COCHITI REACH: an overview of the human-made headwaters of the Middle Rio Grande.

The people of New Mexico have always been close with their river systems. The cultural heritage in New Mexico has shared a symbiotic relationship with the Rio Grande, the San Juan, the Gila, the Pecos, and the Canadian rivers for generations. Most of these traditions value not only the quality and quantity of water for drinking and agriculture, but also for the riparian ecosystems to support life and provide economical, social, and even spiritual well being to humankind. However, due to poor management techniques, these services are becoming more and more threatened as the years go by.

The Middle Rio Grande over the past hundred years has seen drastic changes to the once large dynamic river system. American settlers started coming to the Middle Rio Grande in herds after the Homestead Act of 1862. They quickly manipulated the landscape through non-sustainable grazing practices, which still has detrimental affects today. After the Great Depression, public work efforts lead to large water projects on the rivers of New Mexico. Large earthen dams were constructed on major tributaries of the Rio Grande, greatly affecting the hydrology, geomorphology, and riparian ecosystems throughout the New Mexico's reach of the Southwest's longest river system.

Over the past several decades, numerous restoration programs have been created in attempt to reverse some of the human-induced negative changes associated with the Middle Rio Grande. Such programs and partnerships include the River Ecosystem Restoration Initiative, the Urban Waters Middle Rio Grande Conservation Initiative, the Middle Rio Grande Bosque Initiative, and the Middle Rio Grande ESA Collaborative Program, just to name a few. Most of these groups use adaptive management techniques to create restoration projects to address the issues affecting the system. Most of the major restoration sites can be broken into project areas in the four major reaches of the Middle Rio Grande; the Cochiti Reach, the Albuquerque Reach, the Isleta Reach, and the San Acacia Reach. This report examines the major issues and a few solutions in regards to the Cochiti reach of the Rio Grande.
Figure 1. Map of Cochiti Reach (Image Provided by Cynthia G. Abeyta, Middle Rio Grande Bosque Initiative)

The Cochiti reach extends from the Cochiti Dam downstream 28 miles to the Highway 44 Bridge in Bernalillo. Most of the land in this part of the Rio Grande is owned by the Santa Ana, San Felipe, Santo Domingo, and Cochiti Pueblos. Much like the rest of the Middle Rio Grande, major anthropogenic activities through numerous diversion structures, dams, and levees can be found in the Cochiti Reach. Most of the water used from the river in this area is for irrigation and public supply to the local communities. A few tributaries feed into this reach such as the Jemez River, Galisteo Creek, the Santa Fe River, Borrego Canyon, and the Arroyo Tonque, all of which are only ephemeral in nature. However, when water does flow through them they transport large amounts of sediment into the river.

Over time and from physical constraints, this part of the Rio Grande has experienced drastic changes. Some of the major issues include: main channel incision, poor width-to-depth ratios, coarsening of channel bed, low sediment loads, and a disconnection with the floodplain. But due to the geography of the Cochiti Reach, there are some workable features available for restoration programs. This area doesn't have a strong influence of physical structures close to the river, which allows for room to properly manage and restore the floodplain as opposed to the lower reaches. Another feature of the reach includes a tendency to experience inundation and connectivity to the floodplain at certain flows. At optimal flows, the current state of the reach could create prominent seed sources and change sediment loads for dynamic sediment transport processes. These potential processes could be beneficial for riparian ecosystems, both within Cochiti reach and downstream. One of the most important features is the availability of water. Pulse flow events can be directly managed coming out of Cochiti Dam, creating ideal opportunities for restoration projects and goals. In order to better understand the restoration needs for the Cochiti reach, a better understanding is critical for assessing the changes and desired riparian ecosystems, geomorphology, and the hydrology for the region.

Rio Grande Silvery Minnow -Sara Gerlitz

Species Background and Overview
Figure 2. Rio Grande Silvery Minno, (H. Amarus), Albuquerque BioPark Fish Hatchery Photo Credit: Sara Gerlitz, 10/2013

The Rio Grande silvery minnow (Hybognathus amarus) was formerly the most spatially abundant and common fish with habitat ranging throughout all major tributaries. The minnow’s historic habitat included larger order streams of the Rio Grande Basin such as the Rio Chama, the Pecos River and throughout the lower reaches found along the Texas-Mexico border extending to the Gulf of Mexico (TetraTech, 2004).

The last of six minnow speices remaining in the Rio Grande the decline of the species is correlated to habitat alteration spurred from a lack of water availability due to diversions. Specifically, water diversions for irrigation are likely to be the most influential factor in the species demise. For example, irrigation canals constructed in the 1950s paralleling the river at San Acacia and flowing downstream to the Elephant Butte Reservoir have played a major role in river dynamics including return flow and water table levels. These canals caused habitat modifications and conditions in which large reaches of the river have dry, or no-flow for a given period of the irrigation season. As a result of similar habitat modifications and subsequent fragmentation, the species is found in less than 7% of its historic range from Cochiti Reservoir downstream to Elephant Butte Reservoir. This state of the species led to it’s federal listing under the Endangered Species Act in 1994. (TetraTech, 2004).

The resultant attempts to mitigate jeopardy and prevent a “take” on the species, the Rio Grande habitat management regimes proposed in the 1990s and early 2000s have become the catalyst for river restoration projects. The four major reaches of the Middle Rio Grande in New Mexico are diverse in ecology, geology and human alterations and the current population of the Rio Grande Silvery Minnow (RGSM) has been observed in limited areas, specifically south of Albuquerque, out of range for the designated Cochiti reach. (TetraTech, 2004).

The critical habitat designation in the Rio Grande for H. amarus as defined by the U.S. Fish and Wildlife Service in 2003, extends from Cochiti Dam in Sandoval County, N.M. to a utility line crossing the Rio Grande, a permanent identified landmark in Socorro County, New Mexico (TetraTech, 2004). As mentioned previously, this reach does not currently support a viable population of H. amarus, yet is important in terms of water delivery to the lower reaches. Cochiti Dam’s influence on the lower reaches directly correlates to the species survival downriver as much of the historic habitat of the Cochiti reach has been destroyed.

The designation is mainly justified in terms of the fragmentation of habitat and the effects this imposes on spawning. The minnow spawns in open water, utilizing natural drift for semi-buoyant eggs that passively relocate during development. The drifting occurs for approximately three days after hatching in which larval fish are at the will of the river. This species, H. amarus, although not fully understood, seemingly relies on snowmelt as the catalyst for reproduction. Runoff during May or June is the general spawning period however it may be possible for spawning to occur during precipitation induced high-flow events during the summer. This cue to spawn is based on assumptions and modern day observations (Bovee, 2008).

Passive drifting of eggs and larvae exposes them to displacement downstream from irrigation infrastructure like the San Acacia Dam or Socorro Main Canal. Once displaced, the fish are effectively denied reentry into the Rio Grande upstream from the dam or rerouted completely through agriculture canals. Concerns over these issues led the U.S. Fish and Wildlife Service, in its 2003 Biological Opinion, to direct the Bureau of Reclamation (BOR) to restore connectivity between the reaches upstream and downstream from San Acacia Dam (Bovee, 2008).

The middle Rio Grande, as described earlier in this publication, is not a stream in which natural flow regimes are presently functioning. In a highly managed, anthropocentric induced spring snowmelt, the minnow’s life cycle and the habitat in which it is supported is greatly altered from the conditions in which it evolved. Within a habitat study conducted in 2009 in the reaches where there is a population of minnow, several observations were made by Bovee, et. al. In the lower reaches, specifically from Isleta Dam and San Acacia Reach, H. amarus was most often found in areas of low or moderate water velocity (for example, eddies formed by debris piles, pools, and backwaters). The minnow was rarely found in habitats with high water velocities, like found in the main channel an often deep and swift conditions persist during water availability. Specifically, the juveniles, or Young-of-year (YOY) were virtually entirely found in shallow areas where low-to-no velocity water existed. A narrower criteria range of flows were observed at study sites along the Rio Salado, Rio Puerco and in the Sevilleta. Optimum flows ranged from 20 ft3/s and 40 ft3/s with habitat reductions beginning around 60 ft3/s. This sort of velocity condition occurs in secondary channels, main channel margins, backwaters, and around deposits of large woody debris (Bovee, 2008).

As illustrated in the figure below, for adults, a broader spectrum of velocity and habitats, including the main channel was observed. Suitable hydraulic habitat as found by Bovee et. al, reached maximum values at discharges of 40 ft3/s to 80 ft3/s. Optimum flow range for adult was considered to be at 60 ft3/s. At discharges larger than 150 ft3/s, the area of suitable hydraulic habitat for adults declines within the lower reach. Of note, is when loaded woody debris was factored into hydraulic habitat designations, the optimum flow range increased to a range of 60 ft3/s to 200 ft3/s with a peak habitat area achieved at 150 ft3/s. Minnow of both life stages showed preference for eddies formed by debris piles, particularly during winter where very low water flow conditions are the norm and as found in the lower reaches, loaded woody debris could be considered a preference factor (Bovee, 2008).

Table 1. H. AMARUS, OPTIMUM FLOW RANGES (Bovee, et. al, 2008).

To align the velocity issue with habitat fragmentation and connectivity concerns is the issue of species populations and life span alterations and genetic diversity. Since 2001, the U.S. Fish and Wildlife Service has rescued the minnow from intermittent and drying pools. This lack of connectivity throughout the Isleta and San Acacia reaches poses a situation in which an “incidental take” in terms of the Endangered Species Act requirements are concerned. A conceptual model from Cadwell, et. al, addresses the physical stressors that the fish undergo in fragmented, low-flow situations which in turn, could affect the conservation and protection of the species (Cadwell, et. al, 2009).


The USFWS, in order to minimize this risk to the endangered fish has implemented a rescue program during times of low-to-no flows when habitat fragmentation is created. A rescue protocol was developed and implemented by the Service in 2003 which required locating isolated pools, collection of fish, and transport by bucket, amphibious vehicle, or distribution truck to optimal habitat. As illustrated in the figure below since 2001 to 2007, approximately 550,000 RGSM have been rescued from isolated and receding pools and transported to perennial reaches of the middle Rio Grande (Cadwell, et. al, 2009).

In sum, the above discussion of the Rio Grande Silvery Minnow demonstrates the needs and issues of the species within the lower reaches of the Middld Rio Grande. Due to the lack of restoration projects within Cochiti reach, the goals of recovery will not be addressed within this section. For specific information the 2009 Recovery Implementation Plan (RIP) Final Draft Action Plan, the Middle Rio Grande Endangered Species Collaborative Program (MRGESCP) outlines elements, actions, and associated tasks specific to recovery of the species. Additional discussion is located further into this document in the "Restoration Proposals" section, addressing the needs of both ESA listed species.

Table 2. H. Amarus, USFWS rescues 2001-2007, (Cadwell, 2009).

Southwestern Willow Flycatcher - Jourdan Adair

Species Background and Overview

As its name suggests, the southwestern willow flycatcher (Empidonax traillii extimus) breeds throughout the southwestern United States and winters in Central America and northern South America (Finch et al., 2000). Pictured in Figure 4, a typical southwestern willow flycatcher (SWFL) is 5.75 inches in length and weighs only 0.42 ounces (USFWS, 1995). In accordance with the Endangered Species Act, the U.S. federal government officially listed the SWFL as an endangered species in 1995 (USFWS, 1995).
SWFL Pic.jpg
Figure 4: Southwestern Willow Flycatcher. From

Threats facing the southwestern willow flycatcher are largely due to anthropogenic influence on the landscape, such as the construction of dams. For the Cochiti reach, constrained between Cochiti Dam and Angostura Diversion Dam, this is most certainly the case. Currently, the SWFL is faced with the following threats (USFWS, 2002):
  • Removing, thinning, or destroying riparian vegetation
  • Water diversions and groundwater pumping which alter riparian vegetation
  • Overstocking or other mismanagement of livestock
  • Recreational development
  • Changes in flood and fire regimes due to dams and stream channelization
  • Establishment of invasive/non-native plants
  • Cowbird parasitism (inhibit reproductive success, therefore further reducing population levels)

In order to diagnose and treat the issues facing the southwestern willow flycatcher, it is necessary to first determine what is needed for SWFL survival. With regards to breeding habitat, features that are important to the SWFL include floodplain size and distance to water, riparian vegetation density and heterogeneity, rainfall and permanence of water bodies, vegetation serial stage and temporal stability, and size and location of riparian patches (Hatten et al., 2010). Characteristics of preferred habitat of the SWFL include dense riparian forests and shrub-like vegetation (Graf et al., 2002) associated with rivers and wetland areas such as lakes and reservoirs. Historically, the SWFL nested in willows, seep willow, boxelder, buttonbush, and cottonwood (FWS Protected Species). However, these native species are now being threatened by invasive species growing along the floodplain that is already suffering from separation from the main channel and therefore the absence of critical ecological flows.

During the nesting season (i.e., May to June), nesting sites are often found near surface water or saturated soil, indicating the need for overbank flooding and connectivity between the main channel and floodplain (Unitt). Because the breeding season is so late in the spring and among the shortest of North American songbirds, the SWFL breeding cycle is especially sensitive to any disruptions.

While its endangered status is relevant across the southwest, the SWFL has become especially important to the middle Rio Grande as both and indicator species and a regulatory mechanism for protecting the fragile middle Rio Grande ecosystem. And while there is no current critical habitat in the Cochiti reach, improvement of the ecological conditions within the reach is still crucial for the sake of restoring connectivity with the downstream reaches and for maintaining a healthy environment throughout the middle Rio Grande to support SWFL migration.

Restoration Goals

In its 2009 Recovery Implementation Plan (RIP) Final Draft Action Plan, the Middle Rio Grande Endangered Species Collaborative Program (MRGESCP) outlines elements, actions, and associated tasks specific to recovery of the southwestern willow flycatcher. The elements, which describe general species or Program needs, include Element 2.1: Flycatcher Territory Establishment and Nesting Success, Element 2.2: Flycatcher Research, Monitoring, and Adaptive Management, and Element 2.3: Flycatcher Populations Outside of the Program Boundaries. Each element includes a sub-section of action(s) that broadly explain the activities that need to be completed in order to meet those needs. According to the Program, the breakdown of each element and their respective actions is as follows:
    • Element 2.1: Flycatcher Teritory Establishment and Nesting Success
      • Action 2.1.1: Create necessary habitat for territory establishment and nesting and determine viability of SWFL populations
      • Action 2.1.2: Create necessary hydrologic conditions for territory establishment and nesting
    • Element 2.2: Flycatcher Research, Monitoring, and Adaptive Management
      • Action 2.2.1: Assess, identify, and prioritize science activities that address Program goals
      • Action 2.2.2: Develop and implement monitoring programs
    • Element 2.3: Flycatcher Populations Outside of the Program BoundariesFeasibility of application to Cochiti Reach
      • Action 2.3.1: Coordinate and share information with rangewide database of SWFL detections and territory locations (ex: SWFL forums)

The plan goes even further to suggest tasks that break down the specific steps needed to implement each aforementioned action. As mentioned in the RIP, necessary habitat and flow conditions for SWFL survival must be determined in order to establish appropriate restoration goals and actions. These goals must be specific to the reach to ensure utmost feasibility. However, it is important the restoration projects do not end at the implementation stage. While often more difficult in practice than in theory, there must be established measures of success of restoration, and most importantly, long-term sustainability of these measures.

Another approach to SWFL recovery is in the 2002 Final Recovery Plan for the Southwestern Willow Flycatcher, developed by the Southwestern Willow Flycatcher Recovery Team, Technical Subgroup. In a similar formatting to the MRGESCP RIP, the Recovery Plan explains nine actions needed to recover the SWFL: (1) Increase and improve currently suitable and potentially suitable habitat, (2) Increase metapopulation stability, (3) Improve demographic parameters, (4) Minimize threats to wintering and migration habitat, (5) Survey and monitor, (6) Conduct research, (7) Provide public education and outreach, (8) Assure implementation of laws, policies and agreements that benefit the flycatcher, and (9) Track recovery progress (USFWS 2002).

While several recovery plans have been developed to directly address the current condition of the southwestern willow flycatcher, there are still unanswered questions that need to be addressed. For example, the importance of duration of water on reproductive success, the relationship between food availability and water availability, and the relationship between water availability and the duration of water on survival (Copeland et al., 2009). If these fundamental gaps in knowledge can be closed, then perhaps a much more sustainable balance can be achieved between human usage and species survival.

Geomorphology - Josh Page

Figure 5. Changes in River Geomorphology of Cochiti Reach from 1918 - 2001 (Richard & Julien, 2003)

River geomorphology is the dynamic science connecting the relationships between hydrology, changes in the sediment regime, local geologic conditions, and anthropogenic alterations that all lead to physical changes to a river system (Schumm, 1977). There are a few geomorphic properties that greatly affect the quality of habitats for the silvery minnow and southwestern willow flycatcher. One is the shape of the river, which involves the cross sectional geometry and the flow patterns. If there is little interaction between the river and the floodplain, due to the shape of the river itself, healthy
habitats are less likely to be established. The gradient of the channel greatly affects sediment transport processes leading to changes in bed materials, turbidity, and flow regimes. The height of the channel bank causes issues with connecting to the floodplain at certain discharges. Because these processes develop at variable rates, with large flood events causing drastic changes that can take years to reach an equilibrium state, it is vital to have comprehension of the past and present conditions in order to successfully plan restoration strategies for the future.

Historic Content
In the late 1800s, a few decades after the Homestead Act, a significant change in the Rio Grande’s geomorphology was visible. An issue of sediment aggradation was occurring from arroyo cutting, overgrazing, and irrigation for agriculture (Harperet al., 1943). This lead to the river bed rising which caused major flood events in Albuquerque. This also caused a rise in the water table which destroyed agriculture lands according to Harper. Due to the high levels of sediment deposition, estimation of 75 billion pounds, major water projects were constructed in the 1920s to control the sediment loads in the Rio Grande (Scurlock, 1998). A large amount of dams were constructed in the few decades after the 1920s, but the one that greatly affected the Cochiti Reach of the Rio Grande wasn’t built until 1973. The Cochiti Dam was built as a flood and sediment control structure to prevent additional aggradation below the dam (Scurlock, 1998) (Richarge 2001).

Figure 6. Cross-section of the Rio Grande below Cochiti Dam from 1971-1998 (Richard & Julien, 2003)

The Cochiti Reach pre-1973 has a completely different geomorphology then after the construction of the dam. Before the Cochiti Dam the river bed was dominantly sand with grain sizes around 0.2mm; after the construction of the dam the river bed transformed into cobble and gravel with grain sizes around 20mm in just 10 years (Richard and Julien, 2003) (Salazar, 1998).
Graphical Changes.JPG
Figure 7. Summary of Geomorphic Changes of Rio Grande, Cochiti Reach from 1915-2005 (Richard & Julien, 2003)
The pre-dam mean suspended sediment load was 1,455 mg/L which decreased by 96% to 64 mg/L after the construction was complete (Richard, 2001). Aerial photographs indicate that the river itself was once wide and braided, but post-dam photos show how it changed into a single narrow channel, figure 5 (Salazar, 1998). According to study conducted by Mussetter Engineering, the mean width of the Cochiti Reach in 1917 was 968 ft compared to the drastic change of 314 ft in 1992. One study shows the change in elevation, due to incision, at a cross section below the Cochiti Dam, figure 6. CO-18 is a cross section from a set of Cochiti range lines created by the USGS in 1970. It is located just above the San Felipe Pueblo by the Arroyo Tonque (Richard, 2001). From this study it’s apparent that part of the main channel has dropped 2 meters in elevation due to the construction of the Cochiti Dam, just after a decade. The influence Cochiti Dam had on the Rio Grande is obvious. As a means to control aggredation and flooding, the dam itself was very useful. However it was almost too efficient, causing large amount of incision and a disconnection between the main channel and the floodplain. The dam has caused this part of the Rio Grande to move towards a more equilibrium state, which has negative impacts of ecological functions. The dynamic nature of a river system allows for lateral movement and bank flooding which creates healthy habitats. The current narrow and deep design of the channel has decreased peak flows. As a result of climate change, even with large strom events the flows needed to reconnect the channel to the floodplain are impossible due to restriction from the dam itself. Figure 7, is a graphical representation of these changes on an annual basis complied by a case study completed in 2005 (Richard et al., 2005). It shows a great summary of the changes in river morphology over the past 90 years. It’s important to note the drastic changes in the number of channels, the bed materials, and the sinuosity that occurred after the construction of the dam in 1973.

Hydrology - Kevin Scales
In many ways, the water manager of the Cochiti reach seeking to optimize species welfare and human demands for water has an easier job than those of the lower reaches. Cochiti typically has more flow available, being at higher elevation and in a cooler climate. There are fewer human demands for water in the reach, and the endangered species set consisting of the Silvery Minnow and Southwestern Willow Flycatcher have low populations in the region. The converse of this statement is that the Cochiti reach must be sensitive to the water demands for sections further downstream, as well as to constraints that exist in lower reaches. Thus, management of the Cochiti faces the fewest intrinsic requirements of any reach, but the greatest number of requirements by the other reaches.

The main issue at hand for water management at Cochiti dam is how to shape and time the releases of flood pulses. Flood pulses are essential for the spawning and feeding of the Silvery Minnow and the Southwestern Willow Flycatcher. They must have some depth characteristics and some duration to be ideal, though those exact quantities may be unknown. (See the relevant sections for the specifics for each species). The converse issue is that pulses increase channel incision, which is already a serious concern in Cochiti reach. Sufficiently large pulses may overbank downstream human habitats (in Albuquerque, for example). Also, the needs of agriculture dictate that some releases take place during the growing season, regardless of the species’ needs. Finally, Cochiti Lake is among the more efficient locations to store surface water along the Middle Rio Grande because of its lower evaporative losses compared to downstream sites. As an illustrative example, the pan evaporation coefficients at Cochiti peak at 12.95 inches in June, with an annual rate of 88.01 inches, while water stored at Elephant Butte is subject to a high of 16.37 inches in June, and 112.41 inches annually.[1] Thus, there is an incentive to store as much water as possible at Cochiti to avoid losses.

At present, large flood pulses do not emerge from the Cochiti Reservoir. The storage ability of the dam makes it possible to control otherwise large flood pulses that might pass through, and years of low flow have meant that the reservoir capacity has not been strained. A flow exceedance curve for the past seven years is shown below in Figure 8.
Figure 8: Flow Exceedance Curve for the Cochiti gage, Oct. 1, 2007 to Apr. 20, 2014[2]

The other significant change arising from the creation of Cochiti Dam is the impoundment of river sediments in the reservoir. Prior to the dam, the typical stretch of river was sandy, wide, and braided. Post-dam, the downstream Cochiti reach became gravelly, incised, and meandering. (See the geomorphology section for more on the channel bed.) Much of this can be attributed to the sediment collection within the reservoir. Richard and Julien present suspended sediment data for pre and post dam mean sediment concentrations and single day maxima; their results are reprinted below in Table 3.

Table 3: Comparison of Rio Grande suspended sediment concentration before and after closure of Cochiti Dam (November 1973). Pre and post dam values are the mean values for that time period. Average change is the difference between pre and post dam values expressed as a percentage of the pre-dam value.[3]

From the table, we can see that the placing of a reservoir does more than just retain water. While the residence time for liquid water may be measured in weeks or months, sediments that deposit in the reservoir tend to stay in place indefinitely. In fact, assuming maximum capacity of 900M m3 and and 50% exceedance flow rate of 750 cfs, the residence time for water in Cochiti Lake is ~490 days. If we use the long term mean of 1070 cfs, the water residence time becomes under one year (344 days). Yet, the sediment concentration below the dam has declined dramatically compared to the concentration in higher unregulated portions of the river. (The decline even above the dam is attributed to a shift from wet to dry conditions over the period included in the table.) Typically, 1000 acre-ft of sediment are captured annually.

Flood Pulses
Cochiti Dam can release up to 14790 cfs from the reservoir.[4] Physical theory states, however, that this pulse would not be what is observed any significant distance downstream. Friction and storage along the route suggests that the peak value of a flood wave will decline with propagation distance. The process of predicting a downstream hydrograph from a known upstream hydrograph is called flood routing. So long as the peak flood flow in Albuquerque or other important downstream sites does not exceed the capacity of whatever flood containment structures exist, a constraint on our pulse release will be met. It is unlikely that pulses as high as 14790 cfs will be required to prevent extinction of the Silvery Minnow or the Southwestern Willow Flycatcher, but it is useful to be aware of our constraints.

Typically, modeling of the flood pulse starts with a simple differential equation of the form:
This simple-looking equation states that the change in storage along the reach must equal the difference in the known input hydrograph (I) and the unknown output hydrograph (O). Whatever flows in must either flow out or be stored in the section of river; it is a form of the continuity equation.[5] In a flow section with uniform geometry, it might be possible to generate reasonably analytical results, but in a section of real river, considerable numerical approximation is required. The shape of the stream bed for any river is complicated, and more so for shallow, braided, slow-moving rivers like the Rio Grande. Further, the shape changes irregularly as the water level rises, so only brute-force computation over a detail topographic map can produce the detailed outcomes needed. Such as study is beyond the scope of this document. The approach, however, can be described.

A common method is the Muskingum method, which relates volume to inflow and outflow by the equation
where Kt is a travel time constant and MissingImage5.JPG is a weighting factor that indicates relative contributions of input and output to the hydrograph. A good first guess is that MissingImage6.JPG and Kt is the mean travel time, computed with reach length and mean velocity.

We can discretize the continuity equation
and rearrange
We can then discretize the Muskingum equation and plug it into the continuity equation. The result is
Then we simply step through the solution at the desired time steps.

Conceptually, this is not difficult to understand. In practice, the values of x and Kt will vary with sections. At the very least, the river will need to be broken down into sections without lateral inflow or outflow, as neither is considered in this method. In practice, historical observations are needed to know the values of x and Kt that are needed. Like much in river flow analysis, the future is predicted only once the past is documented.

A Method to Resolve Conflicting Goals**[6]**
In Cochiti or any other reach of the Rio Grande, we fully expect that the optimum flood pulse or other water release schemes will differ depending on our goals. The perfect pulse for spawning among the Silvery Minnow will not be the same as for habitat sustenance among the Southwestern Willow Flycatcher. They may be similar. The geomorphological needs of maintaining the channel against incision and bank erosion will likely call for entirely different pulse characteristics, if any at all, as erosion increases dramatically with flow. There will also be costs associated with any water release. These costs can be direct, such as the cost of paying a gate tender to travel onsite and open the gates. They can also be indirect, such as the costs associated with human use of the river (aquatic activities like fishing, boating, swimming, hiking in the Bosque, etc). There is, of course, also the costs of not having the water available for other uses later. The dichotomy between species preservation and agricultural uses is always present. In a realistic system, we would also need to consider downstream users, as any flood pulse release will travel downstream until it reaches a barrier (Elephant Butte perhaps), the ocean (unlikely), or it evaporates away (which is ongoing).

Thus, the welfare maximizing agent in control of hydrology for the reach needs to balance competing claims. One tool for this is to set up a constrained optimization problem using the method of Lagrange multipliers. It allows us to maximize some objective, which we must define, subject to various inviolable constraints. Mathematically, we set up the problem thusly:


Our g1 … gm are constraints on the solution sets. Here:


Here, W(P,D) is a welfare function, or objective function, a measure of the overall benefit that comes from a flood pulse of peak amplitude P and duration D. It is presumed that the overall shape of the pulse is known in advance (square pulse, bell curve, etc), and if a more complicated pulse characterization with more variables is required, more variables may be specified beyond just P and D. The objective is sum of the various individual welfare functions for the Silvery Minnow (F), the Flycatcher (B), the stream geomorphology (G) and minus the costs (C). The constraints are that the pulse amplitude and duration each have a peak beyond which they must not go, P0 and D0. This could be because of overbanking in a populated area, damage to bridges or other infrastructure, or be based on the needs of the species in the river. If no such constraint actually exists, then that line may be left out of the formulation. We also face physical constraints that the pulse amplitude and duration be intrinsically positive numbers. If the math comes out that a negative pulse is needed, we need a way to prevent that answer from arising. In practice, this may not be an issue, but it is wise to plan for it. Finally, the volume of water available is limited. The final constraint shows this, with the product of peak and duration, scaled by some parameter k that is based on the pulse shape, be less than the maximum volume we either have in Cochiti Lake or less than the maximum we are willing to release. Here, k would equal 1.0 for a square pulse. Its value for other shape would depend on the definition of duration (width of the base, width at root-mean-square value, etc).

The method of solution is to use the Lagrange multiplier method. Set up:


The λi and μi are called Lagrange multipliers. (The choice of letters is based on the type of constraint, zero or non-zero, and is unimportant.) We then take the partial derivatives of MissingImage14.JPG with respect to P and D and set equal to zero.


We have the appearance of two equations with seven unknowns, but recall that we still have the five constraints given, so this provides two equations, five inequalities, and seven unknowns, which is sufficient to solve the system. The inequalities are resolved by setting up complementary slackness conditions. Formally:

So for our system:
The solution is finally determined by a series of cases, which is where the actual analysis can become tedious. It is possible that any, all, or none of the constraints is active. A nice interior solution would provide P and D values that are well away from the constraint values, so none of them would even be relevant. However, after a lengthy extreme drought, the mathematics might try to state that the ideal is to release negative amounts of water, essentially putting water into the system from out of nowhere. Or the species might thrive on too much water over time or at once or both for us to provide. Unfortunately, it may be necessary to go through each combination of constraints being active or inactive (ignoring obvious contradictions such as P>P0 and P<0 simultaneously) and then checking each solution candidate to find out which one really is the welfare maximizer of all possible options.

The shortcomings of this approach are not hard to discover. We need, in precise mathematical formulation, a welfare function for the Silvery Minnow, the Flycatcher, and the geomorphological concerns. The expressions need not be accurate (although the more so the better), but they must be expressible in mathematical terms, using some unit. Among environmental economists, it is typical to try to assign a dollar value to the costs and benefits. For direct or indirect costs, this is natural. To put a value on the Silvery Minnow spawning season, this may be considerably more difficult, particularly if we are trying to evaluate restoration for the fish and birds intrinsically, and not just for our human valuation of their benefit to us.

Hydrology References

[1] Western Resource Climate Center,, accessed 4/20/14.
[2] USGS Data,, accessed 4/20/14
[3] (Richard & Julien, 2003)
[5] This section follows the discussion in Hornberger, George M. et al, Elements of Physical Hydrology, The Johns Hopkins University Press, 1998, ch5.
[6] The mathematical discussion follows sections 3.3 and 3.5 in Sydsæter, Knut et al, Further Mathematics for Economic Analysis, Prentice Hall, Gosport, 2008.

Restoration Proposals

Realistic Restoration Proposal

The southwestern willow flycatcher and Rio Grande silvery minnow are rarely found in the Cochiti reach of the middle Rio Grande due to high incision in the main channel and the replacement of native habitat with invasive species like the salt cedar and Russian olive. While there may be limited hope for reintroduction of these two species directly downstream of Cochiti, the reach has the unique opportunity of having the ability to alter flows at Cochiti Dam in fulfilling the ecological and hydrological needs of the two endangered species throughout the lower reaches.

For the purpose of riparian vegetation recruitment and creating an ideal spawning environment in the reaches downstream of Cochiti, we propose following the proposed contingency plan for the implementation of a temporary deviation from the U.S. Army Corps of Engineers’ operation of Cochiti Lake and Jemez Canyon Dam. This planned deviation would occur from late February through June in order to achieve two primary goals: (1) facilitate spawning and recruitment flows for RGSM and (2) provide seasonal over bank flooding opportunities to create ideal habitat for SWFL. The plan includes two proposed scenarios, one for the purpose of recruitment and the other for the purpose of overbanking to meet the needs of both the RGSM and SWFL.

The first proposed action is for a temporary storage pool of between 5,000 and 20,000 acre-ft at Cochiti Lake and a subsequent release of the stored water to provide a minimum spawning and recruitment flow at the Albuquerque gage of 3,000 cfs for seven days. The second proposed action is for a temporary combined storage pool of 45,000 acre-ft at Jemez Canyon Dam and Cochiti Lake and a subsequent release of an amount of the stored water necessary to provide a minimum flow at the Albuquerque gage of 5,800 cfs for five days. For both plans, it should be taken into account that the channel is confined to a maximum flow capacity of 7,000 cfs released from Cochiti. In terms of the runoff hydrograph, the water for either proposed action would be stored on the ascending limb (i.e., when native flows exceed downstream demands) and would be released at the peak and descending limb.

Idealistic Restoration Proposal

There is no lack in restoration project types across the breadth of endangered species and river restoration fields. The species recovery attempts over the last several decades within the Middle Grande have utilized many techniques, approaches, experiments and theories in regards to the Rio Grande Silvery Minnow and Southwest Willow Flycatcher and the hydrologic realities faced within New Mexico. Several of these techniques can be found within the attached list of Recommended Actions found at the end of this document detailing hard restoration protocols like inducing channel modifications, stabilizing banks, flow manipulations, etc. The combinations of restoration actions addressed by the multitude of recovery plans as listed and discussed in previous sections, lead towards the need to reign in the associated costs and may be benefited through thinking outside-the-box. Beyond strategies found within the realistic proposal section, other less direct forms of action could be utilized.

Within the Sacramento River, located in the Central Valley of northern California, a similar recurring set of endangered species challenges exists. Fish, bird, plant and other wildlife listings in this California watershed and the existing demands on the river have spurred a multitude of restoration proposals including a conservation easement land acquisition proposal by The Nature Conservancy (TNC). Through focusing on floodplain connectivity of two tributaries, the Big Chico and Mud Creek, to the Sacramento, "soft" restoration is an interesting case study in which the MRG issues could be handled. The restoration plan prepared by the TNC is a secondary phase of a four phase restoration plan of a small section of river. The restoration goals address issues of agricultural land out of production or under threat to flooding and erosion and seizes on an opportunity to redesign the landscape. Bordered by state park lands, three parcels located within the delta of the three variable sized streams was targeted. Although this unique set of conditions on the Sacramento and ideal property structures do not directly apply to the Cochiti Reach section of the Middle RIo Grande viewing ESA restoration projects through other case studies across the West should be considered (The Nature Conservancy, 2002).

Floodplain connectivity across the reach, as described by the California project could focus on tributaries. The Jemez River, Santa Fe River, Galisteo & Borrego Canyons and Arroyo Tonque are the main inflows to the Cochit reach system. As the majority of these tributaries are currently sediment sourced altered by human structures the opportunity to change management practices prevails. Addressing these tributaries influence on the river could be a great stepping stone if river restoration within this upper reach is something society deems as a worthy investment of human capital and the political and social constraints that would be necessary for such action.

No single plan can likely be said to perfectly restore the Cochiti Reach. The issues are too vague, and the difficulties are numerous. Regarding the endangered Silvery Minnow and Willow Flycatcher, we are inclined to agree with the assertion that other reaches provide a better use of resources. However, should restoration efforts lead to any kind of significant reintroduction of the two species into the Cochiti Reach, they would serve well as indicator species and a means of determining the health of the river.

The best statement we can make at this time is that efforts to fix the geomorphic problems in the reach are mostly likely to lead to the conditions that will promote species reintroduction. An incised, rocky stream is non-ideal for either species, and is indicative of excessive sediment removal by the Cochiti Dam. Efforts sufficient to connect the stream back to its flood plains could lead to the restoration of eddies, pools, and overbanked areas that promote living conditions conducive to the minnow and flycatcher. Until that time, we should focus on maximizing our realistic returns on our investment. We've seen general-purpose methods for choosing plans of action among many alternatives. These are only first steps among many towards optimum river management, which remains a feasible goal.

Appendix A: Recommended Actions

Most likely methods for the Cochiti Reach (from USBR Joint Biological Assessment, Part II - Table 3):

1) Infrastructure relocation or setback
2) Channel modification
  • Channel relocation using pilot channels or pilot cuts
  • Island and bank clearing and destabilization
  • Bank line embayment
  • Side Channels (high flow, perennial, and oxbow re-establishment)
  • Longitudinal bank lowering or compound channels
  • Jetty/snag removal
3) Bank Protection/Stabilization
  • Longitudinal features
    • Riprap revetments
    • Other type of revetments
    • Longitudinal stone toe with bioengineering
    • Trench filled riprap
    • Riprap windrow
    • Deformable stone toe/bioengineering and bank lowering
    • Riparian vegetation establishment
4) Transverse features or flow deflection techniques
  • Bendway weirs
  • Spur dikes
  • Vanes or barbs
  • J-hooks
  • Trench filled bendway weirs
  • Boulder groupings
  • Rootwads
  • Large woody debris
5) Cross channel (river spanning) features
  • Grade control
    • Deformable riffles
    • Rock sills
    • Riprap grade control (with or without seepage)
    • Gradient restoration facility (GRF)
    • Low head stone weirs (loose rock)
  • Conservation easement

References pertaining to Joint Biological Assessment, Part II: Most Likely Strategies and Methods by Reach




Allan G Harper, Kalervo Oberg, and Andrew R Cordova. 1943. Man and Resources in the Middle Rio Grande Valley. University of New Mexico Press, Albuquerque.

Bovee, K.D., Waddle, T.J., & Spears, J.M. 2008. Streamflow and Endangered Species Habitat in the Lower Isleta Reach of the Middle Rio Grande. Open File Report 2008-1323,U.S. Geologic Survey, U.S. Department of the Interior.

Cadwell, Colleen A., Cho, SungJim & Remshardt, W.J. 2009. Effects of propagation, augmentation, and salvage on recovery and survival of Rio Grande silvery minnow (Hypognathus amarus), Final Report. U.S. Bureau of the Middle Rio Grade Endangered Species Collaborative Program.

Copeland, S. L., R. D. Beane, C.E. Paradzick, and C. L. Sommers. 2009. Literature review on water needs of the Southwestern Willow Flycatcher. Submitted to U.S. Bureau of Reclamation. Albuquerque, NM, by ERO Resources Corporation, Denver, CO. 57 pp.

Drost, C. A., Paxton, E. H., Sogge, M. K., & Whitfield, M. J. 2001. Food habits of the endangered southwestern willow flycatcher. Final Report to US Bureau of Reclamation, Salt Lake City, Utah.

Finch, D.M., J.F. Kelley, & J.E. Catron. 2000. Migration and winter habitat. p. 71-81. In: D.M. Finch and S.H. Stoleson (ed.) Status, ecology, and conservation of the southwestern willow flycatcher. Gen. Tech. Rep. RMRS-GTR-60. USDA Forest Service, Rocky Mtn. Res. Stat., Ogden, UT.

Graf, W. L., Stromberg, J., & Valentine, B. 2002. Rivers, dams, and willow flycatchers: a summary of their science and policy connections. Geomorphology, 47(2), 169-188.

Hatten, J. R., Paxton, E. H., & Sogge, M. K. 2010. Modeling the dynamic habitat and breeding population of Southwestern Willow Flycatcher. Ecological Modelling, 221(13), 1674-1686.

Julien, P., Richard, G., & Albert, J. 2005. Stream restoration and environmental river mechanics. International Journal of River Basin Management, 3(3), 191-202.

Middle Rio Grande Endangered Species Collaborative Program (Program). 2013. Middle Rio Grande Endangered Species Collaborative Program Recovery Implementation Program Final Draft Action Plan. 45 pp. (Accessed at:

Richard, G. 2001. Quantification and Prediction of Lateral Channel Adjustments Downstream from Cochiti Dam, Rio Grande, NM. Dissertation for Degree of Doctor of Philosophy. Colorado State University, Fort Collins, CO.

Richard, G. and Julien, P. 2003. Dam Impacts on and Restoration of an Alluvial River – Rio Grande, New Mexico. International Journal of Sediment Research, Vol. 18, No. 2, pp89-96.

Richard, G., Julien, P., & Baird, D. 2005. Case Study: Modeling the Lateral Mobility of the Rio Grande below Cochiti Dam, New Mexico.Journal of Hydraulic Engineering © ASCE.

Salazr,C. 1998. Morphology of the Middle Rio Grande from Cochiti to Bernalillo Bridge, New Mexico. Thesis for Degree of Masters of Science. Colorado State University, Fort Collins, CO.

Schumm, S.A. 1977. The Fluvial System. John Wiley and Sons, New York.

Scurlock, D. 1998. From the Rio to the Sierra: An Environmental History of the Middle Rio Grande Basin. General Technical Report RMRS-GTR-5. Fort Collins, CO: US Department of Agriculture, Forest Service, Rocky Mountain Research Station. 440 p.

"Southwestern Willow Flycatcher." Nevada Fish and Wildlife Office. U.S. Fish and Wildlife Service, 16 Apr. 2014. Web. 4 May 2014. <>.

Tetra Tech EM Inc. 2004. "Habitat Restoration Plan for the Middle Rio Grande, Middle Rio Grande Endangered Species Act Collaboration Program", Habitat Restoration Subcommittee.

The Nature Conservancy, 2002. "Restoration of the Confluence Area of the Sacramento River, Big Chico and Mud Creeks," CALFED grant application.

Unitt, P. 1987. Southwestern Willow Flycatcher.

U.S. Fish and Wildlife Service (USFWS). 1995. Endangered and threatened wildlife and plants: final rule to list the southwestern willow flycatcher (Empidonax traillii extimus) as endangered with proposed critical habitat. 60 Federal Register 10694-10715.

U.S. Fish and Wildlife Service (USFWS). 2002. Southwestern Willow Flycatcher Recovery Plan. Albuquerque, New Mexico. i-ix + 210 pp., Appendices A-O.

Yong, W., and D. Finch. 1997. Migration of the Willow Flycatcher along the Middle Rio Grande. Wilson Bulletin 109:253-268.