediment, whereas the Rio
Grijalva transported 24,134 ¥103tons annually
(before the dams and reservoirs were constructed;
West et al. 1969).
Little information was found on water chemistry
for the Grijalva–Usumacinta or its tributaries. Water
quality is assumed to be good in the upstream reaches
where there is little development and extensive vege-
tation, but it likely deteriorates with industrial devel-
opment in the lower reaches.
River Biodiversity and Ecology
According to Abell et al. (2000), the Usumacinta–
Grijalva basin is located within the Grijalva–
Usumacinta freshwater ecoregion, which also
encompasses the southern portion of the Yucatán
Peninsula (including the Rio Candelaria). The region
is poorly studied in terms of its biodiversity and
ecology, although some information exists on its
fishes and aquatic insects.
Riparian trees and brush of wetland forests
include Andira galeottiana,Pachira acuatica,Bra-
vaisia integerrima,Bravaisia tubiflora, blood-
woodtree, gregorywood, Paquira aquatica, willow,
and mimosa. Important plants of the mangrove
flooded zones are button mangrove, black mangrove,
white mangrove, and American (or red) mangrove
(Breedlove 1981, Ocaña and Lot 1996). Emergent
aquatic plants include bent alligator-flag, common
cattail, southern cattail, and common reed, and
American eelgrass is an important submerged species
(Lot and Novelo 1988). Floating aquatic plants
form dense covers in places: water snowflake in clear
waters, whereas dotleaf waterlily and nonnative
water hyacinth are common in stagnant water and
disturbed areas of the lower parts of the basin.
Information on the invertebrates of this system
is sparse. Among mollusks there are applesnail
(Mexican), minute hydrobe, Aroapyrgus clenchi,A.
pasionensis,Cochliopina infundibulum,Pachychilus
chrysalis, and P. pilsbryi (A. Contreras-Ramos,
personal communication).
Ríos Usumacinta–Grijalva
Ch23.qxd 3/24/05 2:18 PM Page 1047
Reports on aquatic insects are also rare, but
Bueno-Soria et al. (in press) describe the fauna in
major distributaries of the lower Grijalva system and
at other sites within this region. Taxa reported here
are all from the Grijalva system: the Rio Carrizal,
Rio Samaria, and Rio Mezcalapa–Grijalva. Although
these lists probably only include a fraction of the taxa
in this region, they give some idea of the diversity
of insect communities in these rivers. The fauna
includes six families of mayflies, five families of
odonates (dragonflies and damselflies), twelve fami-
lies of aquatic bugs, nine families of aquatic beetles,
and five families of aquatic flies, but only three fami-
lies of caddisflies. The hellgrammite Corydalus luteus
is also present in the Grijalva–Usumacinta basin (A.
Contreras-Ramos, personal communication).
Among the mayflies from Bueno-Soria et al. (in
press) were Baetidae (Baetis,Camelobaetidius),
Ephemeridae (Hexagenia), Heptageniidae (Heptage-
nia,Stenonema), Leptophlebiidae (Leptophlebia,
Traverella), Polymitarcyidae (Campsurus), and Tri-
corythidae (Leptohyphes,Tricorythodes). Odonates
included Gomphidae (Archaeogomphus,Phyllocycla,
Progomphus), Libellulidae (Libellula,Miathyrria
marcella,Pachydiplax,Tauriphila), Protoneuridae
(Protoneura,Neoneura), Calopterygidae (Hetae-
rina), and Coenagrionidae (Argia,Argiallagma [=
Nehalennia], Heteragrion,Zonagrion). The bugs
included Belostomatidae (Belostoma), Corixidae
(Tenagobia), Gelastocoridae (Nertha), Gerridae
(Rheumatobates,Trepobates), Hydrometridae
(Hydrometra), Pleidae (Paraplea), Veliidae (Microv-
elia,Platyvelia,Rhagovelia), Macroveliidae (Macr-
ovelia), Mesovelidae (Mesovelia), Gerridae
(Metrobates,Neogerris), Naucoridae (Ambrysus,
Pelocoris), Nepidae (Ranatra), and Notonectidae
(Martarega). Caddisflies included only Hydropsy-
chidae (Smicridea), Hydroptilidae (Neotrichia,
Ochrotrichia), and Leptoceridae (Nectopsyche), but
the hydropsychids were found at most sites. Bueno-
Soria et al. found more genera of aquatic beetles than
any other order, including Hydrophilidae (Anacaena,
Berosus), Dytiscidae (Brachyvatus,Laccophilus,
Pachydrus,Thermonectus), Hydrophilidae (Derallus,
Enochrus,Helochares optiata,Paracymus,Tropis-
ternus ovalis), Haliplidae (Peltodytes), Hydraenidae,
Gyrinidae (Gyretes boucardi), Hydrochidae
(Hydrochus), Limnichidae, Noteridae (Hydrocan-
thus,Suphisellus), and Scirtidae (Ora). Chironomi-
dae were the most widely distributed dipterans, but
they were not identified beyond family by Bueno-
Soria et al. (in press). In addition to the taxa collected
in the river segments mentioned already, Bueno-Soria
et al. also provide more extensive lists of the aquatic
insects of Tabasco.
Miller (1986) mentions a total of 115 fish species
known from the Grijalva–Usumacinta system in
Mexico (Minckley et al. 2005), but a recent compi-
lation for the entire state of Chiapas documents
the presence of 111 species. These are from at least
52 genera in 29 families, with 76 (74%) species
native freshwater (primary and secondary), 18 (17%)
marine, and the remainder marine forms now iso-
lated in freshwater. Four families contain 58% (68)
of the species: 33 cichlids (30%), 22 poeciliids
(16%), 9 characids (8%) and 4 profundulids (3.6%)
(Rodiles-Hernández 2004). These same families
provide most of the large number of endemic species
(60 to 70). Noteworthy species are from the Characi-
dae (longjaw tetra), Profundulidae (headwater killi-
fish), Poeciliidae (widemouth gambusia, Chiapas
swordtail, sulphur molly, upper Grijalva livebearer),
and Cichlidae (white cichlid, Angostura cichlid,
Montechristo cichlid, Usumacinta cichlid, freckled
cichlid, Teapa cichlid). Among the euryhaline species
are threadfin shad, longfin gizzard shad, Maya sea
catfish, freshwater toadfish, Gulf silverside, Maya
needlefish, Mexican halfbeak, Mexican mojarra,
and freshwater drum. In addition, there are 11
endangered species: Pénjamo tetra, Lacandon sea
catfish, pale catfish, Olmec blind catfish, Chiapas kil-
lifish, Palenque priapella, Yucatán molly, Chiapas
cichlid, tailbar cichlid, Petén cichlid, and Chiapa de
Corzo cichlid.
Nonnative species from four families have been
introduced primarily for aquaculture: Cyprinidae
(common carp and grass carp), Salmonidae (rainbow
trout), Centrarchidae (largemouth bass), and Cichli-
dae (blue tilapia, redbelly tilapia, Nile tilapia,
Mozambique tilapia, and jaguar guapote) (Rodiles-
Hernández 2004). Some species are fished commer-
cially, such as tropical gar, common snook, blue
catfish, white mullet, giant cichlid, and tilapias.
Though the diverse fish fauna of the Usumac-
inta–Grijalva has been relatively well studied, recent
discovery in the Rio Lacantún (upper Usumacinta) of
a new species in an entirely new catfish family illus-
trates how little we know of this portion of North
America (Rodiles-Hernández et al. 2000, Rodiles-
Hernández et al. 2004). This large species (up to
500mm standard length) is relatively common, with
its description based on over 30 specimens, some
of which were obtained from local residents who
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1048
include it in their diets. Its evolutionary origins
remain enigmatic, with extensive and detailed analy-
ses of morphology and DNA sequence data failing
to reveal any obvious relationships to any of the
world’s other catfish families. Obviously, if some-
thing so unusual, so large, and so obvious can remain
unknown to science for so long in this remote basin,
many more discoveries can be anticipated.
Amphibians and reptiles associated with riparian
habitats of the Usumacinta–Grijalva system include
river crocodile, swamp crocodile, common snapping
turtle, tortugas blanca, and tortugas casquito.
According to Abell et al. (2000), there are 82 species
of native aquatic herpetofauna in the Grijalva–
Usumacinta freshwater ecoregion of which 12 are
endemic. Mammals include neotropical river otter,
West Indian manatee, water opossum, tepezcuintle,
greater bulldog bat, and Baird’s tapir (March et al.
1996, Rodiles-Hernández et al. 2002).
Ecosystem Processes
There are no known studies of ecosystem
processes in the Usumacinta–Grijalva system.
Although much of the region is forested, ecosystem
processes are likely to be quite variable, particu-
larly between the high-gradient uplands and the
influence of flooding in the low-gradient Tabascan
Human Impacts and Special Features
The Usumacinta–Grijalva system is not only
Mexico’s largest river system but probably retains
more invaluable natural features than any other large
system, as well as having great archeological value as
the center of the ancient Mayan culture. It contains
part of the largest remaining tropical rain forest
north of the Amazon and, in recognition of its impor-
tance, several parks and biological reserves have been
established in the basin, including the Pantanos de
Centla Biosphere Reserve (a RAMSAR wetland),
Laguna del Tigre National Park in Guatemala (also
a RAMSAR wetland), Selva Maya (Maya Forest),
Montes Azules Biosphere Reserve, Lacantun Bios-
phere Reserve, and Maya Biosphere Reserve in the
Petén of Guatamala (the region’s largest “protected
Although the Usumacinta and Grijalva rivers
include some of the most remote areas of Mexico, the
drainage systems are impacted by anthropogenic
activities. Deforestation is occurring in upper por-
tions of the drainage basin, resulting in loss of
Chiapas’s once extensive tropical forests. The
Usumacinta continues to be a free-flowing river, but
the Grijalva has two large hydroelectric dams that
dampen the streamflow regime and remove large
amounts of sediment from the system. Unfortunately
there have not been any detailed studies to examine
the influence of these activities on the lower reaches
of the watershed, particularly the hydrology and
riparian ecology. Studies on this subject in other
regions typically report a loss of plant and aquatic
diversity associated with a change in the flood regime
and modification to the stream channel morphology.
In the lower reaches of the basin, where the Usumac-
inta and Grijalva interconnect through multiple
bifurcating distributaries, the extensive wetland
complex is being impacted by engineering activities.
Several larger channels have been straightened to
reduce flooding, and dikes have been constructed
along the channel banks to maintain a navigable
depth for transport of raw materials and agricultural
products. However, the petroleum industry has prob-
ably created the most substantial damage in the low-
lands. Oil pipelines and canals have drained many
wetlands, and dredge spoil has created topographic
barriers that alter wetland hydrology. In many cases
the reduction of water and sediment to floodplain
and deltaic wetlands results in dramatic changes
to the ecology and hydrology of these sensitive
A major concern for the undammed Usumacinta
River is that construction of a series of dams on the
main stem appears imminent. Since at least 1987
construction of a large dam at Boca del Cerro, about
9km south of Tenosique, has been publicly discussed.
In 1987 and 1992, proposals to build a dam at Boca
del Cerro met with stiff opposition and were can-
celled. In 2002, however, evidence surfaced that a
new plan was being developed by Mexico’s Federal
Commission of Electricity (CFE) for a hydroelectric
dam at this site. It appears an agreement was reached
between the Mexican and Guatemalan governments
that calls for a dam at Boca del Cerro and a series of
five upstream dams; however, little information has
been released. This plan is on the watch lists of many
environmental and archeological organizations, and
major international protests are anticipated. Such a
dam or series of dams on the Usumacinta would have
enormous ecological, archeological, and sociological
impacts, potentially flooding thousands of square
kilometers of tropical rain forest rich in unexplored
biodiversity, as well as ancient Mayan ruins and arti-
facts. Upstream and downstream impacts on riparian
ecology could be enormous.
Ríos Usumacinta–Grijalva
Ch23.qxd 3/24/05 2:18 PM Page 1049
The Rio Candelaria drains 10,755km2of the western
Yucatán (Gunn et al. 1995) and is one of the few
significant rivers within the Yucatán Peninsula, an
enormous karstic region of southeastern Mexico (Fig.
23.15). The river flows over 250km through dense
jungle and wetlands to the coastal zone, where it
empties into the Laguna de Terminos, a large brack-
ish lagoon in western Campeche (see Fig. 23.1). Most
of the basin is in Mexico, although 50km of channel
extends into the Petén of northern Guatemala to
account for 1158km2in that country (Gunn et al.
1995). The Candelaria is a large river for the Yucatán
and should be considered atypical of that region;
most of the Yucatán lacks significant rivers, as its
hydrology controlled primarily by interior-draining
cenotes aligned along large faults (INEGI 1981a).
The ancient Maya occupied the Yucatán land-
scape from 3000 to 1100 years ago, and at the height
of their civilization had a large population within
the Candelaria region. Ancient Mayan ruins located
throughout the region have become popular destina-
tions for international tourists. Although the Maya
located throughout the Candelaria basin, the major
ancient Mayan population centers were located in the
upper basin, in the Petén of Guatemala and at
Calakmul within the Maya Biosphere Reserve.
Because the basin drains the old Mayan heartland, it
is not surprising that surficial landscape processes
have been anthropogenically modified for several
thousand years (Pope and Dahlin 1989, Beach 1998).
Landscape modification to exploit water resources
was necessary because the permeable and fractured
limestone results in the water table being as much as
100m beneath the surface (Lesser and Weidie 1988).
The large Mayan population was decreasing by the
post-Classic period (900 to 1500), before the Spanish
conquest. The Spanish did not conquer the Yucatán
until a couple of decades after conquering the Aztecs
in central Mexico, and the population quickly plum-
meted, but by the mid-nineteenth century was recov-
ering. Because of the absence of significant mineral
wealth, this region remained relatively undeveloped
until the twentieth century, and in comparison to
central Mexico remains largely indigenous. There was
even a movement for independence after Mexico
became independent from Spain. The Yucatán did
not have extensive agriculture; however, henequen, a
fibrous product from the agave plant used for manu-
facturing rope, has long been produced. Although its
impact on hydrology would be difficult to speculate,
the conversion of dense forest to henequen represents
a significant land-use change for portions of lowland
Yucatán, principally in the northwestern Yucatán
state of Campeche.
Physiography, Climate, and Land Use
The Rio Candelaria flows through two major phys-
iographic provinces, the Yucatán (YU) and the
Mexican Gulf Coastal Plain (CP) (Grubb and Carillo
1988). More specifically the upper portion of the
Candelaria basin is within the Southern Hilly Karst
Plain of the Yucatán, where most of the surface
drainage of the Yucatán is found (Lesser and Weidie
Few rivers drain the Yucatán because, as is
common in karst settings, high infiltration and fea-
tures such as swallets and cenotes prevent develop-
ment of large surface drainage systems. Smaller
rivers are often not connected to a surface drainage
network, but instead are often diverted into solution
depressions. In these settings the concept of a surface
drainage divide may be inadequate because of the
numerous subsurface passages that transport water
across surface drainage boundaries. The northern
Yucatán has a large concentration of cenotes, but
almost no surface water features (Lesser and Weidie
1988). In comparison, the Candelaria region is
referred to as the “Lake District” by Gunn et al.
(1995) and has a higher degree of surface drainage
features, such as poljes, large solution depressions
that form lakes or wetlands. In this portion of the
Yucatán small river channels connect these features
and comprise the Candelaria drainage network.
Three rivers, the Caribe, Esperanza, and upper
Candelaria, comprise the Candelaria headwaters, all
funneling water from irregular networks of intercon-
nected wetlands and small rivers toward the north-
west, where they converge to form the main channel
of the Candelaria on the Coastal Plain. These head-
waters all drain the swampy and marshy Southern
Hilly Karst Plain physiographic region. The Rio
Caribe begins in the east, near the border between
the states of Campeche and Quintana Roo, west of
the Maya Biosphere Reserve. Like the rest of this
headwater region, the lithology here consists of hor-
izontal bedded Eocene and Miocene carbonate
deposits, including limestone, dolomitic limestone,
and dolomite (Lesser and Weidie 1988). Here the El
Tigre River flows through a poorly drained landscape
with extensive wetlands. East of the swampy low-
lands that make up the poorly defined drainage
divide are the headwaters for the Rio Hondo, the
largest Mexican river draining into the Caribbean
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1050
that serves as the international border with Belize.
The Xpujil Hills, with a maximum elevation of
375masl, appear to represent the drainage divide. In
general these conical karst hills are ~100m above
the swampy plain (INEGI 1998b, 1998c) and are a
testament to the intensive chemical weathering that
has shaped this landscape. The Rio Esperanza drains
the southeastern portion of the basin, extending to
near the Guatemala–Mexico border (INEGI 1998b,
1998c). Finally, the uppermost reaches of the Rio
Candelaria extend into the Petén of Guatemala,
where the drainage divide between it and Rio San
Pedro (which ultimately flows into the Rio Usumac-
inta) lies in a swampy wetland and is difficult to
delineate. These three major headwater rivers of the
Candelaria flow northwesterly, converging near the
boundary between the Southern Hilly Karst Plain and
Mexican Gulf Coastal Plain, which consists of
Quaternary sediments of fluvial and marine origin.
From here the river flows generally north, although
several resistant outcrops disrupt this pattern in the
vicinity of the small town of Candelaria.
The major soil type in the Southern Hilly Karst
Plain portions of the basin is rendzina (INEGI
1998a), thin clayey soils rich in organics (humus) and
calcium carbonate, reflecting the limestone parent
material. Caliche horizons are common and are an
indication that the intensive chemical weathering that
occurs during the rainy season is followed by rapid
evaporation in the dry season. Like the uplands, soils
within the Gulf Coastal Plain are very clayey, but
these vertisols and gleysols are much deeper. Slicken-
sides, shiny pressure surfaces on soil peds, are an
indication of substantial subsurface soil churning.
Slickensides are common in Coastal Plain soils and
also in the deeper upland soils at the base of hills or
solution cavities (INEGI 1998a, Beach 1998).
The climatic regime for the Candelaria watershed
is tropical monsoon, with distinctive wet and dry
seasons (West and Augelli 1989). Average annual
precipitation from 1951 to 1980 was 150cm at the
village of Candelaria, and this does not likely vary
too much throughout the basin because of the
absence of significant topography (INEGI 1981a).
The majority of precipitation falls between June and
October with the onset of summer trade winds and
tropical cyclones (Fig. 23.16). Average temperature
over the same 29-year period was 24.6°C (Gunn et
al. 1995), and monthly mean temperatures range
from 21°C in December to 28°C in May.
The Rio Candelaria primarily drains from two
terrestrial ecoregions in its upper reaches: the
Yucatán Moist Forests and the Petén–Veracruz Moist
Forests (Ricketts et al. 1999). Here the climate sup-
ports growth of a dense tropical broadleaf rainforest,
part of a much larger forest that extends into Belize
and Guatemala and throughout southeastern
Mexico. Wetland vegetation in the uplands occurs
within solution depressions, varying broadly in com-
position between perennially and seasonally inun-
dated basins. As the river approaches the coast it then
passes through the Pantanos de Centla and Usumac-
inta Mangroves ecoregions. The Pantanos de Centla
is seasonally flooded moist forest with associated
wetlands. In the lower reaches these freshwater wet-
lands merge with saltwater wetlands along the coast
that are dominated by mangroves.
Land use in the upper portions of the basin is
limited to small-scale traditional farming, with maize
most important. Logging and ranching are more
important in the lower reaches of the basin, creating
a mosaic of land-cover types as viewed from satellite
imagery. There is little industrial development in this
River Geomorphology, Hydrology,
and Chemistry
The channel morphology of the Rio Candelaria basin
reflects regional hydraulic and sedimentary controls.
The course of these rivers is controlled by bedrock
fractures and solution cavities. Because of an absence
of sand, the bank material is comprised of resistant
cohesive clayey sediments that result in a narrow
and deep channel. The Rio Caribe, for example,
has an average depth of 3.8m along its lower 40km.
Combined with the low energy, this setting does
not permit the development of a meandering pattern.
The river becomes much larger in the Coastal
Plain, where it receives drainage from the Rios Caribe
and Esperanza (INEGI 1998b, 1998c). River depths
in the lower Rio Candelaria vary from 7.2 to 11.3m
(Gunn et al. 1995).
Karst river floodplains differ from rivers trans-
porting clastic sediments. Because of greater channel
stability, a less variable flow regime, and smaller sed-
iment loads, these rivers tend to develop thick back-
swamp deposits but have smaller natural levees (Pope
and Dahlin 1989). After the headwaters form the
main channel, the river flows as a single channel for
90km within a defined alluvial valley. Several kilo-
meters downstream of the small town of Candelaria
the river becomes an anastomosing channel, flowing
within a network of channels for 35km (INEGI
1998b, 1998c). This probably reflects an increase in
valley width coincident with a reduction in valley
Río Candelaria (Yucatán)
Ch23.qxd 3/24/05 2:18 PM Page 1051
gradient (e.g., Nanson and Croke 1992). A transition
from anastomosing to meandering occurs toward the
coast, and the river flows as a sinuous channel for
25km until it reaches brackish coastal wetlands and
discharges into the Laguna de Terminos. Before
entering the Laguna de Terminos, the Candelaria and
Rio Mamantel, a smaller river to the north, enter into
the much smaller but well-defined Laguna Pandau.
The monthly streamflow pattern for the Rio
Candelaria reflects the seasonal precipitation regime
but also regional hydrogeologic controls. Although
most tropical wet and dry rivers encounter great vari-
ability in streamflow, the Rio Candelaria’s discharge
regime is dampened by the substantial amount of
base flow supplied by groundwater, and by reduction
in runoff due to the wetlands in the uplands (Gunn
et al. 1995). Average daily discharge over the 29-year
period, from 1958 to 1990, was 46.2m3/s. The
highest (105.5m3/s) and lowest (18.5m3/s) average
daily discharges occur in October and April, respec-
tively. Extremes ranged from a low of 8.7 m3/s in
April 1975 to 309.2m3/s in October 1963 (Gunn et
al. 1995). Large events are generally associated with
intense rainfall events from tropical cyclones. In spite
of the high precipitation for the basin, runoff is much
lower than that calculated for the Usumacinta–
Grijalva just to its southwest (compare Figs. 23.14
and 23.16). The highest monthly discharge of the
Candelaria in October results in a runoff of only
2.8cm/mo compared to a runoff of 20cm/mo in the
Usumacinta. Such large differences appear to be due
to groundwater losses in the Candelaria that eventu-
ally return much of its water to the sea through sub-
marine springs.
Surface-water features within the Candelaria
basin are distinct from the Pánuco and the Usumac-
inta–Grijalva systems. In contrast to these two large
systems, the majority of wetlands in the Candelaria
occur in the upper reaches of the watershed within
the swampy karstic plain that Pope and Dahlin
(1989) refer to as the Yucatán Lake District. These
features are poljes, which are large, irregular-shaped
karstic solution depressions that lack the cylindrical
morphology of cenotes. They are classified as either
perennial or seasonal wetlands and further classified
as forest, thicket, or herbaceous, which in part
depends on the amount of sediment filling them.
Collectively these features disrupt traditional surface
flow paths and dampen the streamflow regime of
the Rio Candelaria; however, individually they may
undergo great seasonal variation in hydrology.
Wetlands below ~100m asl tend to be perennial,
because they are close to the regional water table, and
those above ~100m are typically seasonal, repre-
senting perched aquifers supplied by local precipita-
tion and runoff. Evaporation, seepage, and channel
outflow remove water from these systems, and there
is tremendous variability in their ability to hold
water. An important characteristic of these features
is the degree to which they have been infilled by sur-
ficial sediments. Where infilled with clay, these fea-
tures may hold water for much of the year, and if
supplied by a channel may even flood adjacent areas
during the rainy season. The rapid change in water
results in considerable shrinking and swelling of
wetland soils. The clay expands during the rainy
season and contracts and cracks during the dry
season, thus limiting vegetation to those plants that
can adapt to such a harsh environment.
Although water-chemistry data specific to the Rio
Candelaria were unavailable (Arriaga-Cabrera et al.
2000), the karstic geology of the region should result
in high values for alkalinity/hardness and pH values
greater than 7.
River Biodiversity and Ecology
The Rio Candelaria is classified by Abell et al. (2000)
in the Grijalva–Usumacinta freshwater ecoregion
rather than the Yucatán ecoregion. They point out
that the freshwater biota is largely unexplored in
the Yucatán, and this is certainly true for the Rio
Wetlands in the lower alluvial valley are located
mainly in backswamp environments. The composi-
tion of vegetation on natural levees is similar to
upland forests because the coarser sediments and
slight elevation of these deposits provide better
drainage. Toward the coast, freshwater marsh merges
with brackish swamps, much of which are predomi-
nantly mangrove, but support a variety of salt-
tolerant species (Pope and Dahlin 1989).
Information on invertebrates in the Candelaria
is sparse, although the narrowmouth hydrobe (a
snail), the hellgrammite Corydalus bidenticulatus,
the stonefly Anacroneuria, and the true bug Abedus
have been found (A. Contreras-Ramos, personal
Ayala-Perez et al. (1998) provide the only pub-
lished summary of the fish fauna of the Rio
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1052
Candelaria, but their work was based entirely on
trawling samples in the shallow (3m) 14km2termi-
nal estuary (1 to 22ppt salinity) of this system,
Laguna Panlau, which is fed by both the Rio Cande-
laria and the Rio Mamantél. The economically
important fish faunas of many similar saline lagoons
associated with the Laguna Terminos are relatively
well studied, and their study was typical of that larger
community. All 50 species were from 24 primary
marine families, with the exception of two cichlids
(Mayan cichlid and one unidentified species) and the
threadfin shad. Dark sea catfish, bay anchovy,
rhombic mojarra, checkered puffer, silver perch,
ground croaker, sand seatrout, and spotted seatrout
were the dominant species. We found no published
reports on the fish fauna of the upstream freshwater
system; however, a search of museum collections pro-
duced a list from a single collection (University of
Michigan) from a tributary of the Rio Candelaria at
a highway bridge 40km southeast of Candelaria in
1982. Not surprisingly, the fauna of this system is
closely related to that of the Rio Grijalva. Only one
of the 17 species collected is not also recorded from
the Grijalva and only one of the species collected here
was also taken in the estuary by Ayala-Perez et al.
(1998). Nine species of cichlids dominated the 1982
collection, comprising numerically about 75% of the
222 specimens: firemouth cichlid, blackgullet cichlid,
yellow cichlid, yellowjacket, Mayan cichlid, yellow-
belly cichlid, Montechristo cichlid, redhead cichlid,
and giant cichlid. Cyprinodontiformes included stip-
pled gambusia, Champoton gambusia, shortfin
molly, picotee livebearer, and pike killifish. The two
characids were banded tetra and Maya tetra, and a
single catfish specimen was pale catfish.
Ecosystem Processes
As pointed out by Arriaga-Cabrera et al. (2000),
very little or almost nothing is known about most
aspects of this river system. No ecosystem studies
have been done on the Candelaria but it would
appear to present an interesting contrast in its hydro-
logical and organic matter budget to most rivers.
Human Impacts and Special Features
Occupation of the Candelaria basin by the ancient
Maya resulted in significant anthropogenic landscape
disturbance much earlier than in most other regions
of Mexico. Until the last few decades, however, this
area had received relatively little human disturbance
for several hundred years. Thus, much of the basin
retains features of a relatively pristine landscape.
Although it is part of the vast karstic Yucatán, which
loses most of its water to groundwater seepage, the
Candelaria is unique in retaining sufficient surface-
water flow to represent one of the largest Yucatán
rivers. Although its biodiversity and ecological char-
acteristics are poorly known, this unusual hydr-
ological system, with anastomosing channels and
wetlands in its headwaters, its meandering channel in
the Coastal Plain, and its coastal mangroves, is an
unusual river ecosystem.
The Maya modified their landscape in two major
ways: forest clearing associated with slash and burn
agriculture and manipulation of the surface hydrol-
ogy for agriculture, which included canals and ter-
races. The immediate impact of Mayan land-use
practices was to increase soil erosion and
lake–wetland sedimentation. This occurred on even
moderate slopes, and several scholars have identified
a distinctive layer of fine sediment associated with
wetland infilling called the “Maya clay” (Beach
1998). Detailed soils and paleoecological analysis has
identified a human imprint on the soils, including an
increase in the amount of ash and pollen types that
differ from natural pollen assemblages. Large-scale
irrigation systems were constructed for agriculture.
Even in the humid Yucatán, high infiltration associ-
ated with limestone greatly limits the availability of
surface water. Within the lower reaches of the
Candelaria, large canals were constructed perpendi-
cular to the main river channel and were often con-
nected to uplands or to a matrix of smaller canals
within backswamps. Canals were also built within
perennial wetlands, although there is debate as to
whether seasonal wetlands were utilized for agricul-
ture. Pope and Dahlin (1989) argue that the rapid
fluctuation of these perched aquifers greatly limited
their usage for agriculture. Ayala-Pérez et al. (1998)
mention that the Maya built an elevated road across
the shallow estuary they studied that is still exposed
during the lowest water levels.
After the collapse of the Mayan empire around
1100 years ago the landscape went fallow and forests
returned. There is considerable debate regarding the
degree to which this landscape is “natural.” It is now
widely recognized that the Maya cleared enormous
tracts of forest during their 3000-year occupation of
this landscape and much of the existing forest does not
predate the Maya collapse. Until the late 1960s there
was little development in this region other than tradi-
tional slash and burn agriculture and nondestructive
extractive forestry. Since then, renewed deforestation
has occurred as a result of large-scale logging opera-
tions and rapid population growth. In the uplands,
Río Candelaria (Yucatán)
Ch23.qxd 3/24/05 2:18 PM Page 1053
forest clearing has mainly been for small-scale slash
and burn (swidden) agriculture. Although traditional
in form, the pace of milpa clearing is more rapid than
in prehistoric times, which may result in this being
much more disruptive in comparison to prehistoric
swidden practice. Coupled with this has been forest
clearing for ranching, particularly in the Coastal Plain,
which can be seen as permanent forest removal.
Although it is difficult to quantify, Gunn et al. (1995)
speculate that these actions are likely increasing runoff
and storm flow through the main channel of the Can-
delaria. However, the coastal area is now within the
Lagunas Terminos Protected Area. There is very little
industry or commercial activities in the lower reaches
of the river. The only significant urban area on the
river is the city of Candelaria, located 57 km above the
river mouth.
The Rio Yaqui is a relatively large 6th order river
basin in northwestern Mexico that drains an area
(73,000km2) between 34°N and 32°N latitude in the
Mexican states of Sonora and Chihuahua and the
extreme southeastern corner of Arizona (Fig. 23.6,
see Fig. 23.17). Based on literature and extensive
fieldwork during the late 1970s, Hendrickson et al.
(1981) provide a useful general overview of physical
aspects of the basin and an account of its biota, with
primary focus on its fishes. That account, as much as
possible, will be updated and expanded upon here).
The main stems of two major subbasins, the
southern Papigochic–Sirupa–Aros and the northern
Bavispe, drain first generally north to northwest in
upper reaches of the Sierra Madre Occidental in
western Chihuahua. The Bavispe makes a broad turn
just south of the international border, adding the
San Bernardino system flowing from Arizona to flow
south to southwest. The Papigochic–Sirupa–Aros
system makes several similar nearly 180° turns as
it wanders through valleys and intervening deep
canyons to eventually join the Bavispe and form the
main-stream Rio Yaqui. Although the higher eleva-
tions and intermediate canyon reaches have high
gradients, downstream the low-gradient reaches
23 Rivers of Mexico
FIGURE 23.6 Upper Rio Yaqui (Photo by W. E. Doolittle).
Ch23.qxd 3/24/05 2:18 PM Page 1054
meander through a desertic forest to cross the
Coastal Plain of Sonora to the Los Algodones Estuary
about 24km southwest of Ciudad Obregón.
Human impacts on the Rio Yaqui date to pre-
Columbian times, when the basin was occupied by
the prehistoric Cahita peoples, believed to have given
rise to present indigenous groups of the region, such
as the Rarámuri (Tarahumara), Yaqui, Mayo, and
Pima. Apache bands later roamed higher elevations
of the basin, and Pueblo cultures were also present,
mostly in Chihuahua. European influence began with
the establishment of Spanish missions in lower ele-
vations beginning in the early sixteenth and seven-
teenth centuries, with relatively little incursion into
the Indian-occupied areas upstream but strong influ-
ence on the tribes of the more accessible lower
reaches of the drainage.
Physiography, Climate, and Land Use
Most of the upper Yaqui basin lies in the Sierra
Madre Occidental (SC) physiographic province,
where it flows through rocky, complex, and often
deep scenic canyons. A small part of the upper basin
in Arizona lies in the Basin and Range (BR) province
and a small part of the Buried Ranges (BU) province.
As the river drains from the mountains it flows across
lower elevations of the Buried Ranges province along
the coast of the Gulf of California.
The Papigochic and Bavispe drainages in the
eastern portion of the basin originate on the high
rolling plains to the east of the crest of the Sierra
Madre, and at least parts of these systems were
captured from former drainages into closed basins
to the east by headwater erosion of west-draining
streams (Hendrickson et al. 1981, Minckley et al.
1986). These systems then drop, sometimes quite
precipitously, from conifer-dominated forests and
high-elevation grasslands (especially in the far south-
eastern reaches of the basin) to descend through
the highly diverse Sierra Madre Occidental Pine–
Oak Forests terrestrial ecoregion (Ricketts et al.
terrestrial.html). The northernmost portion of the
drainage, including in southeastern Arizona, drains
from the arid Chihuahuan Desert ecoregion, typified
by creosote bush, yucca, and various cactus species,
as well as grasslands. The lower Yaqui flows prima-
rily through the Sonoran–Sinaloan Transition Sub-
tropical Dry Forests ecoregion, a transition between
the Sonoran Desert to the northwest and the Sinaloan
Dry Forests to the southeast. Thus, it includes a mix
of desert vegetation, such as various cacti, and sea-
sonally deciduous trees of the Sinaloan Forest. The
very lowest portion of the Yaqui basin north of the
westward-trending river (see Fig. 23.17) is part of
the Sonoran Desert ecoregion.
The climate of the Yaqui basin ranges from tem-
perate in the mountains to very dry in the desert
(INEGI 2000a, 2000b). Mean monthly air tempera-
tures and precipitation were estimated from seven
weather stations throughout the basin. Monthly
temperatures ranged from 11°C in January to 26°C
in July (Fig. 23.18), with an annual mean of 18.4°C.
Daily nighttime temperatures, however, commonly
fall well below 0°C in winter at high elevations, and
daytime temperatures in low and middle elevations
commonly exceed 40°C in summer. Throughout the
basin most of the precipitation falls (as rain) in July
and August (over 11cm/mo), comprising about half
the annual amount of 48cm (see Fig. 23.18). Precip-
itation is highest in the higher eastern mountains,
where substantial accumulations of snow are com-
mon, and least in the lower mountains and foothills
of the western side of the basin.
Human population density in the basin averages
only 7 people/km2, with Ciudad Obregón, Sonora,
the largest city at about 345,222 people (INEGI
2000a, 2000b). In spite of this low density, agricul-
ture and logging are pervasive. Currently, agriculture
is fairly diverse, but only occupies 3% of the basin,
with the largest and most productive portion on the
lowermost Coastal Plain. Wheat, soybeans, cotton,
garbanzo, corn, cattle, poultry, pork, and even
shrimp farms are found, and apples and other
orchard crops are important in the upper Papigochic.
Forestlands represent 75% of the basin, and inten-
sive logging of pine forests is obvious over much of
the Sierra Madre in both Sonora and Chihuahua.
Recent years have seen the logging industry starting
to exploit the diverse oak forests of middle eleva-
tions. Some of the few inaccessible and sometimes
small but significant headwaters areas that remain in
a more or less natural state are being considered for
designation as natural protected areas.
River Geomorphology, Hydrology,
and Chemistry
The river can be separated into four subbasins with
somewhat varied physical, hydrological, and flow
characteristics (INEGI 2000a, 2000b). In the south-
east, the Papigochic, Sirupa, and Aros rivers drain
parts of both Sonora and Chihuahua, whereas to the
north the Bavispe system drains large areas of the
Río Yaqui
Ch23.qxd 3/24/05 2:18 PM Page 1055
same two states and a small part of Arizona. The
third-largest subbasin, the lower Rio Yaqui, harbors
the system’s reservoirs and most of its irrigated agri-
culture. The small Moctezuma–Nacozari subbasin
lies on the west margin of the basin.
The name Rio Yaqui is first applied starting at
about 600masl for the reach extending downstream
from the confluence of the Aros and Bavispe rivers.
Here the natural uncontrolled discharge of the Aros
merges with the regulated flows of the Bavispe to
form a river with a mean width of 60m. The Rio
Yaqui here is characterized by a well-developed
channel with alternating riffles, runs, and pools, with
a diversity of depth and velocity regimes, thus pro-
viding a wide variety of habitats for riverine biota.
Substrates range from boulders, cobble, and well-
sorted gravels with little embedding in higher reaches
to high concentrations of fine benthic organic sedi-
ments with woody debris, sand, and finer particles.
The discharge of the Yaqui is now regulated by
three large reservoirs: La Angostura (Lázaro
Cárdenas), Plutarco Elías Calles (El Novillo), and
Alvaro Obregón (El Oviachic). Two of these,
Plutarco Elías Calles and Alvaro Obregón, are on the
main-stream Yaqui, and La Angostura is farther
upstream on the Bavispe (see Fig. 23.17). The habitat
produced by these dams is intermediate between a
free-flowing river and a more typical deep and estab-
lished lake. At least below La Angostura, discharge
occasionally is dropped to zero for days at a time.
Sediments in the reservoirs are primarily the soils of
the ancient canyons and bedrock and boulders that
are overlain or embedded with sand and silt.
Mean annual discharge from 1976 through 1979
was 78.5m3/s, although much of this flow is now
diverted before reaching the Gulf of California.
Monthly discharge and runoff are now relatively
evenly distributed throughout the year by the dams
(see Fig. 23.18). Highest flows occur in July and
August, when precipitation exceeds 10cm/mo.
The lower section of the lower basin includes that
portion of river from just below El Oviachic Reser-
voir to the river’s mouth, a distance of about 100km.
Concrete channels divert most of the flow here and
become the most characteristic habitat, and only
a small portion of the river remains to flow through
the natural channel. The natural habitats here are
now low velocity with an abundance of organic
matter, as the remaining flows and irrigation returns
course through meanders within a narrow zone of the
broad natural channel through the Coastal Plain.
Controlled releases resulting in removal of peak flood
discharges through this reach have allowed
encroachment of riparian vegetation that constrains
the now small river to a narrow channel through
dense vegetation. Tides are present only in the Algo-
dones estuary, the natural mouth of the main course
of the Rio Yaqui, though a significant portion of the
river’s discharge now finds its way to sea via other,
man-made routes.
Except for localized point industrial and logging
contamination and somewhat more diffuse agricul-
tural runoff, water quality is generally good along
most of the upper and middle river. Numerous small
sawmills in the Sierra Madre continue to discharge
sawdust to stream channels, sometimes causing
extensive impacts, and roads associated with logging,
as well as general logging practices, have sometimes
greatly increased erosion. Intensive agriculture in the
upper Papigochic drainage has obvious impacts on
streams of that area. Nonetheless, somewhat lower
on these rivers, reservoirs support fisheries, mostly of
nonnative species, that are exploited both by local
residents and foreign and domestic sport fishermen.
River Biodiversity and Ecology
The Rio Yaqui is located within the Sonoran fresh-
water ecoregion (Abell et al. 2000), which includes
rivers to the west and south of it that also drain
through the Sonora Desert into the Gulf of Cali-
fornia: the Rio Concepcíon, Rio Sonora, and Rio
Algae and Cyanobacteria
Little information exists on the periphyton in the
river and its wetlands, although species richness
likely is quite high given the variety of habitats.
Aquatic macrophytes occur throughout the river
system, but only 11 species have been collected
from the river and its floodplain, with pondweed,
buttercup, and Nasturtium the most common.
Hendrickson et al. (1981) mention specific occur-
rences of, but did not collect, voucher specimens.
Wetlands in the lower basin are characterized by
a variety of emergent and floating-leaved species,
depending primarily on salinity. Typical species in
freshwater areas include cattail and a few small
ciénegas (desert marshlands), especially in the desert
grassland areas of southern Arizona and nearby
northeastern Sonora and northwestern Chihuahua
(Hendrickson and Minckley 1985), where some have
been given government protection (e.g., USFWS
1994). Halophytes occur in the extensive estuarine
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1056
marshes and also in patches in otherwise freshwater
areas where salt springs emerge. Black mangrove,
white mangrove, and red mangrove dominate in
salt marshes; iodinebush, saltbrush, seepweed, and
Florida mayten dominate in inland areas with
increased salinity. Water hyacinth (nonnative) has
been recorded, but unlike what has happened in some
basins further south, this problematic weed has not
yet established extensively (Arriaga-Cabrera et al.
Riparian corridors in headwaters and middle
elevations consist primarily of Arizona sycamore,
Arizona alder, willows (Goodding, Bonpland), green
ash, and Fremont cottonwood. Expansive mesquite
forests occupy most terraces along larger middle- to
lower-elevation rivers through the Sonoran Desert,
with Fremont cottonwood, willows, common reed,
salt cedar (nonnative), and hackberry associations
along the river’s bank (Minckley and Brown 1994).
Information on invertebrate communities in the
Yaqui is sparse, but at least collections of some taxa
have been made (A. Contreras-Ramos, personal com-
munication). Among the aquatic insects collected in
the system are hellgrammites (Corydalus bidenticu-
latus and Corydalus texanus), mayflies (Siphlonurus
occidentalis,Nixe salvini,Acentrella insignificans),
and stoneflies (Capnia decepta,Mesocapnia frisoni,
and Anacroneuria wipikupa) (Bauman and Kon-
dratieff 1996, Contreras-Ramos 1998, A. Contreras-
Ramos, personal communication). The snail Fossaria
(Bakerilymnaea) bulimoides has also been found, and
isolated populations of the large riverine bivalve
Anodonta sp. occur in at least the Rio Sirupa in
canyon reaches in Chihuahua (D. A. Hendrickson,
unpublished data). Estuarine crustaceans include
Cauque River prawn, blue shrimp, and white
The fish assemblage of the Yaqui is diverse, par-
ticularly because its habitats range from 2500masl to
sea level. A few sources suggest there are at least 107
fish species, including native fishes and marine species
sometimes found in freshwater or in the mouth of the
river (Hendrickson et al. 1981, Minckley et al. 1986,
Campoy-Favela et al. 1989, Calderón-Aguilera and
Campoy-Favela 1993, Castro-Aguirre et al. 1999).
There are at least seven species of Cyprinidae (chubs,
shiners, and minnows), four species of Catostomidae
(suckers), and two species of Poeciliidae (topmin-
nows). Endemic species are few, but include Yaqui
chub (DeMarais 1991, DeMarais and Minckley
1993), Leopold sucker (Siebert and Minckley 1986),
and probably Yaqui trout (undescribed, but see Hen-
drickson et al. 2002, Behnke and Tomelleri 2002),
and Yaqui topminnow (Quattro et al. 1996). Most of
the Cyprinidae and many medium-size species
(Catostomidae and others) prey on aquatic and ter-
restrial insects. Bottom-feeding fishes include Yaqui
sucker and Leopold sucker, as well as the primarily
algal-grazing Rio Grande sucker and Mexican
stoneroller. Only the native roundtail chub and Yaqui
catfish, in adult form, feed almost exclusively on
fishes. Topminnows are the most abundant surface
feeders from the basin in middle to low elevations,
preying on aquatic algae and small insects. An inter-
esting all-female form of topminnow (headwater live-
bearer) (Vrijenhoek 1993, Quattro et al. 1992) occurs
in middle to lower elevations of the basin, represent-
ing the northernmost of a complex of many inde-
pendently evolved asexual (clonal) lineages that exist
as sexual parasites of sexual forms that gave rise to
them via hybridization.
Euryhaline species are an important component
of the lower basin (56 species) and comprise about
of half the Yaqui fish fauna. The most common
and abundant are striped mullet, machete, striped
mojarra, and Heller’s anchovy. Among the natives
with conservation interest is Pacific gizzard shad, for
which the lack of information impedes an appropri-
ate conservation plan (Varela-Romero 1989).
The Rio Yaqui has at least 17 nonnative fishes
that have been introduced over the past century
(Hendrickson et al. 1981, Hendrickson 1984,
Campoy-Favela et al. 1989, Varela-Romero 1989).
Among the most abundant are channel catfish, river
carpsucker, largemouth bass, common carp, Mozam-
bique tilapia, rainbow trout, and green sunfish.
Western mosquitofish is an important nonnative that
threatens the ecologically similar native topminnow.
Nonnative piscivores feed primarily on smaller
forage fishes, such as native cyprinids, catostomids,
and poeciliids. Concern exists about the highly pis-
civorous flathead catfish, which has been reported in
the Yaqui (Leibfried 1991), but no vouchers or sub-
sequent specimens have been recorded. Blue catfish
have long been established in the Bavispe and surely
impact native faunas there. Success of these nonna-
tives appears directly related to the existence of reser-
voirs, which create less diverse habitats and function
as centers for their dispersal. Largemouth bass and
channel catfish now occur far above reservoirs in
both the Bavispe and Papigochic–Aros–Sirupa sub-
Río Yaqui
Ch23.qxd 3/24/05 2:18 PM Page 1057
basins. Genetic interactions with introduced species
also endanger native stocks. Pure stocks of Yaqui
catfish are essentially nonexistent because of hybri-
dization with channel catfish.
A number of Rio Yaqui fishes are considered
endangered or threatened by the Mexican federal
government (SEMARNAT 2002, Varela-Romero
1995). Yaqui chub is the only species considered
endangered, whereas longfin dace, ornate minnow,
Yaqui shiner, Cahita sucker, Rio Grande sucker, and
Yaqui topminnow are considered threatened and
Yaqui sucker, Leopold sucker, and Yaqui catfish are
categorized as under special protection. Yaqui top-
minnow is formally a subspecies of Gila topminnow,
but some populations in the Yaqui appear to be suf-
ficiently different to be elevated to the species level
(Quattro et al. 1996). Recovery efforts for endan-
gered Yaqui fishes are coordinated with efforts in the
U.S. portion of the basin, centered at the Dexter
National Fish Hatchery, New Mexico (Johnson and
Jensen 1991). The final destiny of these native fishes
is the San Bernardino National Wildlife Refuge
(SBNWR), Arizona, and potentially one day former
native habitats in Mexico (USFWS 1994).
Amphibians and reptiles directly associated with
the Yaqui also are diverse. Fourteen species of frogs
and toads, one salamander, five snakes, and three
turtles are found either in the river main stream or
the floodplain much of the year (SIUE 1992, Flores-
Villela 1993, Stebbins 1985). Among the more rep-
resentative amphibians are Tarahumara salamander,
Tarahumara frog, and Chiricahua leopard frog, all
provided legal protection by the Mexican govern-
ment. The introduced American bullfrog is found in
natural habitats of the northernmost Bavispe and in
the lower basin, and culture of the species for com-
mercial purposes has begun. Unfortunately, it has
been amply demonstrated to have severe impacts
on native amphibians and snakes, particularly those
of the genus Thamnophis, which are diverse (six
species) and common in the Yaqui. Sonoran mud
turtle, common slider, and Western box turtle are all
common. A major mammal associated with the Rio
Yaqui is the neotropical river otter, which is occa-
sionally observed in the channelized portion of the
river between the major dams (Gallo-Reynoso 1997)
and also far above the reservoirs.
Ecosystem Processes
No broad-scale studies are available on ecosystem
processes, though much effort has focused on sus-
tainability of agriculture systems in the lowermost
river (
Human Impacts and Special Features
The Rio Yaqui is one of most important rivers of
northwestern Mexico not only because of its inter-
esting biodiversity (Arriaga-Cabrera et al. 2000) but
also because of the economic importance of the agri-
culture its water supports. It is, however, heavily
impacted by humans. Natural discharge and inunda-
tion patterns occur only in the headwaters, upstream
from its three large main-stem dams. Nonetheless,
despite more than a century of exploitation and
extensive, highly modified landscapes (Adams 2002),
many of the river’s tributaries and their riparian
zones have remained relatively intact throughout
substantial portions of the basin.
Impacts are greatest in the lower basin, which
continues to absorb a rapid increase in population
and has been highly developed for intensive irrigated
agriculture, as have the coastal plains of rivers further
south in Sinaloa, such as the Rios Mayo, Fuerte, and
Sinaloa. As with those areas, human communities on
the lower Rio Yaqui floodplain are now experiencing
problems with high pesticide levels (e.g., Arreola-
Lizárraga 1995, Guillette et al. 1998) and water-
supply limitations following recent record droughts.
Water has recently had to be pumped over reservoir
spillways to distribution canals that supply appar-
ently unsustainable levels of agriculture, and local
economies are crashing (Dean 2004).
Aquaculture is now rapidly developing in the
region as a possible alternative. Despite questions
regarding its sustainability, the intertidal areas of
the upper estuary are being converted to extensive
shrimp farms in response to government and pri-
vate incentives. These activities stand to increase
problems related to nonnative species brought in
with aquaculture development. Still, despite its
complex problems related to extensive agricultural
and forestry development, the Rio Yaqui basin
retains long reaches of free-flowing natural desert
mountain rivers, and its biota is at least more intact
than those of similar desert rivers further north, such
as the Gila of Arizona and New Mexico. This
basin should, therefore, remain a high priority for
management and conservation (Arriaga-Cabrera
et al. 2000).
The Conchos basin begins high in the Sierra Madre
Occidental of northwestern Mexico along the North
American continental divide. Most of the basin lie
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1058
within the state of Chihuahua, but the system also
drains a portion of northern Durango (Fig. 23.7,
23.19). From the pine-forested semiarid flanks of the
rugged Sierra Tarahumara, a range within the Sierra
Madre Occidental, the primary tributaries of the
Conchos basin (Rios Chuviscar, San Pedro, Toronto,
Parral, and Florido) flow northeasterly before con-
verging in the arid Chihuahuan Desert. Below the
confluence with the Rio Chuviscar the Rio Conchos
flows northerly to the Rio Bravo del Norte (Rio
Grande), joining that river at Ojinaga, Chihuahua,
across from Presidio, Texas. The drainage area of
the Rio Conchos is estimated at 68,386km2,
which accounts for 14% of the Rio Bravo’s drainage;
however, because the Rio Bravo now has very little
flow upstream of this confluence, the Conchos con-
tributes a large proportion to total discharge below
the confluence. It is thus mostly the Conchos that
Río Conchos
FIGURE 23.7 Upper incised Rio Conchos (Photo by W. E. Doolittle).
Ch23.qxd 3/24/05 2:18 PM Page 1059
provides the essential discharge through Big Bend
National Park, and downstream of the park the
water is heavily used for floodplain irrigation and
municipal purposes by the growing populations
along the Texas–Mexico border. Indeed, there is con-
siderable debate surrounding the issue of water
resources along the Rio Grande, and the contribution
of the Conchos basin has become a controversial
topic between Mexico and the United States. A recent
comprehensive overview of the Conchos basin (Kelly
2001) summarizes the water plan developed by
Mexico’s National Water Commission (Comisión
Nacional del Agua, or CNA) in 1997.
Northern Mexico contained significant numbers
of indigenous peoples. However, unlike Mesoameri-
can Mexico south of the Tropic of Cancer, with
perhaps some exceptions (e.g., Hard and Roney
1998, Schaafsma and Riley 1999, Whalen and
Minnis 2001) and at a somewhat smaller scale, it did
not contain large culture regions (Coe and Koontz
2002). In addition, indigenous peoples within the
area did not have irrigation projects on the scale of
those of more southern populations. The Sierra
Tarahumara are home to the Rarámuri, one of the
largest indigenous groups in Mexico. After the con-
quest of Mexico the Rarámuri people migrated to
this portion of Mexico to escape enslavement in
Spanish silver mines. The Rarámuri, also known as
the Tarahumara, did modify hydrological and sedi-
mentological processes in a couple of ways. On hill
slopes they constructed terraces to reduce runoff and
soil erosion and to allow for more intensive agricul-
ture. Along smaller streams, trincheras (wooden and
stone structures oriented across stream channels)
were constructed to reduce runoff and promote
accumulation of fertile sediments for agricultural
purposes (Doolittle 1985, 2000). The Tarahumara
peoples have been isolated for much of history, and
it is not until recently that they started to be signifi-
cantly impacted by modern society. The principal dis-
turbance to the Tarahumara people today involves
logging of the extensive pine forests of the Sierra
Madre Occidental.
Physiography, Climate, and Land Use
The Conchos basin drains the Sierra Madre Occi-
dental (SC) and Basin and Range (BR) physiographic
provinces (see Fig. 23.19). Its headwaters are prima-
rily within the Sierra Madre Occidental, Mexico’s
most extensive mountain system. The mountains are
aligned north–south and extend from the U.S. border
to central Mexico. Comprised primarily of Tertiary
volcanic rocks, andesite, and rhyolite, some moun-
tains within the Conchos basin contain several ridges
at least 3500masl, such as the headwaters of the Rio
Florido, a tributary draining the Sierra Tarahumara
in northern Durango. Runoff from some of these
peaks also enters Pacific drainages, such as the Rios
Mayo, Fuerte, Sinaloa, and Culiacán, with which the
Conchos headwaters closely interdigitate. The vol-
canic rocks overlay older Cretaceous sedimentary
rocks of marine origin that contain extensive ore
deposits from igneous intrusions. These are fre-
quently exposed where rivers have incised deep
canyons, and have long been exploited for mining.
The lower portions of the basin, greater than half
the total, drain the southern extension of the Basin
and Range province (Hunt 1974). The major tribu-
taries exit the mountains and form the Rio Conchos
within the western fringe of the Basin and Range.
Chihuahua’s Basin and Range topography is charac-
terized by small northwest-trending mountains,
although the ridge and valley topography is not as
symmetrical as that found in the United States. This
series of fault block ridges primarily consists of
uplifted Tertiary igneous or Mesozoic sedimentary
rocks, the elevations of which are generally around
1750masl, although several ridges within the lower
portions of the basin exceed 2000masl. (INEGI
1998e). The low-lying portions include large elon-
gated valleys infilled with Quaternary alluvial
deposits. Along the flanks of the ridges the valleys
have extensive bajadas (coalescing alluvial fans). The
Basin and Range also includes (outside of, but adjoin-
ing, the Conchos basin) large internally draining
playas infilled with Quaternary lacustrine deposits.
These internally draining basins are fed by ephemeral
streams that have undergone repeated shifts in
channel patterns, complicating delineation of these
drainage systems.
The terrestrial ecoregions of the basin correspond
closely with the physiographic provinces (INEGI
1981d, Ricketts et al. 1999). The Sierra Madre
Occidental Pine–Oak Forests ecoregion within the
Sierra Madre Occidental province contains a high
diversity of both pine and oak species. The Chi-
huahuan Desert ecoregion of the Conchos basin is
largely within the Basin and Range province. This dry
desert ecoregion has only xeric plants, such as cre-
osote bush, tarbush, viscid acacia, yucca, and cacti.
The dominant precipitation mechanism within
the Conchos basin is the North American monsoon,
or Mexican monsoon (Douglas et al. 1993), charac-
terized by an abrupt increase in precipitation over the
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1060
basin’s headwaters with more than 50% falling from
mid-June to mid-September. This occurs as the sub-
tropical high pressure cell migrates north, diverting
warm moist air from the Pacific over southwestern
North America. Rather than producing large frontal
storms, the moist air interacts with the region’s
topography to create orographic rainfall with con-
vective thunderstorms. Individual rainfall events thus
tend to be of short duration but intense. Less im-
portant precipitation mechanisms include easterly
migrating tropical cyclones from the Pacific and less
frequently occurring westerly migrating tropical
cyclones from the Gulf of Mexico. However, this
mechanism is also likely influenced by synaptic scale
circulation, particularly the position of the jet stream
and the ridge of high pressure. The upper elevations
of the headwaters (generally 2000masl) also experi-
ence light snowfall during winter months.
Although the basin has a generally warm and dry
climate, there is substantial spatial variation in tem-
perature and precipitation. Annual precipitation at
some points in the Sierra Madre Occidental can
exceed 100cm but is as low as 20cm in some parts
of the Chihuahuan Desert. Temperature varies
greatly with altitude and season, with summer day-
time temperatures in the desert often exceeding 40°C
and winter temperatures in the mountains falling
below .00210°C. Mean monthly precipitation and
temperature for the basin as a whole was estimated
using data from weather stations at Chihuahua
(at the edge of mountains and desert) and Guanacevi
(in the mountains of the Rio Florido basin at
2200masl). Precipitation is highly seasonal, being
less than 1cm/mo from February through April and
increasing to more than 10cm/mo in July and August
(Fig. 23.20). Mean monthly temperatures for the
basin are close to 10°C in midwinter but rise to 23°C
in June. Mean annual precipitation was estimated as
48cm (also see and mean
annual temperature was 18°C.
The economy, and thus land use, in the basin are
primarily agriculture, mining, and forestry. The large
amount of agriculture is made possible by large
irrigation districts in the lower basin in the areas
of Hidalgo de Parral, Camargo-Jiménez, Delicias,
and the lower Conchos. Irrigation has made a variety
of crops possible, including maize, winter wheat,
alfalfa, cotton, and pecans (Kelly 2001). The human
population in the basin was estimated at 1.32 million
in 2000 (about 17 individials/km2) and is projected
to increase to 1.77 million in 2020 (Kelly 2001).
The bulk of this population is concentrated in the
largest cities, such as Chihuahua (677,852), Hidalgo
de Parral (103,185), Delicias (99,137), Camargo
(39,189), and Jiménez (32,966).
River Geomorphology, Hydrology,
and Chemistry
The geomorphology of the Conchos system reflects
significant spatial variability, which would be
expected of an arid basin having close to 3000m of
relief. The upper headwaters in the Sierra Tara-
muhara are deeply incised within the volcanic strata
(see Fig. 23.7). Here the valleys are narrow, and the
rivers are primarily bedrock controlled. However, in
the larger valleys in the mountains the floodplains
widen and the channel is primarily alluvial. In these
segments the water may be diverted for agricultural
irrigation, such as in the Rio Balleza and the upper
Rio Conchos (INEGI 1998d). The river valley near
Delicias, where the Rio San Pedro exits the moun-
tains, is heavily utilized for grazing and extensive
agriculture. Here the valley is 2.5km wide with an
extensive system of irrigation canals, such as Canal
Principal Numero Cinco, which flows from the
Francisco I. Madero reservoir (INEGI 1976a,
1976b). Overgrazing and associated runoff is becom-
ing a significant issue with respect to water quality in
these areas, and recharge of the alluvial aquifer is
being affected, with consequences for sustaining bio-
logically critical low flows. The downstream reaches
of the Conchos are significantly influenced by struc-
tural controls as the river flows through the Basin and
Range province. Here the river channel alternates
between meandering and braided patterns. In some
reaches the floodplain averages 2km wide, whereas
in other reaches it in incised into the uplifted ridges
of the Basin and Range province and lacks an allu-
vial channel (INEGI 1976a, 1976b). Just upstream of
the river mouth, at Ojinago, the valley widens to
3km and the meandering pattern increases in sinu-
osity. A low-flow structure creates a small reservoir,
which is primarily used for diverting water into an
irrigation system (INEGI 1998a).
A 49-year discharge record (1955 to 2003) is
available for the Rio Conchos from the International
Boundary and Water Commission (www.ibwc.state.
gov). Mean annual discharge from 1955 to 1994 was
20.5m3/s, varying from 7.5 to 37.1m3/s (Fig. 23.8).
From 1995 to 2003, however, discharge dropped to
an average of only 3.8m3/s, or only 19% of the long-
term mean. Although it is commonly cited that the
Rio Conchos provides much of the flow of the Rio
Bravo del Norte (Rio Grande) above the U.S. Big
Río Conchos
Ch23.qxd 3/24/05 2:18 PM Page 1061
Bend National Park, flows since 1995 have dramat-
ically reduced this contribution. Part of this flow
reduction has apparently been due to a drought in
the region that has reduced runoff; however, the
decrease also appears due to increased diversions for
agricultural irrigation from dams along the Conchos.
Water in the Rio Conchos is highly regulated by
a series of dams. Of the seven major reservoirs in the
system, La Boquilla (also called Lago Toronto),
located on the main stem of the Conchos near
Camargo, is the largest (see Fig. 23.19). Other major
reservoirs are the San Gabriel on the Rio Florido, the
F. Madero on the San Pedro, and the Luis L. León on
the lower Conchos. All of the major reservoirs are
used for irrigation supply, and La Boquilla is used for
hydropower. The León Reservoir is of particular sig-
nificance because it is the last to discharge water (or
not) before the Conchos enters the Rio Bravo. Not
only have the mean discharge levels at the mouth
of the Conchos been greatly reduced since 1995, but
for extended periods of time water releases have
been reduced to zero by dams at several points along
the lower river (see 2003 reports of the Center
for Space Research at the University of Texas at
Austin, For example, the León
Reservoir has withheld water for several months at a
time, and the only flow into the Rio Bravos has been
from precipitation downstream of the dam.
Because the major dams on the Conchos were
built before even the earliest available discharge
records (1955), it is difficult to know to what extent
they have influenced the natural pattern of flow, but
comparison of mean monthly precipitation to mean
monthly runoff provides some insight. The long-term
average runoff, even from 1955 to 1994, was low,
as would be expected in an arid landscape (see Fig.
23.20). Note that runoff in Fig. 23.20 is multiplied
by 10 to facilitate visualization of the monthly
pattern. Total annual runoff from 1955 to 1994 was
only 0.8cm, or about 2% of total precipitation. Of
particular interest is that the mean runoff pattern
during this time is considerably flatter than might be
expected in a region with highly seasonal precipita-
tion (see Fig. 23.20). This suggests strong regulation
over this 40-year period. From 1995 to 2003,
however, there is not only strong regulation, but
annual runoff has greatly decreased (0.16cm or
0.3% of precipitation). This reduction is particularly
obvious during the dry winter months; mean dis-
charge and runoff from 1995 to 2003 decreased to
less than 10% of prior runoff.
Although water-chemistry and water-quality
information have not been readily available for the
Conchos, a one-time study by Gutiérrez and Borrego
(1999) provides some valuable information. Various
water-quality variables were measured at 105 sam-
pling locations, primarily in the lower river in the
vicinity of Camargo, Delicias, and upstream of
Ojinaga. The natural chemistry of the lower river is
affected by its limestone geology, with a pH exceed-
ing 8 and a Ca concentration greater than 100mg/L.
However, the river is so strongly polluted that it is
23 Rivers of Mexico
FIGURE 23.8 Long-term mean annual discharge (m3/s) on the Rio Conchos at Ojinaga, Chihuahua, Mexico.
Ch23.qxd 3/24/05 2:18 PM Page 1062
difficult to know the baseline chemistry for most
other variables. Kelly (2001) points out that there are
only five water-quality monitoring stations within
the basin but indicates that the most serious problem
is fecal coliform contamination, primarily from
Jiménez and Camargo. She also mentions that the
Rio Florido is contaminated with high levels of oil
and grease from a chemical plant in Camargo.
Because much of the river’s water is used for irriga-
tion, particularly in the Delicias area, agricultural
return flows are highly contaminated with high nutri-
ents and increased salinity. Gutiérrez and Borrego
(1999) provided more specific data and indicated a
great increase in total dissolved solids from 60mg/L
in the headwaters to >700mg/L in the lower river.
They showed that waters returning to the river from
irrigation returns can be almost 2000mg/L. Nutrients
in the lower river were very high, with mean values
of NO3-N ranging from 2.3 to 4.6mg/L in three dif-
ferent river segments. Phosphates ranged from 0.15
to 0.53mg/L. Upstream of the irrigation districts Na
usually was measured at less than 20mg/L, whereas
in the irrigated area near Delicias Na approached
300mg/L at several sampling sites.
River Biodiversity and Ecology
The Rio Conchos has its own freshwater ecoregion
of the same name as part of the Rio Grande complex
of ecoregions (Abell et al. 2000). The river is of par-
ticular interest biologically because of substantial
endemism in its fishes and herpetofauna.
Information on invertebrates of the Rio Conchos
is sparse. One would expect the fauna of the lower
Conchos to be similar to that of the Rio Bravo del
Norte (Rio Grande), but information for the Rio
Grande is also limited (see Chapter 5).
It is surprising that a basin as important as the
Conchos is so poorly collected for fishes. As pointed
out by Hendrickson et al. (2002), “The basin remains
almost completely uncollected above 1700m eleva-
tion,” yet a huge area of the basin drains surfaces
above that elevation to more than 3000masl, and
lower reaches of the basin clearly still require much
more extensive faunal inventories than have been
realized to date. In spite of the paucity of collections,
several recently published accounts of the river’s fish
fauna are available, although none provide a com-
prehensive, detailed overview, and there is consider-
able disagreement about even basic information.
Miller (1978) provides the first account known to
us that attempted to comprehensively inventory the
basin’s fauna, listing 31 species. That work was
updated in Smith and Miller (1986), which lists 35
species, with 7 endemic and 1 nonnative. Abell et al.
(2000), in their continental-scale biodiversity analy-
ses, stated there were 47 native fish species of which
12 were endemic. However, because they included
neither a complete species list nor provided specific
citations from which their data were compiled, it is
impossible to determine the validity of their numbers.
Arriaga-Cabrera et al. (2000) listed over 40 species
for the upper Rio Conchos and Rio Florido, but
sources for the lists are not cited and the lists contain
many errors. Edwards et al. (2002a, 2002b) reported
a total of 44 species from their own collections in the
1990s, but their relatively extensive collections were
limited to lower and middle elevations and are not
comprehensive for the entire drainage.
For purposes of this chapter, therefore, a list of
species was compiled by combining data from Smith
and Miller (1986), Edwards et al. (2002a, 2002b),
Lozano-Vilano (2002), and records from an assort-
ment of available ichthyology collection databases
(including UMMZ, TNHC, CAS, FLMNH, INHS,
ASU, and FMNH, museum codes as in Leviton et al.
1985, Leviton and Gibbs 1988). Based on that com-
pilation and analysis, 53 fishes are recognized with
apparently valid occurrence records in the basin. Of
those, 38 are native (or very probably so; some doubt
remains regarding a few because early collection
records are inadequate to exclude the possibility
that they were present before introductions started),
7 are considered endemic, and 8 are clearly
Documented Conchos basin endemics are
Conchos shiner, bigscale pupfish, bighead pupfish,
Salvador’s pupfish, Conchos darter, yellowfin gam-
busia, and crescent gambusia. Species clearly intro-
duced include goldfish, common carp, warmouth,
inland silverside, white bass, rainbow trout, and
perhaps plains killifish, though the latter’s occurrence
in rivers further south and ability to disperse through
highly saline environments may cast some doubt on
this conclusion.
In addition to the endemics, the remaining native
fauna not surprisingly shares many species with the
Rio Bravo, a river that has also seen major changes
in its fish fauna due to human impacts (Contreras-
Balderas et al. 2003). A few marine derivatives at
least formerly ascended the Rio Bravo to occupy the
Río Conchos
Ch23.qxd 3/24/05 2:18 PM Page 1063
Conchos, including American eel and freshwater
drum. Tropical freshwater components include
Mexican tetra and Rio Grande (or Texas) cichlid.
A diversity of small minnows abound, including
Mexican stoneroller, ornate minnow, red shiner,
roundnose minnow, and several other shiners, includ-
ing Texas shiner, Tamaulipas shiner, Chihuahua
shiner, and Rio Grande shiner, and longnose dace, as
well as at least one species of the western genus Gila
(Gila pulchra or a close relative of it). Catfishes are
relatively diverse, including, in the genus Ictalurus,
headwater, blue, and channel catfishes and an unde-
scribed form. No trout are included as native, though
Hendrickson et al. (2002) point out several lines of
evidence that indicate that a still undocumented trout
may live in remote high elevations of this basin.
Conchos pupfish and blotched gambusia are fairly
wide-ranging cyprinodontiforms. Ranges of both
extend into Texas, but the latter has not been col-
lected in that state for about a decade at least (Hubbs
et al. 1991). Suckers include Yaqui sucker and Rio
Grande sucker, as well as Mexican redhorse. Only
one gar, longnose gar, is known. Although some
debate may exist regarding whether they are native
or not, we consider green sunfish, bluegill, longear
sunfish, and largemouth bass as native until such a
hypothesis can be rejected with more conclusive data;
all are native to nearby streams in both Texas and/or
to the east and south in Mexico. Early collections
were inadequate to rule out their presence long
before introductions started.
Miller (1961, 1978), Contreras-Balderas (1978),
and Contreras-Balderas et al. (2003) have all dis-
cussed the dramatic aquatic faunal changes taking
place in northern Mexico, including the Conchos
basin. By the time of their publications in 1978,
Miller and Contreras-Balderas were separately able
to provide specific examples of dramatic changes
during the first three-quarters of the twentieth
century in three rivers of the Conchos basin at major
population centers (Chihuahua, Camargo, and
Jimenez). By that time the faunas at these sites had
been reduced to anywhere from 5% to 66% of their
former diversity.
Abell et al. (2000) estimated that there are 46
native species of aquatic herpetofauna in the
Conchos, of which 12 are endemic, but provided no
information on the actual species.
Ecosystem Processes
No ecosystem studies have been done on the Rio
Conchos. It is difficult to speculate what preimpact
processes might have been like in this system, other
than being similar to that found in other desert
systems studied in Arizona, New Mexico, and Texas.
At present, however, any ecosystem processes have
been radically altered with the heavy pollution loads,
dams, and dewatering of entire river sections for
months at a time.
Human Impacts and Special Features
The Conchos basin overall is heavily impacted by
humans, but as is the case in many places, fish faunas
and other components of aquatic communities
remain most intact in headwater areas, above the
majority of such impacts. These less-impacted areas
are in mountainous portions of the Conchos basin,
which have extensive tracts of forest. Chiuahuahua
has more forests than any other state in Mexico, and
these are primarily the pine–oak forests of the Sierra
Tarahumara. These forests are essential to both the
terrestrial and aquatic ecology of the region and serve
as a key migratory route for North American birds.
Some tributaries of the Balleza drainage, in particu-
lar, despite extensive logging activities that charac-
terize all parts of the basin at higher elevations,
appear to still harbor relatively intact high- and
middle-elevation aquatic communities in some areas.
Lower reaches, especially below major agricultural
and/or urban areas and below reservoirs, as men-
tioned earlier, retain relatively little of their native
diversity as a result of contamination, dewatering,
and otherwise altered natural regimes. The basin,
however, still retains much of its overall scenic
beauty. Certainly its lower canyon, just upstream of
its mouth at Ojinaga, is spectacular. The Tarahumara
highlands are also a tourist attraction often visited by
those en route to, or coming from, the Rio Fuerte’s
Barranca de Cobre.
There are several major ways in which the
Conchos basin has been negatively impacted by
humans, including logging, dams, surface and aquifer
withdrawals, irrigated agriculture, overgrazing, and
pollution from various sources (Kelly 2001). The
major impact in the headwaters is logging, which is
primarily located within the mountainous head-
waters of the basin. Recent decades have witnessed a
dramatic increase in logging within this region, par-
ticularly by large international logging companies
(Guerrero et al. 2001), and these semiarid moun-
tainous regions are particularly sensitive to defor-
estation, which frequently results in soil erosion
that has downstream consequences. Moreover, the
increased runoff means that less water is available to
infiltrate into the soil, and consequently ground-
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1064
water recharge that supplies streams with base flow
between rainfall events is reduced. A reduction in
base flow has consequences for aquatic species, and
in 14 locations in Chihuahua between 1901 and
1975 it is estimated that over 41% of the fish species
As indicated earlier, the combination of dams,
water withdrawals, and pollution from municipali-
ties, agriculture, and industry has had a profound
effect on the Rio Conchos, particularly in the lower
main stem. The dominant water use (93%) is for irri-
gation, with domestic withdrawals accounting for
6.4% (Kelly 2001). Contaminants are often very
high and flows have sometimes been reduced to zero
for extended periods, all of which is devastating to
aquatic life. Here, as in many arid streams that have
been dammed, there have not only been significant
reductions in total flow and flow variability but a
narrowing of the stream channel as vegetation, par-
ticularly nonnative salt cedar, has encroached upon
the river banks. Much of the natural grasslands have
been transformed by either irrigated crops or over-
grazing and the introduction of nonnative species of
grass (Guerrero et al. 2001).
Water management of the Rio Conchos basin
has been a major point of contention between
Mexico and the United States. According to the 1944
U.S.–Mexico Water Treaty, Mexico is supposed to
provide a certain minimum amount of water to the
United States (Kelly 2001, Center for Space Research
2003). Until the early 1990s Mexico met its treaty
obligations; however, since 1992 it has failed to do
so, arguing that obligations cannot be met due to per-
sistent and extraordinary drought (Center for Space
Research 2003). Simultaneously, the amount of water
used for irrigation has increased.
The Rio Fuerte flows in a generally southwesterly
direction within the southern portion of the state of
Chihuahua to cross the coastal state of Sinaloa (Fig.
23.21). Its major tributaries (Urique, San Miguel,
Verde) arise in the high elevations of the Sierra Madre
Occidental of Chihuahua and Durango. Of particu-
lar interest are portions of middle reaches of the basin
in scenic Barranca del Cobre (Copper Canyon) Park,
home to many Tarahumara (see section on the Rio
Conchos). Coming out of these canyons, the main-
stem river flows primarily across the Piedmont hills
and deltas of the Buried Ranges physiographic
province to empty into the Gulf of California, but its
flow is highly regulated by dams (e.g., Miguel
Hidalgo) at the lower elevations. This is a relatively
arid landscape, particularly in the lower elevations
near the coast, which have been converted to irri-
gated agricultural areas. The reservoirs have been
stocked with many species of nonnative game
The Rio Tamesí flows in a southeasterly direction
within the northeastern state of Tamaulipas, draining
an area of 19,127km2(see Fig. 23.23). It joins the
larger Rio Pánuco near Tampico, 10 km upstream of
the Gulf of Mexico. The basin’s human population is
rather diffuse and generally low, although Ciudad
Victoria (population about 260,000) is located just
outside the northeastern boundary. The Tamesí head-
waters arise high in the semiarid ranges of the north-
ern Sierra Madre Oriental, west of Ciudad Victoria,
and in the Sierra Tamaulipas, a small mountain range
located between the Gulf of Mexico and the Sierra
Madre Oriental. Lithology of the headwater areas
consists predominantly of Cretaceous limestone and
is considered to be heavily karstic. Much of the
streamflow for the Tamesí originates as springs at the
base of the mountains and enters the middle Tamesí
within the Coastal Plain. Forty kilometers from its
confluence with the Rio Pánuco the Rio Tamesí
crosses the Tamaulipas Arch, a south-plunging anti-
cline and an extension of the Sierra Tamaupipas. The
valley has an appreciably low gradient downstream
of the arch, 0.49m/km, with an extensive system of
floodplain lakes. These lakes are utilized for drinking
water by Tampico but, like the lagoons and lakes in
the lower Pánuco, represent an important ecological
component of the Coastal Plain (Fig. 23.9).
The Rio Salado flows in a generally easterly direc-
tion through the northeastern states of Coahuila and
Nuevo León before emptying into Falcon Reservoir
on the Rio Bravo del Norte in Tamaulipas (Fig.
23.25). The basin drains an arid region, spanning
from the headwaters of Rio Sabinas, a major tribu-
tary that begins high in the northern Sierra Madre
Oriental, and flows into the Coastal Plain. A large
dam (V. Carranza) just below the confluence of these
rivers results in a high degree of regulation. Water
from the reservoir irrigates extensive agricultural
areas, particularly for cotton. Much of the main stem
does not flow for extended periods of time. However,
because of the continued international concern with
irrigation and the declining condition of the Rio
Bravo/Grande, the Rio Salado is an important tribu-
tary. Average daily discharge near its mouth in Falcon
Reservoir (near Las Tortilleras, Tamaulipas) from
September 1953 to June 2004 was 10.1m3/s
Additional Rivers
Ch23.qxd 3/24/05 2:18 PM Page 1065
(International Boundary and Water Commission, un-
published data from the International Boundary
and Water Commission,
histflo1.htm). The Salado watershed is increasingly
threatened because of a growing population and
associated land-use change in northern Mexico,
which places greater stress on water supplies such
as springs and streams. For example, Contreras-
Balderas and Lozano (1994) note that in arid South-
western Nuevo León reduced spring flow has
eliminated several endemic species of crayfish and
The Rio Armería flows in a southerly direction in
west-central Mexico, beginning in the state of Jalisco
and passing through the small state of Colima before
entering the Pacific Ocean (Fig. 23.27). The Pacific
coast of Mexico is along an active plate margin and
therefore has a narrow Coastal Plain. The Rio
Armería drains both the Neovolcanic Plateau and
Sierra Madre del Sur, changing its name from Rio
Ayuquila to Rio Armería as it approaches the coast.
A 71km section of river comprises the northeastern
boundary of Sierra de Manantlán Biosphere Reserve
and this basin also receives runoff from the 4240m
asl volcano in the Volcán Nevado de Colima
National Park. Although 60% of the basin is in
forests, its population density is relatively high at
56 people/km2. Much of the river water is diverted
for irrigation in a series of agricultural valleys, with
30% of the basin devoted to agriculture (largely
The Rio Lacanjá is a small tributary of the Rio
Usumacinta located in southern Mexico in the
Chiapas–Guatamala Highlands physiographic
province (Fig. 23.29). This basin has very high
annual precipitation (>200cm) and lies within the
Petén–Veracruz Moist Forests terrestrial ecoregion.
The river is the boundary between two of the most
23 Rivers of Mexico
FIGURE 23.9 Wetlands and lagoons of the lower Rio Tamesí (Photo by P. H. Hudson).
Ch23.qxd 3/24/05 2:18 PM Page 1066
important biosphere reserves in Mexico, Montes
Azules and Lacantún, and 80% of the basin is within
reserves. The river remains in a relatively natural
condition, with fast-flowing runs, a waterfall with
numerous runs and rapids, floodplain clearwater
lakes, a floodplain backwater, and a riparian wetland
(Fig. 23.10).
We acknowledge and thank Norman Mercado-Silva (Uni-
versity of Wisconsin and Universidad de Guadalajara,
Mexico), John Lyons (State of Wisconsin Department of
Natural Resources), and Luis Manuel Martínez-Rivera and
Luis Ignacio Íñiguez-Dávalos (Universidad de Guadalajara,
Mexico) for contributing much of the information for the
one-page summary of the Armería–Ayuquila River. We also
greatly appreciate the information on aquatic invertebrates
provided by Atilano Contreras-Ramos for several of the
rivers. Joachin Bueno-Soria provided an unpublished man-
uscript on the invertebrates of the Grijalva system. José
Campoy-Favela helped with the Rio Yaqui account. We are
most grateful for the contribution of photos by A. H.
Siemens and W. E. Doolittle.
Abell, R. A., D. M. Olson, E. Dinerstein, P. T. Hurley, J. T.
Diggs, W. Eichbaum, S. Walters, W. Wettengel, T.
Literature Cited
FIGURE 23.10 Rio Lacanjá, a tributary of the Rio Usumacinta (Photo by R. Rodiles-Hernández).
Ch23.qxd 3/24/05 2:18 PM Page 1067
Allnutt, C. J. Loucks, and P. Hedao. 2000. Freshwater
ecoregions of North America: A conservation assess-
ment. Island Press, Washington, D.C.
Adams. 2002.
Aguilar-Robledo, M. 1999. Land use, land tenure, and
environmental change in the jurisdiction of Santiago de
Los Valles de Oxitipa, Eastern New Spain, sixteenth
to eighteenth century. Ph.D. diss., Department of
Geography, University of Texas at Austin.
Alcorn, J. B. 1981. Huastec non-crop resource manage-
ment: Implications for prehistoric rainforest manage-
ment. Human Ecology 9:395–417.
Arbingast, S. A., C. P. Blair, J. R. Buchanan, C. C. Gill,
R. K. Holz, C. A. Marin, R. H. Ryan, M. E. Bonine,
and J. P. Weiler. 1975. Atlas of Mexico. University
of Texas at Austin Bureau of Business Research,
Arreola-Lizárraga, J. A. 1995. Diagnosis ecológica de
bahía de Lobos, Sonora, Mexico. Tesis de Maestría
Arriaga-Cabrera, L., V. Aguilar-Sierra, and J. Alcocer-
Durand. 2000. Aguas Continentales y Diversidad
Biológica en Mexico. D. F. Comisión Nacional para el
Conocimiento y Uso de la Biodiversidad (CONABIO).
Mexico City.
Ayala-Perez, L. A., O. A. Avilés-Alatriste, and J. L. Rojas-
Galaviz. 1998. Fish community structure in the
Candelaria-Penlau system, Campeche, Mexico. Revista
de Biología Tropical 46(3):763–774.
Baumann, R. W., and B. C. Kondratieff. 1996. Plecoptera.
In J. Llorente, A. N. García, and E. González (eds.),
Biodiversidad, Taxonomía y Biogeografía de artrópo-
dos de Mexico: Hacia una Síntesis de su Conocimiento,
pp. 169–174. Universidad Nacional Autónoma de
Mexico, Mexico City.
Beach, T. 1998. Soil catenas, tropical deforestation, and
ancient and contemporary soil erosion in the Peten,
Guatemala. Physical Geography 19:378–405.
Behnke, R. J., and J. R. Tomelleri. 2002. Trout and salmon
of North America. Free Press, New York.
Borowsky, R. 1996. The sierra de El Abra of northeastern
Mexico: Blind fish in the world’s largest cave system.
Tropical Fish Hobbyist 44(7):178–188.
Breedlove, D. E. 1981. Introduction to the flora of Chiapas:
Part 1. In D. E. Breedlove (ed.), Flora of Chiapas, pp.
1–33. California Academy of Sciences.
Bueno-Soria, J., S. Santiago-Fragoso, and R. Barba-
Alvarez. In press. Insectos acuáticos. In J. Bueno-Soria,
F. Alvares-Nogueda, and S. Santiago-Fragoso (eds.),
Biodiversity of Tabasco, Mexico. Instituto de Biologia,
UNAM and CONABIO. Mexico City.
Campoy-Favela, J., A. Varela-Romero, and L. Juárez-
Romero. 1989. Observaciones sobre la ictiofauna
nativa de la cuenca del Rio Yaqui, Sonora, Mexico.
Ecológica 1(1): 1–13.
Calderón-Aguielra, L. E., and J. R. Campoy-Favela. 1993.
Bahía de Las Guásimas, Estero Los Algodones y Bahía
de Lobos, Sonora. In S. I. Salazar-Vallejo and N. E.
González (eds.), Biodiversidad Marina y Costera de
Mexico, pp. 411–419. Comisión Nacional Para la
Biodiversidad (CONABIO) y Centro de Investigación
de Quintana Roo, Mexico City.
Castro-Aguirre, J. L., H. Espinosa Pérez, and J. J.
Schmitter-Soto. 1999. Ictiofauna estuarino-lagunar y
vicaria de Mexico. Editorial Limusa, Mexico.
Center for Space Research. 2003. An update of surface
water availability in the Rio Grande basin of Mexico.
University of Texas at Austin Center for Space
Research, Austin.
Coe, M. D., and R. Koontz. 2002. Mexico: From the
Olmecs to the Aztecs. 5th ed. Thames and Hudson,
New York.
Comisión Nacional del Agua (CNA). 2001. Estudio sobre
Disponibilidad y Balance Hidráulico Actualizado de
Aguas Superficiales de la Región Hidrológica no. 16.
C. Rio Armería, internal document by Altiplano de
Ingeniería S.A de C.V.
Contreras-Balderas, S. 1978. Speciation aspects and man-
made community composition changes in Chihuahuan
Desert fishes. In R. H. Wauer and D. H. Riskind (eds.),
Transactions of the symposium on the biological
resources of the Chihuahuan Desert region, United
States and Mexico, pp. 405–431. National Park Service
Proceedings 3. U.S. Department of the Interior,
Washington, D.C.
Contreras-Balderas, S., R. J. Edwards, Mad. L. Lozano-
Vilano, and M. E. Garcia-Ramirez. 2003. Fish bio-
diversity changes in the Lower Rio Grande/Rio Bravo,
1953–1996. Reviews in Fish Biology and Fisheries
Contreras-Balderas, S., and M. L. Lozano-Vilano. 1994.
Water, endangered fishes, and development perspec-
tives in arid lands of Mexico. Conservation Biology
Contreras-Balderas, S., and M. L. Lozano-Vilano. 1996.
Extinction of most Sandia and Potosí valleys (Nuevo
León, Mexico) endemic pupfishes, crayfishes and snails.
Ichthyological Explorations of Freshwaters 7:33–
Contreras-Balderas, S., H. Obregón, and M. L. Lozano-
Vilano. 1996. Punta del Morro, an interesting barrier
for distributional patterns of continental fishes in north
and central Veracruz, Mexico. Acta Biológica Vene-
zolana 16:37–42.
Contreras-Ramos, A. 1998. Systematics of the dobsonfly
genus Corydalus (Megaloptera: Corydalus). Thomas
Say Publications in Entomology, Lanham, Maryland.
Crews-Meyer, K. A., P. F. Hudson, and R. Colditz. 2004.
Landscape complexity and remote classification in
eastern Mexico: Applications of Landsat 7 ETM data.
Geocarto International 19:45–56.
Dean, A. 2004. Drought enters ninth year in birthplace
of the green revolution: Crisis for farmers in the
Yaqui Valley. Encina Columns, Spring 2004. Biannual
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1068
newsletter of the Stanford Institute for International
Studies, Stanford University.
de Cserna, Z. 1989. An outline of the geology of Mexico.
In A. W. Bally and A. R. Palmer (eds.), The geology of
North America: An overview, pp. 233–264. Geologi-
cal Society of America, Boulder, Colorado.
DeMarais. B. D. 1991. Gila eremica, a new cyprinid fish
from Northwestern Sonora, Mexico. Copeia 1991:
DeMarais, B. D., and W. L. Minckley. 1993. Genetics and
morphology of Yaqui chub Gila purpurea, an endan-
gered cyprinid fish subject to recovery efforts. Biologi-
cal Conservation 66:195–206.
Diaz, H. F., and G. N. Kiladis. 1992. Atmospheric tele-
connections associated with the extreme phase of the
Southern Oscillation. In H. F. Diaz and V. Markgraf
(eds.), El Nino: Historical and paleoclimatic aspects
of the Southern Oscillation, pp. 7–28. Cambridge
University Press, Cambridge.
Doolittle, W. E. 1985. The use of check dams for protect-
ing downstream agricultural lands in the prehistoric
Southwest: A contextual analysis. Journal of Anthro-
pological Research 41:279–305.
Doolittle, W. E. 1988a. Intermittent use and agricultural
change on marginal lands: The case of smallholders
in eastern Sonora, Mexico. Geografiska Annaler 70B:
Doolittle, W. E. 1988b. Pre-Hispanic occupance in the
Valley of Sonora, Mexico: Archaeological confirmation
of early Spanish reports. Anthropological Papers of the
University of Arizona. Tucson, Arizona.
Doolittle, W. E. 2000. Cultivated landscapes of native
North America. Oxford University Press, Oxford.
Douglas, M. W., R. Maddox, K. Howard, and S. Reyes.
1993. The Mexican monsoon. Journal of Climate
Dowling, T. E., D. P. Martasian, and W. R. Jeffery. 2002.
Evidence for multiple genetic forms with similar eyeless
phenotypes in the blind cavefish, Astyanax mexicanus.
Molecular Biology and Evolution 19:446–455.
DRSBM–IMECBIO. 2000. Propuesta para que se Con-
sidere la Cuenca del Rio Ayuquila–Armería–Manantlán
(Región Hidrológica XVI, Armería–Coahuayana)
Como Región Hidrológica Prioritaria de la Comisión
Nacional para el Conocimiento y Uso de la Biodiversi-
dad (CONABIO). Prepared by the Dirección de
la Reserva de la Biosfera Sierra de Manantlán–
Comisión Nacional de Areas Naturales Protegidas–
SEMARNAP and the Instituto Manantlán de Ecología
y Conservación de la Biodiversidad. Departamento
de Ecología y Recursos Naturales, Universidad de
Edwards, R. J., G. P. Garrett, and E. Marsh-Matthews.
2002a. Conservation and status of the fish communi-
ties inhabiting the Rio Conchos basin and middle Rio
Grande, Mexico and U.S.A. Reviews in Fish Biology
and Fisheries 12:119–132.
Edwards, R. J., G. P. Garrett, and E. Marsh-Matthews.
2002b. An ecological analysis of fish communities
inhabiting the Rio Conchos basin. In M. d. L. Lozano-
Vilano (ed.), Libro Jubilar en Honor al Dr. Salvador
Contreras Balderas, pp. 43–62. Universidad Autónoma
de Nuevo León, Facultad de Ciencias Biológicas,
Monterrey, Nuevo León, Mexico.
Ekholm, G. F. 1944. Excavations at Tampico and Pánuco
in the Huasteca, Mexico. Anthropological Papers of the
American Museum of Natural History, V. XXVII, Part
V: 508.
Fish, J. E. 1977. Karst hydrogeology and geomorphology
of the Sierra de el Abra and the Valles–San Luis Potosí
region, Mexico. Ph.D. diss., McMaster University,
Hamilton, Ontario.
Flores-Villela, O. 1993. Herpetofauna Mexicana: Annoted
list of the species of amphibians and reptiles of Mexico,
recent taxonomic changes, and new species. Carnegie
Museum of Natural History Special Publication no. 17.
Galindo, I. 1995. La oscilación del sur, El Nino: el caso de
Mexico. In E. Florescano and S. Swan (eds.), Breve
Historia de la Sequia en Mexico, pp. 133–165. Uni-
versidad Veracruzana, Xalapa, Ver, Mexico.
Gallo-Reynoso, J. P. 1997. Situación y distribución de las
nutrias en Mexico, con énfasis en Lutra longicaudis
annectens Major, 1897. Revista Mexicana de Masto-
zoología 2(1): 10–32.
Garcia de León, F. J., D. Gutiérrez Tirado, D. A.
Hendrickson, and H. Espinosa-Pérez. In press. Fishes
of the continental waters of Tamaulipas: Diversity and
conservation status. In J.-L. E. Cartron, G. Ceballos,
and R. S. Felger (eds.), Biodiversity, ecosystems, and
conservation in Northern Mexico. Oxford University
Press, New York.
Grubb, H. F., and J. J. R. Carillo. 1988. Region 23, Gulf
of Mexico Coastal Plain. In W. Back, J. S. Roshein,
and P. R. Seaber (eds.), Hydrogeology: The geology of
North America, pp. 219–228. V. O-2 Geological
Society of America, Boulder, Colorado.
Guerra, L. V. 1952. Ichthyological survey of the Rio
Salado, Mexico. Master thesis, University of Texas,
Guerrero, M. T., F. de Villa, M. Kelly, C. Reed, and B.
Vegter. 2001. The forest industry in the Sierra Madre
of Chihuahua: Economic, ecological and social impacts
post NAFTA. Texas Center for Policy Studies, Austin.
Guillette, E. A., M. M. Meza, M. G. Alquilar, A. D. Soto,
and I. E. Garcia. 1998. An anthropological approach
to the evaluation of preschool children exposed to pes-
ticides in Mexico. Environmental Health Perspectives
Gunn, J., J. F. William, and R. R. Hubert. 1995. A land-
scape analysis of the Candelaria watershed in Mexico:
Insights into paleoclimates affecting upland horticul-
ture in the southern Yucatán Peninsula semi-karst.
Geoarchaeology 10:3–42.
Literature Cited
Ch23.qxd 3/24/05 2:18 PM Page 1069
Gutiérrez, M., and P. Borrego. 1999. Water quality assess-
ment of the Rio Conchos, Chihuahua, Mexico. Envi-
ronmental Internacional 25:573–583.
Hard, R. J., and J. R. Roney. 1998. A massive terraced
village complex in Chihuahua, Mexico, 3000 years
before present. Science 279:1661–1664.
Hendrickson, D. A. 1986. New distribution records for
native and exotic fishes in Pacific drainages of north-
ern Mexico (in English and Spanish). Journal of
Arizona–Nevada Academy of Sciences 18(2):33–38.
Hendrickson, D. A., H. Espinosa-Pérez, L. T. Findley, W.
Forbes, J. R. Tomelleri, R. L. Mayden, J. L. Nielsen, B.
Jensen, G. Ruiz-Campos, A. Varela-Romero, A. M. Van
Der Heiden, F. Camarena, and F. J. García de León.
2002. Mexican native trouts: A review of their history
and current systematic and conservation status.
Reviews in Fish Biology and Fisheries 12:273–316.
Hendrickson, D. A., and J. K. Krejca. 2000. Subterranean
freshwater biodiversity in northeastern Mexico and
Texas. In R. A. Abell, D. M. Olson, E. Dinerstein,
P. T. Hurley, J. T. Diggs, W. Eichbaum, S. Walters, W.
Wettengel, T. Allnutt, C. J. Loucks, and P. Hedao (eds.),
Freshwater ecoregions of North America: A conserva-
tion assessment, pp. 41–43. Island Press, Washington,
Hendrickson, D. A., J. K. Krejca, and J. M. Rodriguez.
2001. Mexican blindcats, genus Prietella (Ictaluridae):
Review and status based on recent explorations. Envi-
ronmental Biology of Fishes 62:315–337.
Hendrickson, D. A., J. C. Marks, A. B. Moline, E. Dinger,
and A. E. Cohen. In press. Combining ecological
research and conservation: A case study in Cuatro
Ciénegas, Mexico. In L. Stevens and V. J. Meretsky
(eds.), Every last drop: Ecology and conservation of
North American desert springs. University of Arizona
Press, Tucson.
Hendrickson, D. A., and W. L. Minckley. 1985. Ciénegas:
Vanishing aquatic climax communities of the American
Southwest. Desert Plants 6:131–175.
Hendrickson, D. A., W. L. Minckley, R. R. Miller, D. J.
Siebert, and P. H. Minckley. 1981. Fishes of the Rio
Yaqui basin, Mexico and United States. Journal of
Arizona–Nevada Academy of Sciences 15(3): 65–
Hendrickson, D. A., and A. Varela-Romero. 2002. Fishes
of the Rio Fuerte, Sonora, Sinaloa and Chihuahua,
Mexico. In M. d. L. Lozano-Vilano (ed.), Libro Jubilar
en Honor al Dr. Salvador Contreras Balderas,
pp. 171–195. Universidad Autónoma de Nuevo León,
Facultad de Ciencias Biológicas, Monterrey, Nuevo
León, Mexico.
Hubbs, C., R. J. Edwards, and G. P. Garrett. 1991. An
annotated checklist of the freshwater fishes of Texas.
Texas Journal of Science Suppl. no. 43(4):1–56.
Hudson, P. F. 2000. Discharge, sediment, and channel
characteristics of the Rio Pánuco, Mexico. Yearbook,
Conference of Latin Americanist Geographers 26:61–
Hudson, P. F. 2002. Floodplain styles of the lower Panuco
basin, Mexico. Journal of Latin American Geography
Hudson, P. F. 2003a. Event sequence and sediment exhaus-
tion in the Lower Pánuco basin, eastern Mexico.
Catena 52:57–76.
Hudson, P. F. 2003b. The influence of the El Nino South-
ern Oscillation on sediment yield in the Lower Pánuco
basin, Mexico. Geografiska AnnalerA85(3–4):
Hudson, P. F. 2004. The geomorphic context of prehistoric
Huastec floodplain environments: Pánuco basin,
Mexico. Journal of Archaeological Science 31:653–
Hudson, P. F., and R. Colditz. 2003. Flood delineation in
a large and complex alluvial valley: The lower Pánuco
basin, Mexico. Journal of Hydrology 280:229–245.
Hudson P. F., R. Colditz, and M. Aguilar-Robledo. In
review. Land use/land cover mapping of floodplain
environments within a large alluvial valley, lower
Panuco basin, Mexico. Environmental Management.
Hudson, P. F., and F. T. Heitmuller. 2003. Local- and water-
shed-scale controls on the spatial variability of natural
levee deposits in a large fine-grained floodplain: Lower
Panuco basin, Mexico. Geomorphology 56:255–
Hulsey, C. D., F. J. García de León, Y. Sánchez Johnson,
D. A. Hendrickson, and T. J. Near. 2004. Temporal
diversification of mesoamerican cichlid fishes across a
major biogeographic boundary. Molecular Phylogeny
and Evolution 31:754–764.
Hunt, C. B. 1974. Natural regions of the United States and
Canada. W. H. Freeman and Company, San Francisco.
INE-SEMARNAP. 2000. Programa de Manejo de la
Reserva de la Biosfera Montes Azules, Mexico.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1976a. Carta Topográfica, 1:50,000, H13-
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1976b. Carta Topográfica, 1:50,000, H13-
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1976c. Carta Topográfica, 1:50,000, H13-
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1978. Carta Topográfica, 1:50,000, H13-
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1981a. Carta Hidrologica Aguas Superfi-
ciales, 1:1,000,000, Merida.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1981b. Carta Hidrologica Aguas Superfi-
ciales, 1:1,000,000, Mexico.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1981c. Carta Hidrologica Aguas Superfi-
ciales, 1:1,000,000, Villahermosa.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1984a. Carta de Efectos Climaticos,
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1070
Regionales Noviembre–Abril and Regionales Mayo–
Octubrei data record from 1921–1980, 1:250,000,
Ciudad Mante, F14-5.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1984b. Carta de Efectos Climaticos,
Regionales Noviembre–Abril and Regionales Mayo–
Octubre, data record from 1921–1980, 1:250,000,
Ciudad valles, F14-8.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1984c. Carta uso del Suelo y Vegetacion, data
record in 1981, 1:250,000, Ciudad Valles, F14-8.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1984d. Carta Geologica, 1:250,000, Ciudad
Mante, F14-5.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1984e. Carta Geologica, 1:250,000, Ciudad
valles, F14-8.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1998a. Carta Edafologica, 1:1,000,000,
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1998b. Carta Topografica, 1:250,000,
Ciudad del Carmen, E15-6.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1998c. Carta Topografica, 1:250,000,
Tenosique. E15-9.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1998d. Carta Topográfica, 1:250,000, San
Juanito, G13-1.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 1998e. Carta Topográfica, 1:250,000,
Ojinaga, H13-8.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 2000a. Síntesis de información Geográfica
del Estado de Chihuahua. Instituto Nacional de
Estadística, Geografía e Informática, Mexico D.F.
Instituto Nacional de Estadística Geografía e Informática
(INEGI). 2000b. Síntesis de información Geográfica del
Estado de Sonora. Instituto Nacional de Estadística,
Geografía e Informática, Mexico D.F.
Jauregui, E. 1995. Rainfall fluctuations and tropical storm
activity in Mexico. Erdkunde 49:39–48.
Jeffery, W. R. 2001. Cavefish as a model system in evolu-
tionary developmental biology. Developmental Biology
Johnson, J. J., and B. L. Jensen. 1991. Hatcheries for
endangered freshwater fishes. In W. L. Minckley and
J. E. Deacon (eds.), Battle against extinction, pp.
199–217. University of Arizona Press.
Kelly, M. E. 2001. The Rio Conchos: A preliminary
overview. Texas Center for Policy Studies, Austin.
Langecker, T. G., B. Neumann, C. Hausberg, and J.
Parzefall. 1995. Evolution of the optical releasers for
aggressive behavior in cave-dwelling Astyanax fascia-
tus (Teleostei, Characidae). Behavioural Processes
Leibfried, W. 1991. A recent survey of fishes from the inte-
rior Rio Yaqui drainage, with a record of flathead
catfish Pylodictis olivaris from the Rio Aros. In Pro-
ceedings XX Desert Fishes Council Symposia, Death
Valley, California.
Lesser, J. M., and A. E. Weidie. 1988. Region 25, Yucatán
Peninsula. In W. Back, J. S. Roshein, and P. R. Seaber
(eds.), Hydrogeology: The geology of North America,
pp. 237–241. V. O-2. Geological Society of America,
Boulder, Colorado.
Leviton, A. E., R. H. J. Gibbs, E. Heal, and C. E. Dawson.
1985. Standards in herpetology and ichthyology. Part
1: Standard symbolic codes and institutional resource
collections in herpetology and ichthyology. Copeia
Leviton, A. E., and R. H. J. Gibbs. 1988. Standards in her-
petology and ichthyology. Standard symbolic codes for
institutional resource collections in herpetology and
ichthyology. Supplement no. 1: Additions and correc-
tions. Copeia 1988:280–282.
Lot, A., and A. Novelo. 1988. El Pantano de Tabasco y
Campeche: la reserva más importante de plantas
acuáticas de Mesoamérica. In Memorias del simposio
internacional sobre ecología y conservación del delta
de los ríos Usumacinta y Grijalva, pp. 537–547.
INIREB y Gobierno del Estado de Tabasco, Mexico
City. Lozano-Vilano, M. d. L. 2002. Cyprinodon
salvadori, new species from the upper Rio Conchos,
Chihuahua, Mexico, with a revised key to the C.
eximius complex (Pisces, Teleostei: Cyprinodontidae).
In M. d. L. Lozano-Vilano (ed.), Libro Jubilar en
Honor al Dr. Salvador Contreras Balderas, pp. 15–22.
Universidad Autónoma de Nuevo León, Facultad de
Ciencias Biológicas, Monterrey, Nuevo León, Mexico.
Lyons, J., G. González-Hernández, E. Soto-Galera, and
M. Guzmán-Arroyo. 1998. Decline of freshwater fishes
and fisheries in selected drainages of west-central
Mexico. Fisheries 23(4):10–18.
Lyons, J., S. Navarro-Pérez, P. Cochran, E. Santana, and
M. Guzmán-Arroyo. 1995. Index of biotic integrity
based on fish asssemblages for the conservation of
streams and rivers in West Central Mexico. Conserva-
tion Biology 9:569–584.
March, I., E. Naranjo, and R. Rodiles-Hernández. 1996.
Diagnóstico para la conservación y manejo de la fauna
silvestre en la Selva Lacandona, Chiapas. Informe Final
ECOSUR, San Cristóbal de las Casas.
Martínez, L. M., A. Carrranza, and M. García. 1999.
Aquatic ecosystem pollution of the Ayuquila River,
Sierra de Manantlán Biosphere Reserve, Mexico. In M.
Munawar, S. Lawrence, I. F. Munawar, and D. Malley
(eds.), Aquatic ecosystems of Mexico: Status and scope,
pp. 1–17. Ecovision World Monograph Series. Black-
huys, Leiden, The Netherlands.
Mayden, R. L., R. H. Matson, and D. M. Hillis. 1992. Spe-
ciation in the North American genus Dionda (Teleostei:
Cypriniformes). In R. L. Mayden (ed.), Systematics,
historical ecology, and North American freshwater
fishes, pp. 710–746. Stanford University Press, Palo
Alto, California.
Literature Cited
Ch23.qxd 3/24/05 2:18 PM Page 1071
Metcalfe, S. E. 1987. Historical data and climatic change
in Mexico: A review. The Geographical Journal 153:
Miller, R. R. 1961. Man and the changing fish fauna of the
American Southwest. Papers of the Michigan Academy
of Science, Arts, and Letters 46:365–404.
Miller R. R. 1978. Composition and derivation of the
native fish fauna of the Chihuahuan Desert region. In
R. H. Wauer and D. H. Riskind (eds.), Transactions of
the symposium on biological resources of the Chi-
huahuan desert region, U.S. and Mexico, pp. 365–381.
National Park Service Proceedings 3. U.S. Department
of the Interior, Washington, D.C.
Miller, R. R. 1986. Composition and derivation of the
freshwater fish fauna of Mexico. Anales de la Escuela
Nacional de Ciencias Biológicas 30:121–153.
Miller, R. R., W. L. Minckley, and S. M. Norris. 2005.
Freshwater fishes of Mexico. University of Chicago
Press, Chicago.
Miller, R. R., and M. L. Smith. 1986. Origin and geogra-
phy of the fishes of central Mexico. In C. H. Hocutt
and E. O. Wiley (eds.), The zoogeography of North
American freshwater fishes, pp. 487–517. John Wiley
and Sons, New York.
Minckley, W. L. 1969. Environments of the Bolsón of
Cuatro Ciénegas, Coahuila, Mexico. Science Series,
University of Texas, El Paso, Texas 2:1–65.
Minckley, W. L., and D. E. Brown. 1982. Wetlands. In
D. E. Brown (ed.), Biotic communities of the American
Southwest, United States and Mexico. Special issue,
Desert Plants 4:222–341.
Minckley, W. L., and D. E. Brown. 1994. Wetlands. In D.
E. Brown (ed.), Biotic Communities: Southwestern
United States and Northwestern Mexico, pp. 222–287,
University of Utah Press, Salt Lake City.
Minckley, W. L., D. A. Hendrickson, and C. E. Bond. 1986.
Geography of western North American freshwater
fishes: Description and relations to intracontinental
tectonism. In C. H. Hocutt and E. O. Wiley (eds.),
Zoogeography of freshwater fishes of North America,
pp. 5l9–614. Wiley Interscience, New York.
Minckley, W. L., R. R. Miller, C. D. Barbour, J. J.
Schmitter-Soto, and S. M. Norris. 2005. Historical
ichthyogeography. In R. R. Miller, W. L. Minckley,
and S. M. Norris (eds.), Freshwater fishes of Mexico.
University of Chicago Press, Chicago.
Mitchell, R. W., W. H. Russell, and W. R. Elliott. 1977.
Mexican eyeless characin fishes of the genus Astyanax:
Environment, distribution and evolution. Texas Tech
University Press, Lubbock.
Morales-Román, M., and R. Rodiles-Hernández. 2000.
Implicaciones de Ctenopharyngodon idella en la comu-
nidad de peces del río Lacanjá, Chiapas. Hidrobiológ-
ica 10(1):13–24.
Muir, J. M. 1936. Geology of the Tampico Region, Mexico.
American Association of Petroleum Geologists, Tulsa,
Nanson, G. C., and J. C. Croke. 1992. A genetic classifi-
cation of floodplains. Geomorphology 4:459–486.
Nelson, J. S., E. J. Crossman, H. Espinosa-Pérez, L. T.
Findley, C. R. Gilbert, R. N. Lea, and J. D. Williams.
2004. Common and scientific names of fishes from the
United States, Canada and Mexico. Special Publication
29. American Fisheries Society, Bethesda, Maryland.
Obregón-Barboza, H., S. Contreras-Balderas, and M. L.
Lozano-Vilano. 1994. The fishes of northern and
central Veracruz, Mexico. Hydrobiologia 286(2):79–
Ocaña, D., and A. Lot. 1996. Estudio de la vegetación
acuática vascular del sistema fluvio-lagunar-deltaico
del Rio Palizada, en Campeche, Mexico. Anales del
Instituto de Biologia, Universidad Nacional Autonoma
de Mexico, Series Botánica, 67:303–327.
O’Hara, S. L., F. A. Street-Perrott, and T. P. Burt. 1993.
Accelerated soil erosion around a Mexican highland
lake caused by prehispanic agriculture. Nature 362:48–
Pope, K. O., and B. H. Dahlin. 1989. Ancient Maya
wetland agriculture: New insights from ecological and
remote sensing research. Journal of Field Archaeology
Quattro, J. M., J. C. Avise, and R. C. Vrijenhoek. 1992.
An ancient clonal lineage in the fish genus Poeciliopsis
(Atheriniformes: Poeciliidae). Proceedings of the
Academy of Natural Sciences of Philadephia 89:348–
Quattro, J. M., P. L. Leberg, M. E. Douglas, and R. C.
Vrijenhoek. 1996. Molecular evidence for a unique
evolutionary lineage of endangered Sonoran desert
fish (genus Poeciliopsis). Conservation Biology
Rauchenberger, M., K. D. Kallman, and D. C. Morizot.
1990. Monophyly and geography of the Pánuco Basin
swordtails (genus Xiphophorus) with description of
four new species. American Museum Novitates
Revenga, C., S. Murray, J. Abramovitz, and A. Hammond.
1998. Watersheds of the world: Ecological value and
vulnerability. World Resources Institute, Washington,
Ricketts, T. H., E. Dinerstein, D. M. Olson, C. J. Loucks,
W. Eichbaum, D. DellaSala, K. Kavanaugh, P. Hedao,
P. T. Hurley, K. M. Carney, R. Abell, and S. Walters.
1999. Terrestrial ecosystems of North America: A con-
servation assessment. Island Press, Washington, D.C.
Rodiles-Hernández, R. 2004. Diversidad de peces conti-
nentales en Chiapas. In M. González-Espinosa, N.
Ramírez-Marcial, and L. Ruíz-Montoya (eds.), Diver-
sidad Biológica de Chiapas, pp. 141–160. Plaza y
Valdés, ECOSUR, COCYTECH, Distrito Federal,
Rodiles-Hernández R., J. Cruz-Morales, and S.
Domínguez. 2002. El sistema lagunar de playas de
Catzajá, Chiapas, Mexico. In E. G. de la Lanza and
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1072
J. L. Garciá-Calderón (eds.), Lagos y presas de Mexico,
pp. 327–337. Distrito Federal., Mexico.
Rodiles-Hernández, R., E. Díaz-Pardo, and J. Lyons. 1999.
Patterns in the species diversity and composition of the
fish community of the Lacanjá River, Chiapas, Mexico.
Journal of Freshwater Ecology 14:455–468.
Rodiles-Hernández, R., S. Domínguez C., and E. Velásquez
V. 1996. Diversidad íctica del Rio Lacanjá, Selva
Lacandona, Chiapas, Mexico. Zoología Informa
Rodiles-Hernández, R., D. Hendrickson, J. Lundberg, and
J. Alves. 2000. A new siluriform family from southern
Mexico. In 80th Annual Meeting American Society of
Ichthyologist and Herpetologists, La Paz, Baja Cali-
fornia, Mexico.
Rodiles-Hernández, R., J. G. Lundberg, and D. A.
Hendrickson. 2004. Diagnosis of the “Chiapas
catfish”: An apparently ancient siluriform lineage from
Mesoamerica. In 84th Annual Meeting American
Society of Ichthyologist and Herpetologists. ASIH,
Norman, Oklahoma.
Ryan, M. J., and G. G. Rosenthal. 2001. Variation and
selection in swordtails. In L. A. Dugatkin (ed.), Model
systems in behavioral ecology: Integrating conceptual,
theoretical, and empirical approaches, pp. 133–
148. Princeton University Press, Princeton, New
Santana-Michel, F. J., H. Luis-Guzmán, R. Enrique-
Sánchez, and G. Ramón-Cuevas. 2000. Flora y veg-
etación del Rio Ayuquila. In Memorias del VIII
Simposio Interno sobre Inventarios, Manejo de Recur-
sos Naturales y Desarrollo Comunitario. Universidad
de Guadalajara, Instituto Manantlán de Ecología y
Conservación de la Biodiversidad, Department of
Natural Resources, Centro Universitario de la Costa
Schaafsma, C. F., and C. L. Riley. 1999. The Casas Grandes
world. University of Utah Press, Salt Lake City.
Secretaría de Infraestructura Urbana y Ecología (SIUE).
1992. Fauna Sonorense. Gobierno del Estado de
Sonora, Secretaría de Infraestructura Urbana y
Ecología, Programa Ambiental Estatal-
Secretaría del Medio Ambiente y Recursos Naturales
(SEMARNAT). 2002. NORMA Oficial Mexicana
NOM-059-ECOL-2001, Protección ambiental-
Especies nativas de Mexico de flora y fauna silvestres-
Categorías de riesgo y especificaciones para su
inclusión, exclusión o cambio-Lista de especies en
riesgo. Diario Oficial De La Federación, Mexico Miér-
coles 6 de marzo de 2002(Segunda Sección):1–85.
Siebert, D. J., and W. L. Minckley. 1986. Two new catosto-
mid fishes (Cypriniformes) from the northern Sierra
Madre Occidental of Mexico. American Museum
Novitates 2849:1–17
Siemens, A. H. 1980. Wetland agriculture in pre-hispanic
Mesoamerica. Geographical Review 70:166–181.
Siemens, A. H. 1983. Oriented raised fields in Central
Veracruz. American Antiquity 48:85–102.
Siemens, A. H., and J. A. Soler-Graham. 2003. Manejo pre-
hispanico del Rio Candelaria, Campeche. Arqueologia
Mexicana 10(59):64–69.
Sluyter, A. 1994. Intensive wetland agriculture in
Mesoamerica: Space, time, and form. Annals of
the Association of American Geographers 84:557–
Smith M. L., and R. R. Miller. 1986. The evolution of the
Rio Grande basin as inferred from its fish fauna. In
C. H. Hocutt and E. O. Wiley (eds.), The zoogeogra-
phy of North American freshwater fishes, pp. 457–485.
John Wiley and Sons, New York.
Stebbins, R. C. 1985. A field guide to the western reptiles
and amphibians. 2nd ed. National Audubon Society.
Tortajada, C., and A. K. Biswas. 1997. Environmental
management of water resources in Mexico. Water
International 22(3):172-178.
Trager, E. A. 1926. The geologic history of the Pánuco
River Valley and its relation to the origin and accumu-
lation of oil in Mexico. American Association of Petro-
leum Geologists Annual Meeting: 667–696.
U.S. Fish and Wildlife Service (USFWS). 1994. Yaqui fishes
recovery plan. U.S. Fish and Wildlife Service, Albu-
querque, New Mexico.
Vannote, R. L., G. W. Minshall, K. W. Cummins, J. R.
Sedell, and C. E. Cushing. 1980. The River Continuum
Concept. Canadian Journal of Fisheries and Aquatic
Sciences 37:130–137.
Varela-Romero, A. 1989. Dorosoma petenense (Günther),
un nuevo registro para la Cuenca del Rio Yaqui,
Sonora, Mexico (Pisces:Clupeidae). Ecológica
Varela-Romero, A. 1995. Perspectivas de recuperación y
cultivo de peces nativos en el Noroeste de Mexico. Pub-
licaciones CICTUS 3:1–6.
Vázquez Sánchez, M. A., and M. A. Ramos Olmos. 1992.
Reserva de la Biosfera Montes Azules, Selva Lacan-
dona: Investigación para su Conservación. ECOS-
FERA, Mexico.
Vrijenhoek, R. C. 1984. The evolution of clonal diversity
in Poeciliopsis. In B. J. Turner (ed.), Evolutionary
genetics of fishes, pp. 399–429. Plenum Press, New
Vrijenhoek, R. C. 1993. The origin and evolution of clones
versus the maintenance of sex in Poeciliopsis. Journal
of Heredity 84:388–395.
Walsh, S. J., and C. R. Gilbert. 1995. New species of
troglobitic catfish of the genus Prietella (Siluriformes:
Ictaluridae) from northeastern Mexico. Copeia
West, R. C., and J. P. Augelli. 1989. Middle America, its
lands and peoples. Prentice-Hall, Englewood Cliffs,
New Jersey.
West, R. C., N. P. Psuty, and B. G. Thom. 1969. The
Tabasco Lowlands of Southeastern Mexico. Technical
Literature Cited
Ch23.qxd 3/24/05 2:18 PM Page 1073
Report no. 70., Coastal Studies Institute, Louisiana
State University, Baton Rouge.
Whalen, M. E., and P. E. Minnis. 2001. Casas Grandes and
its hinterlands: Prehistoric regional organization in
Northwest Mexico. University of Arizona Press,
Whitmore, T. M., and B. L. Turner II. 2002. Cultivated
landscapes of Middle America on the eve of conquest.
Oxford University Press, Oxford.
Wilcox, T. P., F. J. García de León, D. A. Hendrickson, and
D. M. Hillis. 2004. Convergence among cave catfishes:
Long-branch attraction and a Bayesian relative rates
test. Molecular Phylogenetics and Evolution 31:
23 Rivers of Mexico
Ch23.qxd 3/24/05 2:18 PM Page 1074
Relief: 3800m
Basin area: 98,227km2(with Tamesí) 79,100 km2
(without Tamesí)
Mean discharge: 537.4m3/s (with Tamesí)
472.8m3/s (without Tamesí)
River order: NA
Mean annual precipitation: 96cm
Mean air temperature: 20°C
Mean water temperature: NA
Physiographic provinces: Neovolcanic Plateau (NP),
Sierra Madre Oriental (SO), Mexican Gulf
Coastal Plain (CP)
Biomes: Tropical Savanna, Mexican Montane Forest,
Freshwater ecoregion: Tamaulipas–Veracruz
Terrestrial ecoregions: Alvarado Mangroves, Veracruz
Moist Forests, Sierra Madre Oriental Pine–Oak
Forests, Veracruz Montane Forests, Central
Mexican Matorral, Meseta Central Matorral
Number of fish species: >88 (>80 native) (not
including Río Tamesí)
Number of endangered species: 7 fishes
Major fishes: Media Luna cichlid, blackcheek cichlid, Chairel cichlid, slender cichlid, bluetail splitfin, dusky goodea, relict
splitfin, jeweled splitfin, Media Luna pupfish, sheepshead swordtail, short-sword platyfish, delicate swordtail, highland
swordtail, Moctezuma swordtail, Pánuco swordtail, variable platyfish, spottail chub, Pánuco minnow, bicolor minnow,
chubsucker minnow, flatjaw minnow, Río Verde catfish, phantom blindcat, fleshylip buffalo, Mexican tetra
Major other aquatic vertebrates: NA
Major benthic invertebrates: mollusks (Hydrobia tampicoensis, Littoridina crosseana, Lithasiiopsis hinkleyi), crustaceans
(Palaemonetes mexicanus, Procambarus [Ortmannicus] ortmanii, P. [Ortmannicus] acutus cuevachicae, P. [Scapullicambarus]
strenghi, Troglomexicanus perezfafantae, T. huastecae, T. tamaulipenses), hellgrammites (Chloronia mexicana, Corydalus
Nonnative species: tilapias, grass carp, silver carp, channel catfish, blue catfish, largemouth bass, water hyacinth
Major riparian plants: none
Special features: Cascada Tamul, 102m high waterfall (at confluence of Río Gallinas and Río Santa Maria); natural springs;
caves and sinkholes are common, some perhaps among world’s deepest, many with abundant aquatic habitat; some of
highest areas harbor cloud forests (El Cielo Biosphere Reserve)
Fragmentation: dams on Río Moctezuma, Río Tula, Río Topila; dam under construction on lower Río Tamuin
Water quality: highly variable, but poor in many places
Land use: sugarcane farming, citrus, and cattle ranching in Coastal Plain; traditional slash and burn, citrus, and coffee in
mountains; petroleum extraction in lower basin
Population density: low
Major information sources: Contreras-Ramos 1998, A. Contreras-Ramos, personal communication,,
FIGURE 23.11 Map of the Rio Pánuco basin. Physiographic provinces are
separated by yellow lines.
Precipitation or runoff
per month (cm)
Temperature (°C)
(monthly mean)
FIGURE 23.12 Mean monthly air temperature,
precipitation, and runoff for the Rio Pánuco basin.
Ch23.qxd 3/24/05 2:18 PM Page 1075
Relief: 3800m
Basin area: 112,550km2
Mean discharge: 2678m3/s
River order: NA
Mean annual precipitation: 199cm
Mean air temperature: 23°C
Mean water temperature: NA
Physiographic provinces: Chiapas–Guatemala Highlands (CG),
Mexican Gulf Coastal Plain (CP), Yucatan (YU)
Biomes: Tropical Rain Forest, Tropical Savanna
Freshwater ecoregion: Grijalva–Usumacinta
Terrestrial ecoregions: Peten–Veracruz Moist Forests, Chiapas
Depression Dry Forests, Pantanos de Centla, Central American Pine–
Oak Forests, Chiapas Montane Forests, Usumacinta Mangroves
Number of fish species: >112 (103 native)
Number of endangered species: 11 fishes
Major fishes: Pénjamo tetra, headwater killifish, Chiapas killifish,
widemouth gambusia, Chiapas swordtail, sulphur molly, upper Grijalva livebearer, white cichlid, Angostura cichlid, tailbar
cichlid, Petén cichlid, Montechristo cichlid, Usumacinta cichlid, Chiapa de Corzo cichlid, freckled cichlid, Teapa cichlid,
longfin gizzard shad, Lacandon sea catfish, Maya sea catfish, Maya needlefish, Mexican halfbeak, Mexican mojarra
Major other aquatic invertebrates: river crocodile, swamp crocodile, common snapping turtle, tortugas blanca, tortugas
casquito, neotropical river otter, West Indian manatee, water opossum, tepezcuintle
Major benthic invertebrates: mollusks (Pomacea, Aroapyrgus, Cochliopina, Hydrobia, Pachychilus), hellgrammites (Corydalus
luteus), caddisflies (Smicridea, Nectopsyche, Neotrichia), mayflies (Camelobaetidius, Leptophlebia, Traverella, Campsurus),
odonates (Archaeogomphus, Phyllocycla, Protoneura, Neoneura, Heteragrion), beetles (Anacaena, Brachyvatus, Laccophilus,
Helochares, Tropisternus)
Nonnative species: common carp, grass carp, rainbow trout, largemouth bass, blue tilapia, redbelly tilapia, Nile tilapia,
Mozambique tilapia, jaguar guapote, water hyacinth
Major riparian plants: Andira galeottiana, Pachira acuatica, Bravaisia integerrima, bloodwoodtree, gregorywood, Paquira
aquatica, willow, button mangrove, black mangrove, white mangrove, American (or red) mangrove, common cattail,
southern cattail, common reed
Special features: Pantanos de Centla Biosphere Reserve; Laguna del Tigre National Park in Guatemala; Selva Maya, possibly
largest remaining tropical forest in North/Central America; center of ancient Mayan culture; Montes Azules Biosphere
Reserve; Lacantún Biosphere Reserve
Fragmentation: several large dams on Río Grijalva; new dams proposed on Usumacinta
Water quality: NA
Land use: 59% forest, 31% cropland, 8% grassland/savanna/shrubland, 3% urban
Population density: 28 people/km2
Major information sources: Revenga et al. 1998, A. Contreras-Ramos, personal communication,,
Rodiles-Hernández 2004, Bueno-Soria et al. in press, March et al. 1996,
FIGURE 23.13 Map of the Rio Usumacinta basin. Physiographic provinces are separated by yellow lines.
Precipitation or runoff
per month (cm)
(Usumacinta only)
Temperature (°C)
(monthly mean)
FIGURE 23.14 Mean monthly air temperature,
precipitation, and runoff for the Rio Usumacinta basin.
Ch23.qxd 3/24/05 2:18 PM Page 1076
Relief: 375m
Basin area: 10,755km2
Mean discharge: 46m3/s
River order: NA
Mean annual precipitation: 150cm
Mean air temperature: 24.6°C
Mean water temperature: NA
Physiographic provinces: Yucatan (YU), Mexican Gulf
Coastal Plain (CP)
Biomes: Tropical Rain Forest, Tropical Savanna
Freshwater ecoregion: Grijalva–Usumacinta
Terrestrial ecoregions: Yucatan Moist Forests,
Usumacinta Mangroves, Peten–Veracruz Moist
Forests, Pantanos de Centla
Number of fish species: >65 (>17 freshwater, 48
Number of endangered species: none
Major fishes: firemouth cichlid, blackgullet cichlid,
yellow cichlid, yellowjacket, Mayan cichlid,
yellowbelly cichlid, Montechristo cichlid, redhead
cichlid, giant cichlid, stippled gambusia,
Champoton gambusia, shortfin molly, picotee
livebearer, pike killifish, banded tetra, Maya tetra, pale catfish, threadfin shad, bay anchovy, silver perch
Major other aquatic vertebrates: swamp crocodile, neotropical river otter
Major benthic invertebrates: mollusks (narrowmouth hydrobe), hellgrammites (Corydalus bidenticulatus), stoneflies
(Anacroneuria), true bugs (Abedus)
Nonnative species: none documented
Major riparian plants: cattails, breadnut tree, bay palmetto, gumbo limbo, gregorywood, mangroves
Special features: one of the few rivers flowing through the highly karstic region of the Yucatan Peninsula; large water losses to
groundwater; where ancient Mayan civilization developed
Fragmentation: no dams or reservoirs; prehistoric canal systems
Water quality: NA, but perceived to be good
Land use: slash and burn in uplands; cattle ranching in coastal plain
Population density: low, but has increased rapidly since 1960s
Major information sources: Contreras-Ramos 1998, A. Contreras-Ramos, personal communication,,
Gunn et al. 1995, Ayala-Perez et al. 1998
FIGURE 23.15 Map of the Rio Candelaria basin. Physiographic provinces are
separated by yellow lines.
Precipitation or runoff
per month (cm)
+ Groundwater
Temperature (°C)
(monthly mean)
FIGURE 23.16 Mean monthly air temperature,
precipitation, and runoff for the Rio Candelaria basin.
Ch23.qxd 3/24/05 2:18 PM Page 1077
Relief: 2520m
Basin area: 73,000km2
Mean discharge: 78.5m3/s
River order: 6
Mean annual precipitation: 48cm
Mean air temperature: 18.4°C
Mean water temperature: 18.0°C
Physiographic provinces: Sierra Madre Occidental
(SO), Basin and Range (BR), Buried Ranges (BU)
Biomes: Mexican Montane Forest, Desert
Freshwater ecoregion: Sonoran
Terrestrial ecoregions: Sonoran–Sinaloan Transitional
Subtropical Dry Forests, Sierra Madre Occidental
Pine–Oak Forests, Sonoran Desert, Chihuahuan
Desert, Sinaloan Dry Forests
Number of fish species: 107
Number of endangered species: 7 fishes (threatened)
Major fishes: Yaqui trout, Yaqui sucker, Yaqui chub,
roundtail chub, Mexican stoneroller, longfin dace,
Yaqui catfish, Yaqui topminnow, Leopold sucker,
Cahita sucker, Pacific gizzard shad, striped mullet,
striped mojarra, machete, Heller’s anchovy
Major other aquatic vertebrates: water snakes (Thamnophis spp.), Sonora mud turtle, Tarahumara frog, Tarahumara
salamander, Chiricahua leopard frog, neotropical river otter
Major benthic invertebrates: crustaceans (Cauque River prawn, blue shrimp, white shrimp), mollusks (Fossaria [Bakerilymnaea]
bulimoides), hellgrammites (Corydalus bidenticulatus, Corydalus texanus), mayflies (Siphlonurus, Nixe, Acentrella),
stoneflies (Capnia decepta, Mesocapnia frisoni, Anacroneuria wipikupa)
Nonnative species: channel catfish, blue catfish, black bullhead, largemouth bass, rainbow trout, common carp, river carpsucker,
green sunfish, bluegill, western mosquitofish, American bullfrog
Major riparian plants: Arizona sycamore, Arizona alder, Goodding willow, Bonpland willow, green ash, Fremont cottonwood,
common reed, Chinese saltcedar, velvet mesquite
Special features: basin shared between Arizona and México (Sonora and Chihuahua); spectacular canyons (barrancas); high fish
diversity for arid system, but low endemism; protected areas are Tutuaca, Papigochic, and Sierra de Ajos Bavispe
Fragmentation: three major dams on main stem; flow regulation over 50% of the basin
Water quality: NA
Land use: 62% forest, 2% cropland, 33% grassland/savanna/shrubland, 3% urban
Population density: 7 people/km2
Major information sources: Arriaga-Cabrera et al. 2000, Hendrickson et al. 1981, Contreras-Ramos 1998, Baumann and
Kondratieff 1996, Watersheds of the World CD 2003, A. Contreras-Ramos, personal communication,
FIGURE 23.17 Map of the Rio Yaqui basin. Physiographic provinces are separated
by yellow lines.
Precipitation or runoff
per month (cm)
Temperature (°C)
(monthly mean)
FIGURE 23.18 Mean monthly air temperature,
precipitation, and runoff for the Rio Yaqui basin.
Ch23.qxd 3/24/05 2:18 PM Page 1078
Relief: 2700m
Basin area: 68,386km2
Mean discharge: 20.5m3/s
River order: