HYDROECOLOGICAL CHARACTERIZATION OF ARROYO ALAMAR,
TIJUANA, BAJA CALIFORNIA, MEXICO

DRAFT FINAL REPORT

January 19, 2004

Victor M. Ponce, Ana Elena Espinoza, Jose Delgadillo, Alberto Castro, and Ricardo Celis

1. INTRODUCTION
The H. Ayuntamiento de Tijuana (Municipality of Tijuana) has among its current projects the rehabilitation of Arroyo Alamar, a tributary of the Tijuana river. The project will satisfy a host of urban-planning needs, such as the preservation of riparian areas, flood management, planned land use, recreation, landscaping, a green corridor, replenishment of groundwater, improvement of water quality, and compliance with federal stream zoning restrictions.
The project encompasses the 10-km reach of Arroyo Alamar, located between the bridge on the toll road to Tecate, and the channelized reach near the confluence with the Tijuana river. The objective of the project is to rehabilitate the Arroyo Alamar and its flood plain to encourage planned land use and preserve primary hydroecological functions.
The Municipality of Tijuana has developed a preliminary land-use plan which includes a diversity of uses such as agricultural, industrial, urban, recreation, and flood prevention and mitigation. A hydrological study to determine design flood magnitudes has been completed by the Principal Investigator Dr. Victor M. Ponce, in a FY2000 SCERP-funded project. For return periods of 2, 5, 10, 25, 50, 100, 200, 500, and 1000 years, the flood discharges are: 280, 530, 680, 930, 1140, 1310, 1420, 1600, and 1720 m³ s^-1, respectively.
This report continues the hydrological study to its next logical step, i.e., the development of a hydroecological characterization aimed at determining flood levels that are congruent with the proposed land uses. The work includes close consultation with cognizant federal agencies to determine applicable stream zoning restrictions.
Tijuana's city planners envision the rehabilitation of Arroyo Alamar to have the essential character of a green corridor, with multiple land uses in tune with the primary flood-mitigation, aquifer-replenishment, and riparian-habitat functions. In a city with very few large tracts of greenery, the hydroecological rehabilitation of Arroyo Alamar is a highly desirable project. The concrete-channel alternative used in the development of the First, Second, and Third Phases of the Rio Tijuana is no longer a viable alternative, given its marked negative impacts of the landscape and environment. Thus, this project assists the Municipality of Tijuana in furthering their goals of rehabilitating the Arroyo Alamar with an ecologically sound approach.

2. RESEARCH OBJECTIVES
The specific objectives of this project are the following:

To determine the channel properties that are compatible with primary hydroecological functions. These include the cross-sectional geometry, longitudinal slopes, bank protection, and percolation to ground water. The study encompasses the preservation and enhancement of existing riparian corridors, the rehabilitation of degraded riparian areas where warranted, and the selection of alternative land uses.

The determination of the flood frequencies to be implemented in the characterization, and the hydraulic design of the hydroecological channel to convey the selected flood flows.

3. RESEARCH METHODOLOGY / APPROACHES
The research method/approach consists of the following steps:

Assemble land-use information, including existing and planned developments, riparian, agriculture, industry, recreation, tourism, and other areas.

Consultation meetings with officials of the Comisión Nacional del Agua, in Mexicali and Mexico City, to establish appropriate flood frequencies to be used in the design.

Hydrological and ecological design of the rehabilitated channel and its flood plain. This includes the modeling of flood flows using the standard U.S. Army Corps of Engineers HEC-RAS (Hydrologic Engineering Center - River Analysis System) model, Version 3.0. The ecological design includes the selection and establishment of riparian, agriculture, industry, recreation, and other multi-purpose areas within the project site.

To establish the need, where appropriate, to stabilize the stream channel by means of bank protection, grade control, and other suitable means, with the objetive of sustaining the design flood flows.

Fig. 1 Project site along the northeastern portion of the city of Tijuana; outside (solid) lines delimit project area; inside lines delimit federal zone.

Preliminary appraisal based on Mexican hydrological practice indicates that the federal frequency should be at least 10 yr, and the project design frequency should be 1000 yr. The characterization encompasses the entire length of the Arroyo Alamar rehabilitation project, from its upstream end at the bridge over the Tecate toll road to its downstream end about 10 km downstream, to connect with the Second Phase of the Rio Tijuana. This includes three distinct sections:

From the bridge at the Tecate toll road to the bridge at Boulevard Héctor Terán Terán.
From the bridge at Boulevard Teran Teran to the bridge at Boulevard Manuel J. Clouthier.
From the bridge at Boulevard Clouthier to the end of the Second Phase of the Rio Tijuana.

Several site visits were carried out, with the objective of ascertaining the parameters, including riparian areas, friction coefficients, land use, and other influences on the characterization.

The hydraulic modeling was performed using 500 cross sections, each every 20 m. This was necessary to ensure stability and accuracy of the backwater computation, enabling a precise delineation of the design flood stages.

The hydroecological characterization is based on the principle of mixed use of the stream channel, including riparian, agriculture, light industry, recreation, tourism, and water quality management. The cross-sectional design reflects the mixed uses and different flooding risks associated with those uses. A two-frequency compound channel is envisioned, with federal and soft-use zones. The federal zone is left to convey the regulation flood (10-yr frequency); the soft-use zone houses riparian, recreation, and ecotourism zones (1000-yr frequency).

4. PROBLEMS / ISSUES ENCOUNTERED

We used the 5-m topographic information to develop the hydroecological characterization. Detailed topographic information to 1-m resolution, which is required for final design, remains to be developed. The hydroecological design is to be taken as the level of feasibility study, pending more detailed studies when the detailed topographic information is developed.

5. RESEARCH FINDINGS

5.1 Gabion-lined channel systems

To remain stable and satisfy its intended use, a hydroecological channel requires some type of bank protection. Gabions are usually considered as a compromise between concrete lining, riprap or natural vegetation.

A gabion system is wire-enclosed riprap consisting of mats or baskets fabricated with wire mesh, filled with small riprap, and anchored to a slope.

Fig. 2 Layout and dimensions of gabion boxes and mattresses.

Wrapping the riprap enables the use of smaller stone sizes for the same resistance to displacement by water energy. This is a particular advantage when constructing rock lining in areas of difficult access. The wire basket also allows steeper (up to vertical) channel linings to be constructed from commercially available wire units or from wire-fencing material.
Due to their high shear strength, gabion systems provide a highly effective way to control erosion in streams, rivers and canals. They are normally designed to sustain channel velocities of 15 fps or higher. Gabions are constructed by individual units that vary in length from 6 ft to nearly 100 ft (gabion mats); therefore, applications can range anywhere from small ditches to large canals.
Gabion channels are a compromise between riprap and concrete. When the same-size rocks are used in gabions and riprap, the acceptable velocity for gabions is at least 3-4 times that of riprap. Unlike concrete, gabions can be vegetated to blend into the natural landscape.

Fig. 3 (a) Example of vegetated gabion channel.

Fig. 3 (b) Example of vegetated gabion channel.

Fig. 3 (c) Example of vegetated gabion channel.

Gabion channels with vegetation have the following advantages:

They allow infiltration and exfiltration.
They filter out contaminants.
They are more flexible than paved channels (Fig. 4).
They provide greater energy dissipation than concrete channels (Fig. 5).
They improve habitat for flora and fauna.
They are more aesthetically pleasing.
They have lower cost to install, although some maintenance is required.

Fig. 4 (a) Failure of paved channels due to lack of flexibility.

Fig. 4 (b) Failure of paved channels due to lack of flexibility.

Fig. 4 (c) Failure of paved channels due to lack of flexibility.

Fig. 5 Gabion-lined spillway provides energy dissipation.

The Manning's n or roughness coefficient for gabion channels with vegetation depends primarily on the type of vegetation and the size of the stones being used. There is no specific formula for the roughness coefficient for gabion channels with vegetation. The roughness coefficient can be derived by knowing other values of Manning's n and relating them to the specific case. For gabions without vegetation, Manning's n ranges from 0.025 to 0.03. Manning's n for vegetated channels also vary depending on the type of soil, the amount of cover, resistance and retardance. Manning's n for vegetative channels is given by the following formula:

(Eq. 1)

in which R = hydraulic radius; C = retardance coefficient, depending on cover and condition; and S = energy slope (m/m). The retardance coefficient ranges from C = 15.8 for class A vegetation to C = 37.7 for class E vegetation, as shown in Table 1.

Table 1. Classification of vegetative cover depending on degree of retardance.
Retardance Class	Cover	Condition	C Value
A	Weeping lovegrass	Excellent stand, tall, avg.	15.8
A	Yellow bluestem Ischawmum	Excellent stand, tall, avg.	15.8
B	Kudzu	Very dense growth, uncut	23.0
	Bermuda grass	Good Stand, tall, avg.
	Native grass mixture (mixture of bluestems)	Good stand, unmowed
	Weeping lovegrass	Good stand, tall,avg.
	Lespedeza sericea	Good stand, not woody, tall, avg.
	Alfalfa	Good stand, uncut, avg.
	Weeping lovegrass	Good stand, unmoved, avg.
	Kudzu	Dense growth, uncut
	Blue gramma	Good stand, uncut, avg.
C	Crabgrass	Fair stand, uncut, avg.	30.2
	Bermuda grass	Good stand mowed, avg.
	Common lespedeza	Good stand, uncut, avg.
	Grass-legume summer mixture	Good stand, uncut, avg.
	Centipedegrass	Very dense cover, avg.
	Kentucky bluegrass	Good stand, headed, avg.
D	Bermuda grass	Good stand	34.6
	Common lespedeza	Excellent stand, uncut, avg.
	Buffalo grass	Good stand, uncut, avg.
	Grass-legume fall mixture	Good stand, uncut 10 to 13 cm
	Lespedeza sericea	Cut to 5-cm height
E	Bermuda grass	Good stand, cut to 4cm	37.7
E	Bermuda grass	Burned stubble	37.7

There is no specific formula for Manning's n for channels with concrete, gabion or riprap lining. Table 2 shows frequently used Manning's n.

Table 2. Values of Manning's n for selected lining materials
Concrete: Trowel finish	0.012-0.014
Concrete: Float finish	0.013-0.017
Gunite	0.016-0.022
Flagstone	0.020-0.025
Gabions	0.025-0.030
Riprap	0.040-0.070

Riprap channels have a very large range of Manning's n, depending on the stone diameter and the flow depth. The Manning's n of the composite channel is given by the following formula:

(Eq. 2)

in which n_l , n_b , n_r , n_c= Manning's n of the left side slope, bottom, right side slope and composite channel, respectively; P_l , P_b , P_r , P_c= wetted perimeter of the left side slope, bottom, right side slope, and composite channel, respectively.

In gabion channels with vegetation (Fig. 6), the value of Manning's n is estimated by experience.

Fig. 6 A gabion channel with vegetation
(Source: Maccaferri, Inc).

The procedure for placing and filling gabions can be summarized as follows:

Assembling the individual units before placing (Fig. 7);
Placing them and wiring them together and filling the units with rocks (Fig. 8); and
Closing and wiring down the lids (Fig. 9).

Fig. 7 Assembling the individual units before placing.

Fig. 8 Filling the units with rocks.

Fig. 9 Closure of gabions using single lids or mesh rolls.

Gabions can be placed on dry bank (Fig. 10) or under water (Fig. 11). Linings laid in dry conditions are placed directly on a stable slope which is not too steep as to cause the revetment to slide. The units are normally laid down on the slope of the bank, at right angles to the current. However, the units on the bed itself should be laid in the direction of flow.

Fig. 10 Placing gabions on dry banks.

Fig. 11 Placing gabions under water.

When constructing a lining under water, dumping riprap is challenged by the many uncertainties, since it is difficult to obtain a uniform distribution of the material over the whole area to be protected. In order to reduce the risk, the amount of rocks dumped has to be increased by 50%. This problem does not arise with gabions, since the structure is preassembled and of fixed thickness. The gabions are placed using cranes or pontoons (Fig. 12).

The stability of gabion linings depends on the strength of the mesh, the thickness of the lining and the grading of the stone fill. Once the water velocity is known, these parameters can be selected. For longevity of the revetment, the mesh must be protected from corrosion. Gabions are constructed from wires with heavy zinc content and PVC coating.

Fig. 12 Using a pontoon to place the gabion mattress.

The two primary elements in channel design are cross-sectional shape and lining (Fig. 13). Lining is determined by erosion resistance and drainage requirements. Vegetated linings are appropiate for low velocities. Paved linings may be used for soil-water interfaces where the soil and groundwater conditions are such that the soil may erode under the design flow. Generally, velocities over 1.5 m/s require lined waterways.

Fig. 13 Usual types of channel linings in gabions and gabion mats.

For design purposes, uniform flow conditions are usually assumed, with friction (or energy) slope (S_f ) equal to average bed slope (S_o). This allows the flow conditions to be defined by a uniform flow equation such as Manning's. Supercritical flow creates surface waves of depth comparable to the flow depth. For very steep channel gradients, the flow may splash and surge; therefore special considerations for freeboard are required.

Once the basic hydraulic and physical characteristics of the channel have been determined, the selection of the lining and stability analysis can be made. Generally, a 10-yr to a 100-yr storm (1000-yr storm in Mexico) is used to determine capacity, while the 2-yr storm is used for channel stability. After vegetation is fully developed, the channel is considered stable and capacity becomes more critical.

Channels should be designed so that the flow velocity does not exceed the permissible velocity for the type of lining used. It is also important to check outlets for stability. Excessive velocities or grade changes may require protective or stabilizing structures, transition sections, or energy dissipators to prevent erosion or scour.

5.2 Hydrological design of Arroyo Alamar

To perform the hydrological design of the rehabilitated channel includes the determination of flood discharges, for selected design frequencies, and the calculation of water-surface profiles. The U.S. Army Corps of Engineers HEC-RAS (Hydraulic Engineering Center - River Analysis System) model was used to calculate the water-surface profiles.

A hydrological study to determine design flood discharges for 2-yr to 1,000-yr frequencies has been performed by Ponce (2000). The results are shown in Table 3.

Table 3. Arroyo Alamar flood discharges calculated using mathematical modeling.
Return period (yr)	Flood discharge Q (m³/s)
2	280
5	530
10	680
25	930
50	1,140
100	1,310
200	1,420
500	1,600
1,000	1,720

The 1,000-yr frequency is used by the Comision Nacional del Agua to design flood control channels in urban environments. To determine the freeboard, there is a need to go beyond the 1,000-yr frequency. According to USDA Natural Resources Conservation Service (NRCS) methodology, for urban flood control projects such as the Arroyo Alamar rehabilitation, the freeboard hydrograph should be designed to safely pass the Probable Maximum Precipitation (PMP). There is no PMP determination for Mexico. In places where it does not exist, the PMP is commonly approximated as the 10,000-yr frequency.

With the 2-yr to 1000-yr frequency floods determined by mathematical modeling, the Gumbel method was used to calculate the 5,000-yr and 10,000-yr flood discharges. Fig. 14 shows the Gumbel fitting and Table 4 shows the estimated discharges.

Fig. 14 Gumbel fitting for Arroyo Alamar 5,000-yr and 10,000-yr flood discharges.

Table 4. Estimated discharges using the Gumbel method.
Return period (yr)	Gumbel variate y	Flood discharge Q (m³/s)
5,000	8.52	2,140
10,000	9.21	2,290

To perform the HEC-RAS modeling, the design channel alignment was obtained from Tijuana officials. The project has a total channel length of 9880.849 m. The upstream point, with invert elevation 80 m, is at the bridge over Cañon del Padre, on the toll road to Tecate (Fig. 15). The downstream point, with invert elevation 40 m, is at the beginning of the concrete-lined reach of Arroyo Alamar, near the confluence with the Tijuana river (Fig. 16). This provides an average channel slope of 0.004048.

Fig. 15 Upstream end of Arroyo Alamar rehabilitation project, at Puente Cañon del Padre.

Fig. 16 Downstream end of Arroyo Alamar rehabilitation project, at the beginning of the
concrete-lined reach, near the confluence with the Tijuana river.

The project reach, of length 9880.849 m, was subdivided into reaches at an equidistance of 20 m, for a total of 494 reaches and 495 cross sections. The small reach interval (20 m) was adopted to ensure the accuracy of the water-surface profile computation. River stations are numbered from 001 to 495.

A prismatic channel of a chosen cross-sectional geometry was adopted for design. The channel consists of a main channel and left and right overbank channels (flood plains). The bottom width of the main channel is 40 m, and the main channel depth is 3.8 m. The side slopes of the main channel are 2 horizontal to 1 vertical.

The overbank channels are 20 m wide each, with channel depth 3.0 m. and side slopes 2 H:1 V. The overbank channels drain into the main channel with a 1% transversal slope. Figure 17 shows the channel design.

Fig. 17 Arroyo Alamar channel design.

In HEC-RAS, cross-sectional data is entered from left to right, looking in the downstream direction. The left channel bottom x-coordinate was specified as 100 m, and the corresponding cross-sectional coordinates were calculated using a spreadsheet. The x-coordinate left overbank limit is 92.4 m, and the x-coordinate right overbank limit is 147.6 m.

Figure 18 shows typical cross sections generated by HEC-RAS for the discharge of 550 m³/sec.

Fig. 18 (a) Result generated by HEC-RAS for the flood discharge of 550 m³/sec at cross section 001.

Fig. 18 (b) Result generated by HEC-RAS for the flood discharge of 550 m³/sec at cross section 495.

Several types of loss coefficients are utilized by HEC-RAS to evaluate energy losses. These are the Manning's n for friction (boundary) loss, contraction and expansion coefficients to evaluate transition loss, and bridge and culvert loss coefficients. At the present level of approximation, all secondary energy-loss coefficients have been neglected.

Appropriate values of Manning's n are significant to the accuracy of the calculated water surface profiles. The Manning's n value depends on a number of factors, including surface roughness, amount and type of vegetation, channel irregularities, channel alignment, scour and deposition, presence of obstructions, size and shape of channel, stage and discharge, seasonal changes, temperature, and bed material load.

Manning's n values for inbank channel flow was estimated at 0.035, and for overbank flow at 0.075. These values are consistent with established practice. The values of Manning's n recommended by HEC-RAS for natural streams are shown in Table 5.

Table 5. Manning's n values recommended by HEC-RAS for natural streams.
Type of channel and description	Minimum	Normal	Maximum
1. Main channels
a. Clean, straight, full stage, no rifts or deep pools	0.025	0.030	0.033
b. Same as above, but more stones and weeds	0.030	0.035	0.040
c. Clean, winding, some pools and shoals	0.033	0.040	0.045
d. Same as above, but some weeds and stones	0.035	0.045	0.050
e. Same as above, lower stages, more ineffective slopes and sections	0.040	0.048	0.055
f. Same as "d" but more stones	0.045	0.050	0.060
g. Sluggish reaches, weedy, deep pools	0.050	0.070	0.080
h. Very weedy reaches, deep pools, or floodways with heavy stands of timber and brush	0.070	0.100	0.150
2. Flood Plains
a. Pasture no brush
1. Short grass	0.025	0.030	0.035
2. High grass	0.030	0.035	0.050
b. Cultivated areas
1. No crop	0.020	0.030	0.040
2. Mature row crops	0.025	0.035	0.045
3. Mature field crops	0.030	0.040	0.050
c. Brush
1. Scattered brush, heavy weeds	0.035	0.050	0.070
2. Light brush and trees, in winter	0.035	0.050	0.060
3. Light brush and trees, in summer	0.040	0.060	0.080
4. Medium dense brush, in winter	0.045	0.070	0.110
5. Medium dense brush, in summer	0.070	0.100	0.160
d. Trees
1. Cleared land with tree stumps, no sprouts	0.030	0.040	0.050
2. Same as above but heavy sprouts	0.050	0.060	0.080
3. Heavy stand of timber, few down trees, little undergrowth, flood stage below branches	0.080	0.100	0.120
4. Same as above but with flood stage reaching branches	0.100	0.120	0.160
5. Dense willows, summer straight	0.110	0.150	0.200
3. Mountain streams, no vegetation in channel, banks usually steep, with trees and brush on bank submerged
a. Bottom: gravels, cobbles, and few boulders	0.030	0.040	0.050
b. Bottom: cobbles with large boulders	0.040	0.050	0.070

The boundary condition is necessary to establish the starting water-surface elevation at either end of the channel system. Since the flow is subcritical, the boundary condition is specified at the downstream end. The average channel slope S_o= 0.004048 has been specified as the downstream boundary condition in the present study.

Flood discharge information is required at each cross section in order to compute the water-surface profile. The flow value is entered at the upstream end of the reach, and it assumes that the flow remains constant until another flow value is encountered with the same reach. The flood discharges corresponding to 10-, 50-, 100-, 500-, 1000-, 5000- and 10000-yr return periods were used to calculate water-surface profiles with HEC-RAS. In addition, the flood discharge of 550 m³/sec, estimated as the 10-yr flood by Comision Nacional del Agua, was also considered.

The HEC-RAS model results are shown in Table 6. Flow depths vary from 3.278 m to 6.779 m; inbank flow velocities vary from 3.6 to 6.04 m/s; overbank flow velocities vary from 0.77 to 1.56 m/s. The design flow velocity for inbank gabion lining is that corresponding to the 1000-yr flood: 5.45 m/s. For the 1000-yr design flood discharge, the freeboard is 1.117 m.

Table 6. HEC-RAS results showing flow depths, mean velocities, Froude numbers, and freeboard for corresponding return periods.
Return period (yr)	Discharge (m³/s)	Flow depth (m)	Inbank flow mean velocity (m/s)	Overbank flow mean velocity (m/s)	Froude number	Freeboard (m)
-	550	3.278	3.60	-	0.68	0.522
10	680	3.706	3.87	-	0.69	0.094
50	1140	4.810	4.70	0.77	0.72	2.190
100	1310	5.147	4.62	1.18	0.73	1.873
500	1600	5.677	5.31	1.17	0.75	1.323
1000	1720	5.883	5.45	1.25	0.75	1.117
5000	2140	6.554	5.90	1.49	0.77	0.450
10000	2290	6.779	6.04	1.56	0.77	0.221

040117