Estimating Predevelopment
Hydrology in the Middle Rio
Grande Watershed, New Mexico
Prepared for
U.S. Environmental Protection Agency
Office of Wastewater Management
Water Permits Division
Municipal Branch
Prepared by
John Kosco, P.E.
Khalid Alvi, P.E.
Mustafa Faizullabhoy, P.E.
it
TETRATECH
10306 Eaton Place, Suite 340
Fairfax, VA 22030
April 2014
EPA Publication Number 832-R-14-007
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM April 2014
(This page was intentionally left blank.)
It
TETRATECH
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Contents
Executive Summary v
1 Introduction 1
2 Rainfall Analysis 3
2.1 Calculating the 95th percentile rainfall event 3
2.2 Rainfall analysis for Albuquerque MS4 area 5
3 Watershed Characterization 12
3.1 Geology and Soils 15
3.2 Land Cover 20
4 Runoff Analysis 23
4.1 Methodology 23
4.2 Results 29
5 Conclusions and Recommendations 33
6 References 34
Tables
Table 2-1. Comparison of rainfall percentiles using three methods 10
Table 3-1. Hydrologic soil group descriptions 15
Table 3-2. Land cover summary (2006 NLCD) 20
Table 4-1. TR-55 CN assumptions for the Albuquerque region (Source: NRCS 1986) 24
Table 4-2.Runoff depths for various rainfall percentiles and CNs 30
Figures
Figure 2-1. Guidance on creating a rainfall frequency spectrum (Hirshman and Kosco 2008) 4
Figure 2-2. Albuquerque precipitation gauges with dates of coverage 6
Figure 2-3. Comparison of percentile storms for Albuquerque-area precipitation gauges 7
Figure 2-4. Distribution of storm events at Albuquerque International Airport (1948-2012) 8
Figure 2-5. Cumulative long-term hourly precipitation diurnal for the Albuquerque International
Airport (NCDC 290234) 9
Figure 2-6. Comparison of rainfall duration curves for the Albuquerque International Airport using
three storm classification methods 11
Figure 2-7. Comparison of rainfall duration curves for the Albuquerque International Airport using
three storm classification methods (85th to 95th percentile range) 11
Figure 3-2. Albuquerque Urbanized Areas and Municipal Boundaries 13
Figure 3-3. Study area 2010 Urbanized Areas and elevation 14
It
TETRATECH
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM April 2014
Figure 3-4. Geology in the study area 16
Figure 3-5. Slope in study area 17
Figure 3-6. Hydrologic soil groups 18
Figure 3-7. Study area runoff potential 19
Figure 3-8. Existing land cover 21
Figure 3-9. Existing imperviousness 22
Figure 4-1. Predeveiopment CNs (Good hydrologic condition) 25
Figure 4-2. Predeveiopment CNs (Poor hydrologic condition) 26
Figure 4-3. Percent imperviousness and land cover distribution summarized by land cover type 27
Figure 4-4. Post-developed CNs (Poor pervious hydrologic condition with impervious mix) 28
Figure 4-5. Nomograph depicting runoff depth as a function of rainfall and CN 29
Figure 4-6. Predeveiopment runoff for 95th percentile rainfall (good hydrologic condition) 31
Figure 4-7. Predeveiopment runoff for 95th percentile rainfall (poor hydrologic condition) 32
It
TETRATECH
iv
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Executive Summary
The city of Albuquerque is located in the heart of the Middle Rio Grande watershed in New Mexico. The
climate of the watershed is arid with an average annual rainfall of approximate 9.5 inches per year
(NOAA 2013). The local climate affects soils and vegetation, which in turn affect runoff and hydrology.
The goal of this study was to determine representative predeveiopment hydrology conditions for Middle
Rio Grande Watershed. Managing stormwater to predeveiopment hydrological conditions for newly
developed and redeveloped sites is a goal of the existing Phase I Albuquerque municipal separate storm
sewer system (MS4) permit and the proposed Middle Rio Grande watershed MS4 permit to improve
water quality. These permits use a percentile storm event approach as a surrogate for mimicking
predeveiopment hydrology. The Phase I Albuquerque permit requires capture of the 90th percentile storm
event. The proposed Middle Rio Grande watershed MS4 permit is proposing a similar standard. This
study aims to clarify the link between the goal of mimicking predeveiopment hydrology in this watershed
and the percentile storm event approach used in the permits.
A recent case study conducted by Tetra Tech under contract with the U.S. Environmental Protection
Agency (EPA) Office of Research and Development focused on investigating both site- and regional-
scale stormwater management questions in the Middle Rio Grande watershed (Shoemaker et al. 2013).
That study provided some useful baseline data for this analysis.
The first step to determine representative predeveiopment hydrology conditions for Middle Rio Grande
Watershed was to characterize local rainfall patterns to identify a range of 24-hour rainfall depths where
measureable runoff might first occur. It was important that the methods applied were proven to be
representative given the arid climate and infrequent rainfall. The second step was to characterize the
physical characteristics within the regulated MS4 boundary of the watershed in terms of soils, geology,
and land cover to define predeveiopment hydrologic conditions. Finally, TR-55 was used to assess the
combinations of rainfall and land cover conditions where runoff was expected to occur. Given the well-
draining nature of soil conditions in the regulated MS4 area, sensitivity tests suggested that the threshold
where measureable runoff occurs would most likely be somewhere between the 90th and 95th percentile
24-hour rainfall depths (0.615 to 0.78 inches).
Given these results, the performance standard to capture the 90th percentile storm event in the current
Phase I Albuquerque permit and the proposed Middle Rio Grande watershed MS4 permit is a reasonable
surrogate for mimicking predeveiopment hydrology for this watershed.
Managing stormwater to pre-development runoff condition will reduce water quality impacts on the
receiving water as development occurs in the watershed. It is anticipated that it will also provide cost
savings for new development and achieve multiple benefits such as reducing local flooding, reducing
drought impacts, making communities more resilient to extreme wet weather events, making
neighborhoods more livable, reducing the urban heat island effect, reducing energy demands, and
improving air quality. Those benefits are also not explicitly quantified in this analysis, but could be
possible indicators to evaluate in future analyses.
It
TETRA TECH
V
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM April 2014
(This page was intentionally left blank.)
It
TETRATECH
vi
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
1 Introduction
The city of Albuquerque is located in the Middle Rio Grande watershed in New Mexico. The Rio Grande
actually has its headwaters in Colorado. By the time the river reaches Albuquerque, New Mexico, the
drainage area encompasses approximately 14,000 square miles. The climate in the Albuquerque area is
arid with an average annual rainfall of approximate 9.5 inches per year (NOAA 2013). The local climate
affects soils and vegetation, which in turn affect runoff and hydrology.
The goal of this study was to determine representative predeveiopment hydrology conditions for Middle
Rio Grande Watershed. Managing stormwater to predeveiopment hydrological conditions for newly
developed and redeveloped sites is a goal of the existing Phase I Albuquerque municipal separate storm
sewer system (MS4) permit and the proposed Middle Rio Grande watershed MS4 permit to improve
water quality. These permits include a post construction standard which uses a percentile storm event
approach as a surrogate for mimicking predeveiopment hydrology. The Phase I Albuquerque permit
requires capture of the 90th percentile storm event. The proposed Middle Rio Grande watershed MS4
permit is proposing a similar standard. This study aims to clarify the link between the goal of mimicking
predeveiopment hydrology in this watershed and the percentile storm event approach used in the permits.
A recent case study conducted by Tetra Tech under contract with the U.S. Environmental Protection
Agency (EPA) Office of Research and Development focused on investigating both site- and regional-
scale stormwater management questions in the Middle Rio Grande watershed (Shoemaker et al. 2013). In
that study, the System for Urban Stormwater Treatment and Integration Analysis (SUSTAIN) was used to
identify cost-effective stormwater management strategies through cost-benefit optimization that reduced
E. coli loading by 66 percent, based on target requirements established by the Middle Rio Grande E. Coli
Total Maximum Daily Load (TMDL). A literature search conducted during that study provided
meaningful guidance about the types of management practices that are commonly used in the region with
estimates of expected treatment effectiveness (Gautam et al. 2010; MRGARWG 2008), as well as a list of
practices that did not work well in arid climates (LaBadie 2010). That study provided some useful
baseline data for this analysis.
Tetra Tech technical direction for this effort was as follows:
• Determine representative predeveiopment hydrology conditions for Middle Rio Grande
Watershed, Albuquerque, NM. Model natural, predeveiopment conditions considering the
appropriate parameters including rainfall, soil types, land cover, evaporation, and others to
estimate the average predeveiopment hydrology for the watershed. The watershed has been well
studied, including a recent SUSTAIN modeling project (Shoemaker et al. 2013), and therefore
baseline data are readily available.
• Estimate an average natural retention/runoff ratio in order to estimate a corresponding percentile
storm event for the watershed that the performance standard can be based on given the results of
the predeveiopment hydrology analysis.
• Provide a chart showing the 95th, 90th, 85th percentile storm event for this watershed based on
local precipitation.
This report describes the analysis that was conducted in fulfillment of these technical directions. First, an
analysis of local rainfall data was conducted to characterize storm behavior (in terms of volumes and
frequencies). That analysis component culminated with estimates of 85th, 90th, and 95th percentile storms.
Second, regional spatial data were analyzed to characterize the spatial variability and distribution of
geological and physical conditions that are typical of natural conditions in the regulated MS4 permit area
of the watershed. Finally, a modeling analysis was conducted using TR-55 to test the combinations of
rainfall and physical watershed conditions to identify the combinations where measurable runoff first
It
TETRA TECH
1
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM April 2014
occurred under predeveiopment conditions. Sensitivity tests were performed to assess the impact of a
range of different percentile rainfall depths and land cover hydrologic conditions on projected runoff
volume. Results of this analysis provide clarify the link between predeveiopment hydrology and the
percentile storm event approach of the performance standard in the MS4 permits for this watershed.
It
TETRATECH
2
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
2 Rainfall Analysis
Understanding the regional weather patterns is essential to accurately reflect expected volume, intensity,
and duration of expected storm events with high spatial variability in meteorology. Given the arid desert
climate of Albuquerque where rainfall is sparse, this understanding becomes even more important. The
area has 300 days of sunshine and about 9 inches of rainfall annually (NOAA 2013).
The method used to calculate the 95th percentile rainfall event is discussed in section 2.1. The rainfall
analysis conducted for the Albuquerque MS4 area is discussed in section 2.2.
2.1 Calculating the 95th percentile rainfall event
Chapter 4 of the Center for Watershed Protection's (CWP) guidance document for building an effective
post-construction stormwater management program (Hirshman and Kosco 2008) recommends using daily
time step data as an approximation for estimating 24-hour rainfall distributions. Figure 2-1, taken from
the CWP guidance, describes the process of developing a rainfall frequency spectrum which is used to
calculate percentile storms for an area. In general, a weather station with at least 30 years of daily rainfall
records is used. Small storms of less than 0.1 inch are edited out and the entire rainfall record is sorted
from largest to smallest and numbered. A percentile is then assigned based on the total number of rainfall
records (for example, the 10th largest storm out of a total record of 500 days of recorded rainfall greater
than 0.1 inch would be in the 98th percentile (500-10)/500 x 100% = 98%).
Why are small storms not included in calculating the percentile rainfall event?
The rainfall from minor storms may be entirely stored in surface depressions and eventually lost to
evaporation or infiltration. As a result, no runoff is produced.
Schueler (1987) developed a Simple Method for estimating storm pollutant load export delivered from
urban development sites. From the analysis of National Urban Runoff Program (NURP) data and storm
events recorded at National Airport, Schueler found that the runoff coefficient needed to be corrected to
eliminate the portion of annual rainfall which does not produce any direct runoff. The analysis found that
about 10% of the annual rainfall volume is so slight that no appreciable runoff is produced.
As also discussed in Section 4, the NRCS Curve Number (CN) method is a simple, widely used and
efficient method for determining the approximant amount of runoff from a rainfall event in a particular
area. TR-55 uses the curve number method and approximates the initial abstraction (Ia) to be equal to 0.2S
where S is related to the Curve Number (CN) as:
5 = iooo_10
CN v '
Runoff (Q) is computed as:
q=^L or (2)
(P+Ia)+S (P+0.85) v '
Where
Q = runoff (inches)
P = rainfall (inches)
S = potential maximum retention after runoff begins (inches)
Ia = initial abstraction (inches), or the amount of water before runoff, such as infiltration, or
rainfall interception by vegetation
TETRATECH
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
If you assume la = 0.1 inch and back calculate CN you get a CN of 95. So all CN's less than 95
inherently have an initial abstraction greater than 0.1-inch. Urban districts (commercial and business
areas) are assigned up to a 95 curve number for the poorest soil conditions, so rainfall below 0.1 inches is
not expected to produce runoff.
Also, Pitt (1999) found that in Milwaukee, rains less than about 0.05 inches did not produce noticeable
runoff and rain events less than 0.5 inches account for most of the events but little of the runoff and
pollutant load discharges and are easiest to control.
A Rainfall Frequency Spectrum (RFS) is a tool that stormwater managers should use to analyze and develop local stormwater
management criteria and to provide the technical foundation for the criteria.
Over the course of a year, many precipitation events occur within a community. Most events are quite small, but a few can
create several inches of rainfall. An RFS illustrates this variation by describing how often, on average, various precipitation
events (adjusted for snowfall) occur during a normal year.
The graph below provides an example of a typical rainfall frequency spectrum and shows the percentage of rainfall events
that are equal to or less than an indicated rainfall depth. As shown, the majority of storm events are relatively small, but there
is a sharp upward inflection point that occurs at about 1 inch of rainfall (90% rainfall event). The 90% rainfall depth is the
recommended standard for the Water Quality Volume (see Table 4.7).
a
5
o
1 T
1-year, 24-hour storm = 2.4"
Target for Channel Protection (CP)
Maximize Runoff Reduction (RR) for All
Runoff Producing Events Up to the 1-year,
24-hour storm
0
0%
90% Rainfall Event = 1
Recommended Water
Quality Volume (WQv)
Percentile
Rainfall Frequency Spectrum
for Mlnneapolis-St. Paul, MN
(1971-2000) with several
noteworthy rainfall events
identified (adapted from
MSSC, 2005).
Guidance on creating an RFS is provided below. If a community is large in area or has considerable variation in elevation or
aspect, the RFS analysis should be conducted at multiple stations.
1. Obtain a long-term rainfall record from an adjacent weather station (daily precipitation is fine, but try to obtain at least 30
years of daily record). NOAA has several Web sites with long-term rainfall records (see http://www.nesdis.noaa.gov). Local
airports, universities, water treatment plants, or other facilities might also maintain rainfall records.
2. Edit out small rainfall events than are 0.1 inch or less, as well as snowfall events that do not immediately melt.
3. Using a spreadsheet or simple statistical package, analyze the rainfall time series and develop a frequency distribution that
can be used to determine the percentage of rainfall events less than or equal to a given numerical value (e.g., 0.2,0.5,1.0,
1.5 inches).
4. Construct a curve showing rainfall depth versus frequency, and create a table showing rainfall depth values for 50%, 75%
90%, 95% and 99% frequencies.
5. Use the data to define the Water Quality storm event (90th percentile annual storm rainfall depth). This is the rainfall depth
that should be treated through a combination of Runoff Reduction (Table 4.6) and Water Quality Volume treatment
(Table 4.7).
6. The data can also be used develop criteria for Channel Protection (Table 4.8). The 1-year storm (approximated in some
areas by the 99% rainfall depth) is a good standard for analyzing downstream channel stability.
7. Other regional and national rainfall analysis such as TP-40 (NOAA) or USGS should be used for rainfall depths or intensity
greater than 1 year in return frequency (e.g., 2-, 5-, 10-, 25-, 50-, or 100-year design storm recurrence intervals).
Figure 2-1. Guidance on creating a rainfall frequency spectrum (Hirshman and Kosco 2008)
TETRATECH
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
2.2 Rainfall analysis for Albuquerque MS4 area
Tetra Tech reviewed rainfall records from NOAA's Global Historical Climatology Network-Daily
(GHCND) precipitation gauges in the Albuquerque area. Precipitation data at the Albuquerque
International Airport (NCDC 290234) was used in this analysis because it represented the gauge with the
longest period of daily rainfall records. However, Tetra Tech also reviewed data from other gauges in the
Albuquerque area to compare rainfall records. As illustrated in Figure 2-2, four gauges have daily rainfall
records with three other gauges having less than daily records and therefore not appropriate for a
percentile rainfall analysis. Figure 2-3 compares the percentile storms for the four gauges with daily
rainfall records. The percentile storms from all four gauges fell in a relatively narrow range, but the
Albuquerque International Airport gauge was generally the lowest.
It
TETRA TECH
5
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
USC00290903
/1895-1982
USC00292100:
1982-2011 (91% coverage for daily')'
AJSC00298011
¦19S3z1979|
^USC00290225:MB
,1991;2011[(100%^coverag?for,daily)
USC00298873
Hi 910-19621
¦BWl'alUSC00290234:S..,^
J r-1931 -2011 (100% coverage for daily)
USC00290231:
,1991-2011 (100% coverage for daily)'
BHlL/'MMt' m
[legend
I
GHCND Station
-i ;j i,
; Urban Outlines (Regional
Albuquerque
Precipitation Gauges
WGS_1984_Web_Mercator_Auxiliary_Sphere
Map produced 04-Apr-2014
0 12 4
I Kilometers
0 12 4
I Miles
It
TETRATECH
SEPA
United States
Environmental Protection
Agency
Figure 2-2. Albuquerque precipitation gauges with dates of coverage
TETRATECH
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Rainfall Distribution
(1991-2014)
ra
cc
3.0
2.5
a>
.c
u
a.
ai
~
S 2.0
1.5
£ 1.0
0.5
0.0
USC00292100 (Corrales)
USC00023050 (Airport)
— USC00290231 (Vail
ey)
USC00290225 (Foothills)
75%
85%
Percentile Storm Event
95%
Figure 2-3. Comparison of percentile storms for Albuquerque-area precipitation gauges
Using observed data from the Albuquerque International Airport from January 1, 1948, through
December 31, 2012, individual precipitation events were categorized assuming a 6-hour inter-event
interval and a minimum storm size of 0.1 inches. The resulting precipitation event distribution from that
period (with about 22 events per year) is presented in Figure 2-4. Of the precipitation events summarized,
over 80 percent were less than 0.5 inch, and 97 percent were less than 1 inch. Knowing the storm
distribution in an arid environment with well-draining soils is important because only the largest of
storms are likely to generate runoff under predeveiopment conditions.
TETRATECH
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Excludes storm intervals < 0.1 inches
<0.25 0.25-0.5 0.5 -0.75 0.75- 1.0 1.0- 1.25 1.25- 1.5 1.5 -2.0 >2.0
Storm Depth
Figure 2-4. Distribution of storm events at Albuquerque International Airport (1948-2012).
Chapter 4 of the Center for Watershed Protection's (CWP) guidance document for building an effective
post-construction stormwater management program (Hirshman and Kosco 2008) recommends using daily
time step data as an approximation for estimating 24-hour rainfall distributions. That was the approach
used for estimating the 85th, 90th, and 95th percentile rainfall depths in the Albuquerque MS4 region.
Recognizing that weather patterns vary significantly by location, it is possible that the use of a static 24-
hour timeframe may introduce a temporal bias in certain climate regions. Nevertheless, the longer the
historical record used, the less impact of that assumption becomes. To test the sensitivity of the static 24-
hour timeframe methodology for weather in Albuquerque, two other analytical methods were compared
with the CWP-recommended approach to see if they produced similar results. Precipitation data at the
Albuquerque International Airport (NCDC 290234) for data collected between January 1, 1948 and
December 31, 2012 were used for all three methods. Also, total storm values less than or equal to 0.1
inches were excluded from the analysis. The three methods were as follows:
1. Recommended: Using a daily time step precipitation data (static 24-hour timeframes)
2. Using a rolling 24-hour window with hourly time step precipitation data
3. Using a storm separation technique with a location-specific inter-event time
For Method 1, daily precipitation totals were first ranked in descending order based on depth and all
values less than or equal to 0.1 inches were excluded. This resulted in 1,898 samples (i.e. daily totals
greater than 0.1 inches) for performing the percentile calculation.
For Method 2, a daily total precipitation depth was computed for each hour that consisted of that hour's
value plus the previous 23 hours. Those daily totals were then ranked in descending order based on depth
and all values less than or equal to 0.1 inches were excluded. This resulted in 36,194 samples for
performing a percentile calculation, compared to 1,898 for Method 1.
For Method 3, performing storm separation first required defining an inter-event time, or the number of
dry hours observed between storms. Unlike the other two methods, storm separation results in storms with
variable durations. Local regulatory frameworks or guidance often define an inter-event time based on
rainfall patterns or management practice draw-down times. In the absence of such specific guidance,
It
TETRATECH
8
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM April 2014
patterns in observed precipitation data can be used to set a representative inter-event time. A long-term
hourly precipitation diurnal was developed for each month using long-term observed data at the
Albuquerque International Airport. This diurnal presented as Figure 2-5 expresses long-term cumulative
precipitation as the percent that fell during each hour or each month.
Percent of Annual Precipitation by Month and Hour (1/1/1948 through 12/31/2012)
January
February
March
April
May
June
July
August
September
October
November
December
Figure 2-5. Cumulative long-term hourly precipitation diurnal for the Albuquerque International
Airport (NCDC 290234).
Representation of long-term precipitation using this monthly-hourly diurnal format highlights systematic
temporal and seasonal patterns that may not be as evident from time series plots. The data show that
roughly 10 percent of the total annual rainfall falls in the six hour window between 3:00 PM and 9:00 PM
in July, while another 11 percent falls between 3:00 PM and 11:00 PM in August (both outlined in black
on the graph above). Based on the analysis presented in Figure 2-5, a six hour inter-event time was
selected as the locally-representative threshold criteria for performing storm separation. Similar to the
previous two methods, precipitation totals for each resulting storm interval were also ranked in
descending order based on depth and all values less than or equal to 0.1 inches were excluded. This
resulted in 1,435, the fewest number of samples for performing a percentile calculation.
The ranked precipitation series were used to calculate the 85th, 90th, and 95th percentile storm depth for
each of the three methods. A comparison of the three methods and resulting percentile storm depths is
presented below in Table 2-1.
It
TETRATECH
9
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Table 2-1. Comparison of rainfall percentiles using three methods
Calculation
Method
Observed
Data
Time Step
Number
of Samples
Percentile Rainfall Depths (inches)
85th
90th
95th
Method 1 - Fixed 24-
hour window
Daily
1,989
0.53
0.615
0.78
Method 2 - Rolling 24-
hour window
Hourly
36,194
0.50
0.60
0.77
Method 3 - Storm
separation
Hourly
1,435
0.53
0.635
0.84
The results showed that the three methods produced very similar rainfall duration depths, suggesting that
using daily rainfall as a 24-hour storm approximation in arid Albuquerque produced representative
estimates for the 85th, 90th, and 95th percentile rainfall depths. There was a long enough historical record
to produce a diverse sample space of daily rainfall totals using a fixed 24-hour stepping time window. As
expected, using the rolling 24-hour window produced the smoothest curve because it used the highest
number of samples. In cases where only short historical records are available that method would offer
some advantages. It also minimizes the impact of imposing a human construct (i.e. the clock or calendar)
into the analysis, since any continuous 24-hour period could technically be considered a day. The storm
separation approach results deviated from the other two methods at higher percentiles. This also made
sense given that storm separation imposes no maximum duration on storm periods. Figure 2-6 presents
the final storm percentile curves for all three methods derived using rainfall data from the Albuquerque
International Airport (NCDC 290324) from 1/1/1948 through 12/31/2012. Figure 2-7 shows the same
curves zoomed into the 85th to 95th percentile range.
It
TETRATECH
10
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Albuquerque international Airport (NCDC 290234)
January 1,1948 - December 31,2012
2.4
—Fixed 24-hour Interval
2.2
—Rolling 24-hour Interval
Storm Interval Separation
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Percentile
Figure 2-6. Comparison of rainfall duration curves for the Albuquerque International Airport using
three storm classification methods.
Albuquerque International Airport (NCDC 290234)
January 1,1948-December 31,2012
1.0
—Fixed 24-hour Interval
— Rolling 24-hour Interval
Storm Interval Separation
0.9
0.8
0.7
0.6
0.5
85%
90%
95%
100%
Percentile
Figure 2-7. Comparison of rainfall duration curves for the Albuquerque International Airport using
three storm classification methods (85th to 95th percentile range).
It
TETRATECH
11
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
3 Watershed Characterization
Located in north-central New Mexico, the focus area for this analysis is the regulated MS4 permitted area
in the Middle Rio Grande watershed. The 2000 and 2010 Census urbanized areas along with municipal
boundaries are shown on Figure 3-1. The 2010 Census urbanized area boundary was used to indicate the
MS4 permitted area in the watershed as depicted on Figure 3-3. It covers an area of approximately 250
square miles (650 square kilometers), with an average elevation of approximately 1,600 meters (5,200
feet) above sea level. The Middle Rio Grande watershed drains to the Rio Grande River. The Rio Grande
River begins in the Rocky Mountains in Colorado and flows through New Mexico on its way to the Gulf
of Mexico. The Rio Grande River bisects the study area and is the subject of an E. coli TMDL that was
approved by EPA in 2010. The current Phase I Albuquerque MS4 permit has four co-permittees including
the city of Albuquerque, Albuquerque Metropolitan Arroyo Flood Control Authority, New Mexico
Department of Transportation, and University of New Mexico. The Phase II MS4 permit covers the
following entities: City of Albuquerque, Kirtland Air Force Base, NM Dept. of Transportation, City of
Rio Rancho, Village of Los Ranchos de Albuquerque, Bernalillo County, Sandoval County, Southern
Sandoval County Arroyo Flood Control Authority, Town of Bernalillo, Village of Corrales, Pueblo of
Isleta, Pueblo of Sandia, and Pueblo of Santa Ana. The proposed Middle Rio Grande watershed MS4
permit would supersede the existing Phase I and II permits and could also permit the Eastern Sandoval
Flood Control Authority, EXPO (State Fairgrounds) and Sandia Laboratories, Dept. of Energy.
It
TETRATECH
12
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
San Felipe Pueblo
Santa Ana Pueblo
Bernalillo
Algodones
Rio Rancho
[wYAVyW
Placitas
Pueblo of Sandia Village
Corrales
North Valley
Los Ranchos
CXXXXXXXXXXXXXX
xxxxx>oo
8888888^9999^
>00088$8$8SSSS
'yVxAA
XXXxX, JJS
arnuel
>oooo666ood<
Swyywwx
OwXxxX
>^vOvsa
AXWWWWy\ ,
KX>
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Albuquerque Urbanized Areas
Elevation
I Kilometers
11
¦ Miles
Soil Survey Geographic (SSURGO^J>rfsbase for [Rio Grande-Albuquerque, NM], Available online at
http://www.arcgis.com/apps/OnePane/basicviewer/index.html?appid=a23eb436f6ec4ad6982000dbaddea5ea. Accessed [11/11/2013].
Legend
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
3.1 Geology and Soils
The study area lies in the valley between the Sandia Mountains to the east and the divide between the Rio
Grande and Rio Puerco drainage basins to the west. The valley is filled with thick alluvial deposits which
allow for much of the rainfall to infiltrate (Figure 3-4). Other geologic formations are present in the study
area including clastic and volcanic formations. The majority of the study area is fairly flat, with slopes
less than 5 percent (Figure 3-5). The data presented in this section are derived from the Soil Survey
Geographic Database (SSURGO) which is maintained by the U.S. Department of Agriculture Natural
Resources Conservation Service.
The National Cooperative Soil Survey publishes soil surveys for each county within the U.S. These soil
surveys contain predictions of soil behavior for selected land uses. The soil surveys are designed for many
different uses, including land use planning, the identification of special practices needed to ensure proper
performance, and mapping of hydrologic soil groups (HSGs).
HSGs refer to the grouping of soils according to their runoff potential. Soil properties that influence the
HSGs include depth to seasonal high water table, infiltration rate and permeability after prolonged
wetting, and depth to slow permeable layer. There are four groups of HSGs: Group A, B, C, and D as
described in Table 3-1. Figure 3-6 presents the spatial distribution of HSGs in the study area. The
majority of the regulated MS4 area in the watershed (70 percent) is HSG B. Sandier soils (HSG A) tend
to be present in areas with high slope (greater than 5 percent slope).
Table 3-1. Hydrologic soil group descriptions
HSG
Group Description
A
Sand, loamy sand or sandy loam types of soils. Low runoff potential and high infiltration rates
even when thoroughly wetted. Consist chiefly of deep, well to excessively drained sands or
gravels with a high rate of water transmission.
B
Silt loam or loam. Moderate infiltration rates when thoroughly wetted. Consist chiefly or
moderately deep to deep, moderately well to well drained soils with moderately fine to moderately
coarse textures.
C
Soils are sandy clay loam. Low infiltration rates when thoroughly wetted. Consist chiefly of soils
with a layer that impedes downward movement of water and soils with moderately fine to fine
structure.
D
Soils are clay loam, silty clay loam, sandy clay, silty clay or clay. Group D has the highest runoff
potential. Low infiltration rates when thoroughly wetted. Consist chiefly of clay soils with a high
swelling potential, soils with a permanent high water table, soils with a clay pan or clay layer at or
near the surface and shallow soils over nearly impervious material.
The SSURGO layer also provides independently estimated runoff potential on a relative scale ranging
from "Negligible" to "Very High," as shown in Figure 3-7. About 91 percent of the regulated MS4 area
in the watershed has either "Low" or "Very Low" runoff potential, while about another 6 percent has
"Medium" runoff potential. Less than 3 percent of the MS4 area has "High" or "Very High" runoff
potential.
It
TETRATECH
15
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Survey Geographic (SSURGCj.prab.
liable online at http://datagateWay.nrcs
Albuquerque Urbanized Areas
Geology
I Kilometers
11
¦ Miles
Legend
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Soil Survey Geographic (SSURG
http://www.arcgis.com/apps/OneP
Albuquerque Urbanized Areas
Slope
I Kilometers
11
¦ Miles
aase for [Rio Grande-Albuquerque, NM], Available online at
rne/basicviewer/index.html?appid=a23eb436f6ec4ad6982000dbaddea5ea. Accessed [11/11/2013],
Legend
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Soil Survey Geographic (SSURG^^Hfabase for [Rio Grande-Albuquerque, NM]. Available online at
http://wwwarcgis.com/apps/OnePane/basicviewer/index.html?appid=a23eb436f6ec4ad6982000dbaddea5ea
Albuquerque Urbanized Areas
Hydrologic Soil Group
I Kilometers
11
¦ Miles
Soil Survey Geographic
Legend
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Legend
Albuquerque
03 Ri° Grande
Stream
Runoff
Medium
Very low
Negligible
Soil Survey Geographic (SSURG©)JJ
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
3.2 Land Cover
The objective of this study was to characterize predeveiopment conditions for storm water management
targets; therefore, land cover, which is mostly a reflection of existing conditions, is not as important to the
analysis as underlying geology or soils. Land use and impervious data are provided by the Multi-
Resolution Land Characteristics Consortium (MRLC). The MLRC distributes the National Land Cover
Database (NLCD), a 30-meter resolution land cover database. Land cover in the study area is comprised
predominately of developed land uses with shrub/scrub and pasture common near the river and along
steep slopes in the northwest (Figure 3-5 and Figure 3-8). Impervious areas are common throughout the
study area due to the developed state of the watershed (Figure 3-9).
Table 3-2. Land cover summary (2006 NLCD)
Land cover
Percent of MS4
area (%)
Total land area
(acres)
Developed (high, medium, low intensity and open space)
66%
106,650
Grassland and shrubland
22%
35,127
Pasture and crops
9%
14,205
Wetland and water
3%
5,321
Other (forest and barren)
<1%
784
It
TETRATECH
20
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Albuquerque Urbanized Areas
Land Use Classification (2006)
I Kilometers
11
¦ Miles
Legend
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Albuquerque Urbanized Areas
Percent Imperviousness
I Kilometers
11
¦ Miles
Legend
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
4 Runoff Analysis
TR-55 is perhaps the most widely used model for predicting hydrology in the United States. It is a
simplified empirical method that uses the Soil Conservation Service runoff equation for estimating runoff
volumes from rainfall and times of concentration as a function of precipitation and land conditions. The
simplifications often limit the use of the procedures for runoff estimation because they provide less
accurate results than more detailed deterministic hydrology models would provide. However, the effects
of simplification are more of a concern when predicting peak discharges or runoff hydrographs. The TR-
55 documentation advises users to examine the sensitivity of the analysis being conducted to ensure that
the degree of error is tolerable (NRCS 1986).
For this study, the objective was to identify the rainfall volume associated with predeveiopment
conditions. Consequently, TR-55 was only used to estimate the earliest portions of the hydrograph when
measurable runoff begins. TR-55 approximates the initial abstraction (7a) to be equal to 0.2S where S is
related to the Curve Number (CN) as:
5 = iooo_10
CN v '
Runoff (Q) is computed as:
Q=i£zk21 or (2)
(P+Io)+S (P+0.85) v '
Where
Q = runoff (inches)
P = rainfall (inches)
S = potential maximum retention after runoff begins (inches)
Ia = initial abstraction (inches).
Initial abstraction represents all surface capture or losses before runoff begins. As observed in Equation 1
above, smaller CN values will yield a larger value of S, and in turn, larger initial abstraction values. This
empirical parameter assumes that the initial abstraction is correlated to surface depressions, soil type,
vegetation cover, evaporation, and infiltration. The soil and land use influences are expressed by way of
the CN. Using Equations 1 and 2, a nomograph can be generated to estimate the direct runoff volume as a
function of precipitation and CN. It is recognized that the peak flow would still vary for a given runoff
depth since it is a function of the time of concentration (and Ia/P) for a given drainage area; nevertheless,
because we are only interested in the point of the hydrograph where measureable runoff first occurs, there
is more confidence (and less uncertainty) about that prediction.
4.1 Methodology
Inputs for the TR-55 procedure are derived from local land cover, soil, and precipitation data. Rainfall
inputs to TR-55 were derived from the analysis in Section 2. In fact, rainfall depths between 0.1 extending
all the way to 1.5 inches were evaluated, which included the 85th, 90th, 95th percentile rainfall depths and
beyond.
CNs in TR-55 are typically between 30 and 98, with the smaller CN indicating less runoff. In fact, some
CN values are lower than 30; however TR-55 suggests using 30 as a minimum value for calculations. A
CN of 98 is often used to represent impervious areas. Because Albuquerque is in the arid southwestern
United States, CN values for "Desert Shrub"—where major plants include saltbush, greasewood,
creosotebush, blackbrush, bursage, palo verde, mesquite, and cactus—were used to characterize
TETRATECH ——
23
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
predeveiopment conditions. TR-55 further qualifies CN values for that category into "Poor," "Fair," or
"Good" hydrologic conditions. "Poor" means <30 percent ground cover (e.g. litter, grass, and brush
overstory), while "Good" indicates >70 percent ground cover—"Fair" is between 30 to 70 percent (NRCS
1986). The land cover values for predeveiopment and developed urban areas vary by HSG as shown in
Table 4-1. For reference purposes, CN values for "woods in good condition" and "woods in poor
condition" are also shown for comparison with desert shrub values. Those values might typically have
been used to represent predeveiopment conditions for most of the United States. The differences between
"woods" and "desert shrub" are more pronounced for A and B soils than for C and D soils. The published
"desert shrub" values are also higher than "woods" because rainfall totals are lower in the desert
southwest. For comparison, "desert shrub" values in "good" and "poor" condition were used for
predeveiopment conditions to test the implication of predeveiopment hydrologic condition on estimated
runoff—"good" condition (Figure 4-1) yields less runoff than the "poor" condition (Figure 4-2).
Table 4-1. TR-55 CN assumptions for the Albuquerque region (Source: NRCS 1986)
Land Cover
Curve Number (CN) by Hydrologic Soil Group (HSG)
A
B
C
D
Predeveiopment
Conditions
Woods in good condition1
30
55
70
77
Woods in poor condition1
45
66
77
83
Desert shrub in good condition
49
68
79
84
Desert shrub in poor condition
63
77
85
88
Western Desert
Urban Areas
Natural desert landscaping2
63
77
85
88
Artificial desert landscaping
with impervious weed barrier
96
96
96
96
Impervious
Surfaces
Parking lots, roofs, driveways
(excluding rights-of-way)
98
98
98
98
Mixed Urban
Land Cover
Composite pervious/
impervious land cover
Composite CNs are usually calculated using area-weighting
of directly-connected/unconnected impervious with pervious
1: Shown for reference purposes only for comparison with desert shrub CN values
2: TR-55 assumes the same values as "Desert shrub in poor condition"
It
TETRATECH
24
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Albuquerque MS4 Areas
CN for Predevelopment (Good)
I Kilometers
11
¦ Miles
Legend
Albuquerque
03 Rio Grande
Stream
Curve Number
| 49-54
| 55-59
60-64
65-69
70-74
75-80
81 -85
86-90
91-95
I 96-100
GCS North American 1983
Figure 4-1. Predevelopment CNs (Good hydrologic condition).
It
TETRATECH
25
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Albuquerque MS4 Areas
CN for Predevelopment (Poor)
I Kilometers
11
¦ Miles
Legend
Albuquerque
03 Rio Grande
Stream
Curve Number
| 49-54
| 55-59
60-64
65-69
70-74
75-80
81 -85
86-90
91-95
I 96-100
GCS North American 1983
Figure 4-2. Predevelopment CNs (Poor hydrologic condition).
It
TETRATECH
26
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
The post-developed condition curve numbers were developed by area-weighting poor-condition pervious
CN values with an impervious CN value of 98 for NLCD 2006 land cover and impervious area
distribution. Figure 4-3 shows percent imperviousness and land cover distribution summarized by land
cover type within the MS4 jurisdictional boundary. Although high intensity development is about 75
percent impervious on average, it only makes up 3 percent of the MS4 area. There is some impervious
cover within the undeveloped land areas (about 2 percent on average); however, these are most likely
small distributed structures or hard surfaces that are most likely not directly connected impervious area.
Given the prevalence of well-draining A and B soils, all impervious areas were assumed to be directly
connected for generating composite CN values through area-weighting (as a conservative assumption).
Figure 4-4 is a map of the composite CN spatial distribution within the MS4 jurisdictional boundary.
id
CL
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
75%
I High ¦ Medium ¦ Low ¦ Open BOther
35%
14%
55%
High
34%
Land Cover Distribution
Medium
11%
Low
Open
2%
Other
Land Cover
Developed Land Cover
Figure 4-3. Percent imperviousness and land cover distribution summarized by land cover type.
TETRATECH
27
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Albuquerque MS4 Areas
CN for Post-Developed Condition
I Kilometers
11
¦ Miles
Legend
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
4.2 Results
The TR-55 equations described previously were used to generate a runoff nomograph using regionally-
representative CN and rainfall depth combinations as presented in Figure 4-5. The nomograph presents
runoff depths between 0.01 and 1.5 inches depending on the rainfall and CN combination. Runoff depths
between 0.01 and 0.1 inches were considered as part of the analysis to provide a broader sample space to
define points where measureable runoff first occurs. Three sets of runoff estimates (each set having three
rainfall depths 85th, 90th, and 95th percentile) were generated using TR-55 assuming:
1. Predevelopment CNs and good hydrologic condition (from Figure 4-1)
2. Predevelopment CNs and poor hydrologic condition (from Figure 4-2)
3. Post-development CNs and poor hydrologic condition (from Figure 4-4).
Albuquerque
Runoff Depth - SCS TR55 Method
60 65 70 75 SO 85 90 95 100
Curve Number (CN)
Figure 4-5. Nomograph depicting runoff depth as a function of rainfall and CN,
It
TETRATECH
29
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Because the CWP guidance already excludes rainfall events less than 0.1 inches, selecting an even lower
runoff threshold provides a more conservative definition of "measurable" runoff. For this analysis, the
runoff threshold range between 0.01 and 0.1 inches was considered for the range where measureable
runoff first occurs. The 85th, 90th, and 95th percentile rainfall depths result in measureable runoff occurring
above CN values greater than or equal to 77. These data indicate that under predeveiopment conditions,
there would only be measurable runoff occurring from very small pockets within the study area where the
CN is 77 or greater and around the 95th percentile rainfall depth (0.78 inches of rainfall). Very little
predeveiopment runoff is expected for much of the watershed because HSG is mostly A and B. Table 4-1
summarizes TR-55 runoff simulations for the three rainfall depths and the range of CN values above
which runoff (defined as > 0.01 inches) occurs. Spatial maps of runoff were generated by joining the
underlying TR-55 runoff estimates (plotted in Figure 4-5 and partially summarized in Table 4-1) with the
curve number maps previously shown. Of the six predeveiopment runoff maps that were evaluated, only
the 95th percentile rainfall depth produced sizeable areas of measureable runoff, as plotted in Figure 4-6
for good hydrologic condition and Figure 4-7 for poor hydrologic condition.
Table 4-2. Runoff depths for various rainfall percentiles and CNs
Rainfall Percentile
85th
90th
95th
Rainfall Depth (in.):
0.53
0.615
0.78
Curve Number (CN)
Runoff depth (inches)
<74
76
0.01
78
0.02
80
0.01
0.03
82
0.01
0.05
84
0.01
0.03
0.07
86
0.02
0.04
0.10
88
0.04
0.07
0.14
90
0.07
0.10
0.19
92
0.10
0.15
0.25
94
0.16
0.21
0.33
96
0.23
0.30
0.44
98
0.35
0.42
0.58
100
0.53
0.62
0.78
Runoff: Low -> High
It
TETRATECH
30
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Albuquerque MS4 Areas
Runoff for 95th Percentile Rain
Predevelopment (Good)
I Kilometers
11
¦ Miles
Legend
Albuquerque
03 Rio Grande
Stream
Runoff (in.)
I 0
| 0.02
IB 0.04
| 0.06
0.08
I °-1
0.2
0.3
0.4
0.5
0.7
0.8
GCS North American 1983
Figure 4-6. Predevelopment runoff for 95th percentile rainfall (good hydrologic condition).
It
TETRATECH
31
-------
Estimating Predevelopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
Albuquerque MS4 Areas
Runoff for 95th Percentile Rain
Predevelopment (Poor)
I Kilometers
11
¦ Miles
Legend
Albuquerque
03 Rio Grande
Stream
Runoff (in.)
I 0
| 0.02
IB 0.04
| 0.06
0.08
I °-1
0.2
0.3
0.4
0.5
0.7
0.8
GCS North American 1983
Figure 4-7. Predevelopment runoff for 95th percentile rainfall (poor hydrologic condition).
It
TETRATECH
32
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
5 Conclusions and Recommendations
This analysis evaluates a range of rainfall depths and estimated runoff responses that constitute and
describe predeveiopment conditions. Here are some observations from this analysis in the regulated MS4
area of the watershed:
1. Estimated rainfall depths for the 85th, 90th, and 95th percentile are 0.53, 0.615, and 0.78 inches,
respectively.
2. The regulated MS4 area of the watershed is well drained, fairly flat, and has a low potential for
runoff in areas with low imperviousness (A and B type soils).
3. Under natural/predevelopment conditions, there is little to no measureable runoff generated in the
study area for about 95 percent of all rainfall events.
4. If the predeveiopment condition in the regulated MS4 area of the watershed is defined as the
rainfall depth above which measurable runoff first occurs under natural conditions, then that
threshold is somewhere between the 90th and 95th percentile 24-hour rainfall depths (0.615 to 0.78
inches)
The predeveiopment hydrology of the watershed is based on the critical combinations of rainfall, soils,
and land cover that results in measureable runoff. For this analysis CN is used as a surrogate indicator of
soil and land cover conditions, with higher values indicating more runoff potential. Using the TR-55
equations, the 85th percentile rainfall event (0.53 inches) begins to generate measureable runoff (> 0.01
inches) at a CN of 84; however, under predeveiopment conditions, the worst-case CN values for desert
shrub in poor hydrologic condition are 85 and 88 for C and D soils respectively, but only 77 for B soils
and 63 for A soils, which together represent 98 percent of the soil within the MS4 focus area. Only the
90th and 95th percentile storms begin generating runoff at lower CN values of 80 and 76, respectively.
This analysis excludes all 24-hour storms less than 0.1 inches. For this reason, the level of runoff
considered to be "measurable" was tested within a range of values ranging between 0.01 and 0.1 inches of
runoff. Depending on whether 0.01 or 0.1 inches is considered as the threshold for "measurable" runoff,
even some developed areas (i.e. low intensity or open space) did not generate enough runoff to meet that
threshold.
Given these results, the performance standard to capture the 90th percentile storm event in the current
Phase I Albuquerque permit and the proposed Middle Rio Grande watershed MS4 permit is a reasonable
surrogate for mimicking predeveiopment hydrology for this watershed.
Managing stormwater to pre-development runoff condition will reduce water quality impacts on the
receiving water as development occurs in the watershed. It is anticipated that it will also provide cost
savings for new development and achieve multiple benefits such as reducing local flooding, reducing
drought impacts, making communities more resilient to extreme wet weather events, making
neighborhoods more livable, reducing the urban heat island effect, reducing energy demands, and
improving air quality. Those benefits are also not explicitly quantified in this analysis, but could be
possible indicators to evaluate in future analyses.
It
TETRATECH
33
-------
Estimating Predeveiopment Hydrology in the Middle Rio Grande Watershed, NM
April 2014
6 References
Albuquerque GIS (Albuquerque geographic information systems). 2012. Data downloads.
. Accessed January 14, 2012.
Gautam, M., K. Acharya., and M. Stone. 2010. Best Management Practices for Stormwater Management
in the Desert Southwest. Journal of Contemporary Water Research & Education 146:39-49.
Hirshman, D. and J. Kosco. 2008. Managing Stormwater in Your Community: A Guide for Building an
Effective Post-Construction Program. EPA/833/R-08/001. Center for Watershed Protection. U.S.
Environmental Protection Agency, Washington, DC.
LaBadie, K.T. 2010. Identifying Barriers to LID and GI in the Albuquerque Area.
. Accessed August 23, 2011.
MRGARWG (Middle Rio Grande-Albuquerque Reach Watershed Group). 2008. Middle Rio Grande-
Albuquerque Reach Watershed Restoration Action Strategy. Prepared for the New Mexico
Environmental Department, Santa Fe, NM.
NMED-SWQB (New Mexico Environment Department-Surface Water Quality Bureau). 2010. Total
Maximum Daily Load (TMDL) for the Middle Rio Grande Watershed. New Mexico
Environmental Department, Santa Fe, NM.
NOAA. 2013. 2012 Weather Highlights - Temperature & Precipitation: Albuquerque.
http://www.srh.noaa.gov/abq/?n=climonhigh2012annual-tempprecipabq Accessed April 22,
2014.
NRCS (Natural Resource Conservation Service). 1986. Urban Hydrology for Small Watersheds,
Technical Release 55 (TR-55). U.S. Department of Agriculture, NRCS, Conservation Engineering
Division. 210-VI-TR-55, Second Edition, June 1986.
Pitt, R. 1999. Small Storm Hydrology and Why it is Important for the Design of Stormwater Control
Practices. In: Advances in Modeling the Management of Stormwater Impacts, Volume 7. (Edited
by W. James). Computational Hydraulics International, Guelph, Ontario and Lewis
Publishers/CRC Press. 1999.
http://rpitt.eng.ua.edu/Publications/UrbanHvandCompsoils/small%20storm%20hvdrology%20Pit
t%20j ames98.pdf
Schueler, T. 1987. Controlling Urban Runoff: A Practical Manual for Planning and Designing Urban
BMPs. Metropolitan Washington Council of Governments.
Shoemaker, L., J. Riverson, K. Alvi, J. X. Zhen, and R. Murphy. 2012. Report on Enhanced Framework
(SUSTAIN) and Field Applications to Placement of BMPs in Urban Watersheds. EPA/600/R-
11/144. U.S. Environmental Protection Agency, Washington, DC.
Shoemaker, L., J. Riverson, K. Alvi, J. X. Zhen, R. Murphy, B. Wood. 2013. Stormwater Management
for TMDLs in an Arid Climate: A Case Study Application of SUSTAIN in Albuquerque, New
Mexico. EPA/600/R-13/004. U.S. Environmental Protection Agency, Washington, DC.
U.S. EPA. 2009. Technical Guidance on Implementing the Stormwater Runoff Requirements for Federal
Projects under Section 438 of the Energy Independence and Security Act. EPA 841 -B-09-001.
http://water.epa.gov/polwaste/nps/upload/eisa-438.pdf
It
TETRATECH
34
------- |