US Army Corps of Engineers® Los Angeles District

Prado Basin Ecosystem Restoration and Water Conservation Study

APPENDIX N Water Conservation and Sediment Transport Analysis Report /Orange County Prado Basin Feasibility Study

WATER CONSERVATION AND SEDIMENT TRANSPORT ANALYSIS REPORT

Hydraulics Section Hydrology & Hydraulics Branch U.S. Army Corps of Engineers Los Angeles District

February 2021 Table of Contents 1. INTRODUCTION ...... 6 2. STUDY AREA ...... 6 3. BACKGROUND INFORMATION ON ...... 7 3.1 General ...... 7 3.2 Pertinent Data ...... 8 3.3 Current Water Control Plan ...... 8 4. CLIMATE CHANGE CONSIDERATION ...... 9 4.1 Current Climate and Climate Change Trends ...... 9 4.2 Projected Climate Change and Business Line Impacts ...... 12 5. DESCRIPTION OF WATER CONSERVATION TERMINOLOGY ...... 15 5.1. Yield ...... 15 5.2. Water “Lost” ...... 15 5.3. Present Conditions ...... 16 5.4. Future Conditions ...... 16 5.5. Debris Pool ...... 16 5.6. Buffer Pool ...... 16 6. LOWER SANTA ANA RIVER REACHES ...... 17 7. RESERVOIR SIMULATION ...... 18 7.1. Representative Period of Record ...... 18 7.2. Frequency ...... 18 7.3. Data Collection ...... 19 7.4. Inflow Adjustments and Projections ...... 19 7.5. Evaporation ...... 19 7.6. Elevation-Area-Storage Capacity Curve...... 20 7.7. Present Condition: Flood Season Water Conservation to Elevation 498.0 ft plus Non-Flood Season Water Conservation to Elevation 505.0 ft. (no deviation in place) ...... 20 7.8. Alternative 4: Water Conservation to Elevation up to 505.0 ft during both flood and non-flood seasons...... 21 7.9. Sediment ...... 21 7.10. Analysis Overview ...... 22 7.11. Results ...... 22

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 2 7.12. Sensitivity Analysis ...... 23 7.13. Flood Risk Due To Project Alternative ...... 23 8. SEDIMENT TRANSPORT – UPPER SANTA ANA RIVER (USAR) ...... 24 8.1 Effects of Sediment trap and transition channel ...... 24 8.2 Effects of Increasing Reservoir Water Conservation Elevation ...... 26 9. SEDIMENT TRANSPORT – LOWER SANTA ANA RIVER (LSAR) ...... 27 10. REFERENCES ...... 30

Tables Table 1: Pertinent Data – With Phase II Modifications Prado Dam and Reservoir Table 2: Identified Climate Risks Associated with Measures Table 3: Projected Annual Inflow for 2021 and 2071 Table 4: Monthly Evaporation Rates for Prado Dam Used in HEC-5 Model Table 5: Elevation-Area-Capacity Relationship Prado Dam Present Conditions – Year 2021 Table 6: Elevation-Area-Capacity Relationship Prado Dam Future Conditions – Year 2071 Table 7: HEC-5 Results – Without Project – Present Conditions – 350 cfs Recharge Rate Table 8: HEC-5 Results – Alternative 4 – Present Conditions – 350 cfs Recharge Rate Table 9: HEC-5 Results – Without Project – Future Conditions – 350 cfs Recharge Rate Table 10: HEC-5 Results – Alternative 4 – Future Conditions – 350 cfs Recharge Rate Table 11: Project Benefit Analysis – 350 cfs Fixed Recharge Rate Table 12: Inundation Durations for Prado Dam – Without Project – Present Conditions – 350 cfs Table 13: Inundation Durations for Prado Dam – Present Conditions – 350 cfs Table 14: Inundation Durations for Prado Dam – Without Project – Future Conditions – 350 cfs Table 15: Inundation Durations for Prado Dam – Future Conditions – 350 cfs Table 16: HEC-5 Results – Without Project – Present Conditions – 450 cfs Recharge Rate Table 17: HEC-5 Results – Alternative 4 – Present Conditions – 450 cfs Recharge Rate Table 18: HEC-5 Results – Without Project – Future Conditions – 450 cfs Recharge Rate Table 19: HEC-5 Results – Alternative 4 – Future Conditions – 450 cfs Recharge Rate Table 20: Project Benefit Analysis – 450 cfs Maximum Recharge Rate Table 21: Inundation Durations for Prado Dam – Without Project – Present Conditions - 450 cfs Table 22: Inundation Durations for Prado Dam – Alternative 4 – Present Conditions - 450 cfs Table 23: Inundation Durations for Prado Dam – Without Project – Future Conditions - 450 cfs Table 24: Inundation Durations for Prado Dam – Alternative 4 – Future Conditions - 450 cfs

Figures Figure 1: Santa Ana River Drainage Area Basin Map Figure 2: Prado Basin Study Area Map Figure 3: Interim Water Control Plan – May 2003 Figure 4: Modeling Extent and Trap Scenario Modeled Figure 5: River Bed Profile Change after 10 Years for Base Case Scenario Figure 6: River Bed Profile Change after 10 Years for Trap Scenario

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 3 Figure 7: Comparison of Channel Slope Change after 10 Years Figure 8: Comparison of Average D50 Change after 10 Years Figure 9: River Bed Profile Change after 10 Years for 1st Scenario Figure 10: River Bed Profile Change after 10 Years for 2nd Scenario Figure 11: Deposited and Re-entrained Sediment – 1% Fine Sand Figure 12: Deposited and Re-entrained Sediment – 1% Medium Sand Figure 13: Deposited and Re-entrained Sediment – 5% Fine Sand Figure 14: Deposited and Re-entrained Sediment – 5% Medium Sand Figure 15: River Bed Profiles after 5 and 10 Years – Base Fine Sand Figure 16: River Bed Profiles after 5 and 10 Years – 1% Fine Sand Figure 17: River Bed Profiles after 5 and 10 Years – 1% Medium Sand Figure 18: River Bed Profiles after 15 Years – 1% Fine Sand Figure 19: River Bed Profiles after 15 Years – 1% Medium Sand

Plates Plate 1: Annual Peak Instantaneous Streamflow Santa Ana River at E Street Plate 2: Annual Peak Instantaneous Streamflow Chino Creek at Schaeffer Avenue Plate 3: Annual Peak Instantaneous Streamflow Temescal Creek above Main Street Plate 4: Annual Peak Instantaneous Streamflow Cucamonga Creek near Mira Loma, CA Plate 5: Nonstationarity Analysis of Maximum Annual Flow: Santa Ana River Plate 6: Nonstationarity Analysis of Maximum Annual Flow: Chino Creek Plate 7: Nonstationarity Analysis of Maximum Annual Flow: Temescal Creek Plate 8: Nonstationarity Analysis of Maximum Annual Flow: Cucamonga Creek Plate 9: Range in Projected Annual Maximum Monthly Flows, Huc-1807 Plate 10: Mean Projected Annual Maximum Monthly Stream Flow, Huc-1807 Plate 11: Projected 100-year 24 HR Rainfall RCP8.5 Plate 12: Projected Temperature Rise and Late Century Drought Plate 13: Projected Vulnerability Flood Risk Reduction Business Line for Huc-1807 Plate 14: Projected Vulnerability Ecosystem Restoration Business Line for Huc-1807 Plate 15: Projected Vulnerability Water Supply Business Line for Huc-1807

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 4 Attachments Available Upon Request

Attachment A Prado Basin, , Water Conservation Memorandum of Agreement - 7 July 2006

Attachment B Technical Memorandum, Evaluation of Suitable Discharge Rate from Prado Dam to Surface Water Recharge System - February 5, 2013, OCWD

Attachment C Prado Basin Daily Discharge Estimate for 2021 and 2071, Using the Wasteload Allocation Model - December 2013, Wildermuth Environmental, Inc.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 5 PRADO B A S IN FEASIBILITY STUDY WATER CONSERVATION AND SEDIMENT TRANSPORT ANALYSIS

1. INTRODUCTION

The purpose of the Prado Basin Feasibility Study (Feasibility Study) is to investigate alternatives to restore environmental resources within Prado Basin and downstream of the Prado Dam, within the Santa Ana River (SAR). This study will also evaluate the alternatives to increase the existing volume of water conservation potential from what is currently identified as part of the approved water control plan. This effort in support of the Feasibility Study, as provided in this appendix, will focus on restoring aquatic, wetland and riparian habitats for endangered and other significant species, increasing water conservation benefits and reducing problems caused by sediment trapped by Prado Dam. All the elevations in this appendix are with respect to the National Geodetic Vertical Datum of 1929.

The U.S. Army Corps of Engineers (USACE), Los Angeles District, conducted a water conservation study for Prado Dam in 2004, where five alternatives were evaluated. In this 2004 study, Alternative 4, which is to maximize the water conservation elevation up to 505 ft at any time of year, was selected for further evaluation. This was a favored alternative to pursue with consideration for the local sponsor’s preference, and it introduced little to no impacts with respect flood risk management regulation/operation. In this appendix, this alternative will be referred to as the “Alternative 4”.

The baseline or current approved regulation at Prado Dam is to maximize water conservation up to elevation 498.0 ft in the flood season, and up to 505.0 ft in the non-flood season (Alternative 2 of the USACE 2004 study). Flood season is defined as the period from 1 October through 28/29 February. Non-flood season is the period from 1 March through 30 September. In this appendix, the baseline condition will be referred to as “without project condition”.

In the 2004 water conservation study, the reservoir capacity rating curve was developed using 1998 survey data. In the current study, the reservoir capacity rating curve was updated based on the 2008 survey data. Inflow hydrographs were also updated since the 2004 water conservation study to best reflect existing project conditions.

In addition to the water conservation study, a sediment transport analysis was also conducted for the Santa Ana River reaches upstream and downstream of the Prado Dam.

2. STUDY AREA

The Prado Basin Study Project Management Plan provides the study area information. Prado Basin is located within the Santa Ana River watershed. The Santa Ana River drainage area that is controlled by Prado Dam regulation encompasses about 2,255 square miles of the Santa Ana River watershed, making it the largest watershed in southern California. The study area is approximately 50 square miles within which all storm water and groundwater flows drain into the Prado Basin. Four major

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 6 watercourses drain into Prado Basin: Chino Creek, Cucamonga/Mill Creek, Santa Ana River, and Temescal Creek. The Chino Creek channel drains the western boundary of the study watershed area. The Cucamonga and Day Creeks are concrete lined channels traversing the center of the study area and merging to form Mill Creek before entering the Prado Basin; the Santa Ana River is the main watercourse entering the Basin from the northeast, and Temescal Creek enters near the southern boundary of the Basin. The study area extends downstream of the Prado Dam along the Santa Ana River for approximately 8 miles. Figure 1 shows the Santa Ana River Watershed.

The study area will encompass the Prado Basin to elevation 566 ft and each of the four major watercourses to the approximate area where the channels change physical characteristics or are adjacent to another project. The Chino Creek will be included upstream from the basin to just upstream of Soquel Canyon Parkway where the creek is channelized. The study area will extend upstream along Mill Creek to approximately Hellman Avenue where the creek is channelized. The Santa Ana River portion will extend upstream to Hamner Avenue. The Temescal Creek will be included upstream to Lincoln Avenue where the creek is channelized.

Downstream of Prado Dam, the Santa Ana River continues for about 31.2 miles to the Pacific Ocean. The approximately 8 miles immediately downstream from the dam to Weir Canyon Road represents a relatively natural channel with riparian and aquatic habitat as the river meanders within the banks. The channel segment will be included in the study area. The lower portion of the river has been modified for efficient flood risk management conveyance of storm water runoff to the ocean. Figure 2 shows the Prado Basin study area.

3. BACKGROUND INFORMATION ON PRADO DAM

3.1 General

The Prado Dam was authorized by Congress in 1936 for "flood risk management and other purposes." The primary purpose is to provide flood protection to the metropolitan area of Orange County.

During times of minimal flood threat, the dam can be regulated to control runoff in order to supply water to the OCWD. The valley portion of the watershed above Prado Dam is rapidly urbanizing; consequently, average annual runoff is increasing. In 1985, a hydrology and water conservation study of Prado Reservoir was prepared. In 1988, an analysis for the operation of Prado Dam for additional water conservation was conducted. In 1990, the water control plan was revised to introduce a buffer pool, which allowed the water control manager to limit releases from the dam to facilitate OCWD’s groundwater recharge activities.

Prado Dam is currently operated in accordance with the USACE Prado Dam Interim Water Control Plan (IWCP), contained in the Prado Dam Interim Water Control Manual, dated August 2020 (2020 IWCM) (See Section 3.3 for details).

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 7 3.2 Pertinent Data

Construction of Prado Dam was completed in May 1941, as part of a general plan for the construction of flood risk management facilities in the Santa Ana River Basin. The dam is located on the Santa Ana River 30.5 miles upstream from the Pacific Ocean. For reasons of continuous and rapid urbanization of the watershed which consequently led to generating increased surface runoff, Prado Dam required modification. Authorization for the modification of Prado Dam is contained in the Water Resources Development Act of 1986, (P.L. 99-662). The purpose of this modification is to provide additional capacity for storage of floodwaters and sediment by enlarging the existing Prado Dam and Reservoir and to take advantage of increased downstream channel capacity by increasing the release capacity of the outlet works. The modification authorized by Congress is based on the plan recommended by the Los Angeles District of the USACE, as described in a document entitled, Design Memorandum No. 1, Phase II GDM on the Santa Ana River Mainstem including Santiago Creek, Volume 2 - Prado Dam, dated August 1988.

The pertinent data for Prado Dam with the Phase II modifications in place are shown in Table 1. The Prado Dam Interim Water Control Plan from the 2020 IWCM is provided on Figure 3.

Prado Dam modification per the Phase II GDM includes an earth fill embankment which rises 134 ft above the streambed, with a crest length of 3,050 ft. The reservoir formed by the dam will hold about 334,227 ac-ft of water at spillway crest elevation at 563 (based on 2015 survey), covering about 9,600 acres of land. Prado Dam modification includes raising of the existing spillway crest to elevation 563.0 ft. The newly installed reservoir outlet sill is at elevation 470.0 ft. The new outlet works structure consists of 6 main outlet gates, each 9.75 ft wide by 14.75 ft high, and two low flow 3 ft diameter steel pipe conduits. The low-flow valves alone can pass approximately 450 cfs with the water surface at elevation 505.0 ft.

3.3 Current Water Control Plan

Prado Dam is the key element in the flood risk management system protecting downstream Orange County. The downstream channel is managed by Riverside County for about the first 2.8 miles below the dam and then by Orange County Public Works, for the remaining distance to the Pacific Ocean. The dam achieves flood risk management by capturing large inflows and releasing this impoundment at rates that does not exceed available downstream channel capacity. In general, all runoff impounded behind Prado Dam are stored temporarily then drained as rapidly as possible to create storage space in preparation of any subsequent storm runoff events. During times when there are no threats of significant storm runoff, the dam can also be regulated for water conservation.

The current IWCP officially identifies a maximum flood season top of water conservation pool elevation up to 498.0 ft, and that maximum impoundment for water conservation can be further maximized up 505.0 ft, linearly allowing the impoundment to increase from elevation 498 up to 505 from March 1st to March 10th. After March 10th, the maximum temporary impoundment for water conservation is still up to 505.0 ft, however, all water impound must be evacuated by 1st of September, or sooner if possible, in order to allow maintenance at the dam and reservoir to take place during the drier months.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 8

Further, during the non-flood season, while the pool impoundment elevation remains above 498.0 ft, in accordance with the Biological Opinion and agreements with the USFWS, the dam must maintain a running average discharge of 350 cfs in order to limit the duration of inundation impacts to any sensitive habitat existing within the reservoir. The downstream OCWD recharge facility capacity can vary monthly; therefore, Attachment B of this appendix also provides a technical memorandum documenting the evaluation of the suitable discharge rate from Prado Dam and the estimated OCWD SAR diversion capacity.

In addition, the current IWCP also incorporates the approved multi-year deviation, allowing the flood season water conservation buffer pool elevation to maximize up to 505 ft. With the approved deviation in place, there currently is no seasonal variation to the maximum water conservation pool during the water year. Further, when the current Feasibility Study is complete, the proposed alternative for allowing the flood season buffer pool to go up to 505 ft will also be implemented permanently under the current IWCP. The multi-year deviation will expire in March 2023.

4. CLIMATE CHANGE CONSIDERATION

Scientific evidence shows that in some areas climate change is shifting the climatological baseline about which natural climate variability occurs. Extreme seasonal conditions of rainfall and runoff may become more common in some regions. These conditions may be intensified by future changes in the condition of native vegetation and societal demands for energy and water. Therefore US Army Corps of Engineers (USACE) projects, programs, missions, and operations must assess these potential changes to remain reliable in spite of this baseline shift. A qualitative assessment of potential climate change threats and impacts that may be potentially relevant to this study was conducted to address this issue. This qualitative assessment was performed under the guidance of ECB No. 2018-14 (USACE 2018).

4.1 Current Climate and Climate Change Trends

The coastal regions of southern California are characterized as having a Mediterranean climate with warm to hot, dry summers and mild, moderately wet winters. However, further inland where part of this study area is situated, the areas can turn more semi-arid. The study area is situated across Riverside and San Bernardino Counties. The climate in Riverside County is semi-arid with hot, dry summers and mild, relatively wet winters. The average annual precipitation in the Prado Basin averages between 11 and 19 inches annually (CDM Smith 2016). The region follows a Mediterranean seasonal precipitation pattern with most of the precipitation occurring between November to March. Due to the hills and mountains of the study region, extreme precipitation is heavily influenced and driven by atmospheric rivers which cause orographic enhancement of precipitation (USACE 2015). Atmospheric rivers are responsible for about twenty to fifty percent of the precipitation and streamflow for the study area (USACE 2015). Prado Reservoir does not typically store runoff for

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 9 extended periods of time. Impounded flows from large storm events are released to maintain flood control capacity throughout the rainy season (winter) and facilitate groundwater recharge of OCWD’s facilities. Outside the winter flood season when significant storm runoff is unlikely, dam releases can still be regulated for water conservation.

According to USACE (2015), over the historic period there has been an increasing trend in observed mean, daily minimum, daily maximum air temperature across the southwest, inclusive of the study region. A strong positive linear trend in mean air temperature has been observed over the period 1950-2000 in the winter (December-February) and spring (March-May), a lesser increase in the summer (June- August), and decreasing temperature in the fall (September-November) in the study area (Wang et al., 2008). In the third National Climate Assessment (NCA), Garfin et al. (2014) show temperatures in a recent 22-year span (1991-2012) were above the average for the period 1901-1960 by up to 2◦ F for the California Region, inclusive of this study area. Details on statistical significance were not provided in the NCA, however these trends are consistent with two separate studies. Tebaldi (2012) identified a decadal temperature increase of 0.16 F for California with a 95% confidence interval (C.I.) and a study by Hoerling et al. (2013) identified an increase in average annual daily temperature of 0.9 to 4.5 ◦ F between 1901 and 2010 with 95% C.I. Hoerling et al. (2013) also reported statistically significant (95% C.I.) increase in both minimum and maximum annually averaged daily temperature for the California Region of up to 5.4 F from 1901 to 2010. Similar increases were also documented in the most recent National Climate Assessment (Vose et al. 2017).

Several studies evaluating national precipitation trends have failed to report statistically significant trends in total annual precipitation for the study area. Grundstein (2009) found no significant trend in soil moisture index, which is a function of both precipitation and evaporation (ET) and can be used as a representation of trends in both precipitation and ET. A study by Kunkel et al. (2013) focused on the southwestern U.S. region, including the area of study, found no statistically significant trends in historical annual, seasonal or extreme precipitation from 1895 to 2011. There were also no trends found in the frequency of extreme precipitation events. In the third NCA, Walsh et al. (2014) reported annual precipitation changes from 1991-2012 compared to 1960-2012 for the California Region with areas in southern California showing a decrease in precipitation, however no details about statistical significance were provided with the report. The fourth NCA showed large percent decreases in annual precipitation for coastal southern California, but little change in average winter precipitation for the period 1986 to 2015, compared to 1901 to 1960 (Easterling et al. 2017). These studies show that to date there has been no detectable trend in winter season precipitation that might affect water supplies and stream flows in the basin.

The USACE Climate Hydrology Assessment Tool was used to examine observed streamflow trends in the vicinity of the study area. Since the study focuses on water conservation within the Prado Dam basin, inflows from the four major contributing rivers were analyzed for trends in annual peak instantaneous streamflow as it is these flashy, major storm events that develop pools

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 10 within the reservoir which impact flood risk management, water conservation, and ecosystem restoration activities. Four gages, one for each contributing river or creek, were chosen from the set of USGS gages used for calibration in the Wasteload Allocation Model (WLAM) in the development of hydrographs for the basin (Wildermuth Environmental Inc. 2009). The hydrologic time series of annual peak instantaneous streamflows at Santa Ana River (Gage No. 11059300), Chino Creek (Gage No. 11073600), Temescal Creek (Gage No. 11072100), and Cucamonga/Mill Creek (Gage No. 11073495) are shown on Plate 1 through Plate 4. The Santa Ana River gage had a large gap in the period of record, so only the continuous period of record was analyzed. The p- value is a metric by which to measure statistical significance. Commonly, the threshold for statistical significance is the 95% confidence and a p-value of less than 0.05 is considered an indication of a statistically significant trend. All four gages show a general trend of increasing peak streamflow, however only Cucamonga Creek (p-value=0.0008) had a statistically significant trend. The remaining gages had no statistically significant trends as indicated by high p-values, Santa Ana River (p-value =0.46), Chino Creek (p-value=0.19), and Temescal Creek (p-value=0.29). Although each stream’s subwatershed would be subject to similar large-scale hydrologic drivers of climate change, just one of the streams analyzed showed a significant trend in streamflow. This may be an indication that increases in flow are based on more variable localized factors, such as land development, that are not fully explained by climate change trends.

The USACE nonstationarity tool was used for each of the gages that had at least 30 years of streamflow data at the time of this climate assessment. Stationarity is the assumption that time-series hydrologic data is constant through time. However, the impact of climate change has led to circumstances under which this assumption does not hold true resulting in nonstationarity. The nonstationarity tool works best with continuous data, therefore only the longest period of continuous peak streamflow observations available for each gage was used in the analysis. The presence of non- stationarity in the peak streamflow dataset indicates that there is variance in the mean within the data that is not independent of time. There were no nonstationarities in the hydrologic time series for Santa Ana River or Temescal Creek as can be seen in Plates 5 and 7, respectively. The tool did identify nonstationarities or change-points in the mean of time series data for Chino Creek and Cucamonga Creek. Plates 6 and 8 show that for Chino Creek and Cucamonga Creek, respectively, there is a change of the mean annual peak streamflow within different segments of the time series data. The Chino Creek analysis shows a noticeable shift in mean during the years of 1986-1990. This coincides with a document period of drought within southern California and the data is likely reflecting this temporary shift in conditions. No issues with the instrumentation or data were noted in the gage record that would explain the distinct shift in flows at Chino Creek. A similar decrease in streamflows was observed during this period in the other stream gages and Cucamonga Creek has a non- stationarity indicated in mean flow at the end of this drought period as well. It’s possible that because the other streams lacked sufficient period of record before or after this period the tool could not detect a non-stationarity due to the drought. Also, each stream has its headwaters in distinct parts of the

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 11 upper watershed and more localized variability in these portions of the watershed may cause differences in the trends noted across the streamgages.Periods of drought are associated with or driven by increases in temperature and more extreme precipitation, which are indicators of a changing climate. These temperature and precipitation extremes can have impacts to water supplies and vegetation (Crocket and Westerling 2018). The potential impacts to this project’s business lines are discussed in Sec. 4.2. The heatmap graphical representation shows under which method a non- stationarity was detected and the color of the bar indicates the corresponding statistical parameter the statistically significant change was detected. Chino Creek time series data was determined to have no statistically significant trend. Trend analysis (not shown) of the Cucamonga Creek time series data indicate a statistically significant positive trend of the mean annual peak streamflow.

The National Oceanic and Atmospheric Administration (NOAA) National Environmental Satellite, Data and Information Services (NESDIS) released a report in January 2013 (NOAA 2013) assessing climate trends under high and low emission scenarios into the next 50-100 years for the southwest region. The report indicates that over the period of record (1895-2011) for the southwest region, annual temperatures have generally increased over the past 115 years, however, annual precipitation shows no long-term trend. To account for climate change, the projected meteorological conditions in the region considers the past temperature and precipitation records, as well as the modeled future conditions in the area through 2070. Under the high emissions scenario, a warming trend and increase of days above 95°F and a precipitation trend towards slightly drier conditions in areas of California prone to low precipitation can be expected over the next 50 years. However, these estimates have significant uncertainty and spatial variation.

4.2 Projected Climate Change and Business Line Impacts

Future changes in streamflow should be considered as they have potential impact to not only the Flood Risk Reduction business line, but also Ecosystem Restoration and Water Supply business lines, which are of concern for this project. Studies by Tebaldi et al. (2006) and Wang and Zhang (2008) forecasted increases in the intensity and the number of high-volume storm events by the end of the 21st century for the California Region. A climate change analysis was conducted for the Santa Ana River Watershed as part of the U.S. Bureau of Reclamation Santa Ana River Watershed Basin Study (2013). Their summary findings include a decreasing trend of April 1st snowfall water equivalent (SWE) for water years 1950 -2099 due to warming across the region. SWE on April 1st is used as a proxy for subsequent spring-summer runoff conditions, however the water conservation activities of Prado basin rely heavily on winter precipitation events. A reduction in spring and summer runoff conditions is unlikely to have a major impact on water conservation efforts as large pools are typically not generated during these times of the year. The total annual runoff volume follows the long-term declining trend pattern similar to that of precipitation. The study also reported slight decreases to total runoff in the December-March winter season and the April-July summer season over the simulated years 2050-2099. While decreases in annual runoff may have

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 12 some negative impact on water conservation efforts, the trends of increased magnitude and frequency of peak flows (which predominantly occur during the winter) bodes well for continued generation of impoundments that can be regulated for flood risk management and water conservation.

The USACE Climate Hydrology Assessment tool was used to identify projected trends in annual maximum monthly stream flow for the overall Southern California Coastal Watershed in support of this qualitative assessment. Plate 9 displays the range of the forecast annual maximum monthly stream flows computed by 93 different hydrologic climate models for the period of 2018-2099. The overall projected trend in annual maximum monthly streamflow generally increases overtime as shown in Plate 10, however, this increase is not statistically significant as indicated by the high p- value (p-value = 0.63). This finding suggests that there is no statistically significant change in flood risk in the future in the study area relative to the current time. While other studies (US Bureau of Reclamation 2013 and USACE 2015) indicate decrease precipitation and SWE, an increase of streamflow by the Climate Hydrology Assessment over the larger Southern California Coastal area may be indicative of vulnerabilities associated with urban development, as discussed below.

The Fourth California Climate Assessment (2018) report and toolset were used to understand the projected changes to relevant hydrologic factors in the study area. While model projections show very little change in either direction in the average precipitation, extreme events are projected to increase. The Assessment analyzes meteorological trends under two scenarios. The first scenario is RCP 4.5 scenario in which emissions peak around 2040, then decline and the second is RCP 8.5 in which emissions to rise strongly through 2050 and then stabilize around 2100. Under RCP8.5 the wettest day of the year is projected to increase by up to 30% and the frequency of extreme dry years will increase by double or more by year 2100. Extreme events in the region are driven by atmospheric rivers. Global climate models project up to a 40% increase in the precipitation quantities of atmospheric rivers under RCP8.5 and up to a doubling of days the region experiences such conditions. Plate 11 shows the results of 4 different models with respect to ranges of 100-year 24 hour rainfall totals at two future time periods under RCP85. In this analysis the extreme precipitation events are days during a water year (Oct–Sep) with 1-day rainfall totals above an extreme threshold of 0.37 inches. GCMs project drier soils due to an 80% chance of a decades-long drought between 2050 and 2099. Plate 12 shows the trends in minimum and maximum temperature associated with a late century dry spell from 2051–2070. Droughts associated with increased ambient temperatures may lead to loss of non-drought resistant vegetation which is important for sediment stabilization. Similarly, increased wildfire risk associated with higher temperatures, can lead to increase debris and sediment loads during rainfall events occurring over burn scarred areas. Major winter storms following periods of drought or intense fire seasons are of particular concern for sediment management. Furthermore, increases in temperature is associated with decreased SWE as precipitation is less likely to fall as snow at lower elevations. Although the contribution of SWE to water supply are minor compared to rainfall runoff during winter storms, the impact due to increased temperature is worth noting.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 13 The USACE Watershed Vulnerability Assessment Tool was used to analyze the vulnerability of the study area in impacts to the Flood Risk Management business line (Plate 13), the Ecosystem Restoration business line (Plate 14) and the Water Supply business line (Plate 15). This study area is within the Southern California Coastal Basin (HUC 1807). For this basin, the tool projects a high vulnerability to flood risk throughout the 21st century (epochs 2050 and 2085) in both wet and dry periods. A breakdown of the flood vulnerability indicators for wet and dry periods in each epoch are shown in Plate 13. For all four scenarios, the biggest contributing indicator of an increase in flood vulnerability was acres of urban area within the 500-year floodplain. This indicates that the approximately half the vulnerability is an increase of urbanization and impervious area within the floodplain of the study area leading to increased runoff volumes. The Runoff to Precipitation elasticity indicator measures the percent change in runoff divided by percent change in precipitation and indicates the watershed may be vulnerable to runoff being more sensitive to small increases in precipitation in comparison with other watersheds which in this case is largely driven by increased urbanization as well. Beyond urbanization, the remaining indicators of Flood Magnification (local and cumulative) and Annual Variability of Runoff are driven by changes in hydrologic and meteorological drivers. Flood magnification indicates that there is a vulnerability to flood flow increases within the watershed (local) and in upstream watersheds (cumulative) likely directly related to projections of increase in number and duration of atmospheric rivers bringing extreme events to the region. The annual variability is a measure of how much or little natural water reliability is to be expected from a watershed. As discussed previously, GCMs project increased periods of droughts punctuated with extreme events creating a more variable precipitation and runoff system. Despite being an indication of extreme events impacting flood control and flood risk management efforts, this type of variability may assist water conservation (and thus ecosystem restoration) efforts by allowing more frequent impoundment of water from an increase in winter storm events.. While this does exacerbate flood risk, a large impoundment of water at the end of the flood season may improve ecosystem restoration prospects by providing a source of water later in the season.

While drivers of vulnerability to flood risk management have an impact on other business lines, the tool also independently projects vulnerability to the ecosystem restoration and water conservation business lines for epochs 2050 and 2085 for both wet and dry periods as well. The dominant indicator for ecosystem restoration in all scenarios is the percentage of freshwater plant communities at risk of extinction. Reduced flow rates and increased sedimentation in streams are possible causes of high values in this category. The second highest indicator is monthly coefficient of variation (CV) of runoff. This high indicator value points to high short-term variability in monthly runoff within a year. Both of these indicators are closely aligned with the flood risk management business line indicator of annual variability of natural water supply discussed above.

The projections for vulnerability of the Water Supply Business Line for wet and dry periods for 2050 and 2085 show the dominant indicator to be the Annual or Monthly Variation of Unregulated Runoff. The next indicator is the Change in sediment load due to change in future precipitation.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 14 Sediment load increasing due to storm events triggered by extreme precipitation has the potential to increase infilling within the reservoir which would negatively impact water supply.

With respect to future projections, the strongest signal for change in the study area’s hydrology comes as a result of increased extreme precipitation levels and associated meteorological events. The changes in frequency of extreme events and the previously indicated system vulnerability to increased urbanization will likely lead to an increase in higher peak flows during the flood season, and although this may provide an impoundment later in the season, it does not indicate water supplies to sustain the a non-flood season pool will be available into and during the dry summer months. Extreme precipitation in the study area is driven by atmospheric rivers in this region. While there is analysis to support an increasing trend in the number of atmospheric events, that predicted increase is not well quantifiable to support a quantitative assessment of its impacts on this project’s business lines. There is also a strong consensus amongst the literature that supports a trend of increasing temperatures which poses impacts through increased sediment loads due to loss of non-drought resistant vegetation and greater variability in flows which impacts the ability to maintain water levels. Based on this climate assessment, the recommendation is to treat the potential effects of climate change as occurring within the uncertainty range calculated for the current hydrologic analysis. Table 2 shows a qualitative summary of identified climate risks associated with plan measures.

5. DESCRIPTION OF WATER CONSERVATION TERMINOLOGY

5.1. Yield

Yield as used in this study is, the total volume of water delivered to the OCWD Spreading Grounds on an annual basis. This definition is consistent with Reservoir Yield as defined in the "Glossary of Hydrology", by Shuh-shiaw Lo, dated 1992 and distributed by the Water Resources Publications, which says "Reservoir Yield is the amount of water which can be supplied from the reservoir in a specified interval of time". The yield for Present Conditions is an estimate of the average annual volume of water presently diverted to the OCWD spreading grounds. The alternative operations evaluated in this study allow water to be held at Prado Dam (up to a certain elevation) until it can be accommodated by the downstream spreading grounds, which effectively increases the average annual yield. While the water surface is below the designated pool elevation, reservoir releases are limited to the intake capacity of the downstream spreading facilities.

5.2. Water “Lost”

Water "Lost" as presented in this Appendix is simply the difference between the average annual flow in the Santa Ana River above the Orange County spreading facilities minus the average annual yield for the downstream spreading grounds. In the hydrologic models, it is shown as the volume of water which passes by the spreading grounds diversion. The volume of water actually “lost” to the Pacific Ocean is affected by other factors en route such as evaporation, percolation, etc, mainly as a result of flood events and flood risk management discharges from Prado Dam that exceed intake

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 15 capacity at the spreading grounds.

5.3. Present Conditions

Present Conditions for the modeling purposes presented in this report are for water year 2021 in the watershed, which include urbanization estimated for the year 2021 and inflow from upstream reclamation plants. Present conditions assume that modifications to Prado Dam have been completed with respect to the raised main embankment, new larger capacity outlet works, but still with an unraised spillway structure. The elevation-area-capacity used in the models is based on the 2008 survey for Prado Reservoir. The Present Conditions long-term downstream capacity for daily spreading operations by OCWD will be fixed at 350 cfs, which is a value based on historical capacity of the spreading basins to take flow. Further, the plan of operation in this model for the Present Conditions will be without having an approved multi-year deviation in place to allow the water conservation buffer pool to go up to elevation 505 during the flood season.

5.4. Future Conditions

Future Conditions are for water year 2071. Urbanization adjustments were made to the daily inflows to represent changes over the 50-year project life. Future Conditions yields presented in this appendix include the effects of sedimentation. The sediment yield for Prado Dam was estimated to be 0.75 ac- ft per mi2 per year. Thus, for Future Conditions (WY 2071; 50 years) the sediment accumulation would be about 35,000 ac-ft. This volume was distributed throughout the reservoir up to the future spillway crest (el. 563.0 ft) and the elevation-area-capacity relationships were modified accordingly. The Future Conditions long-term downstream capacity for daily spreading operations by OCWD will be fixed at 350 cfs. It is unlikely that OCWD can realistically increase this capacity due to the infiltration of available lands for infiltration basins.

5.5. Debris Pool

Before flood risk management discharge can be made from the dam, the debris pool (up to elevation 490.0 ft) is built which serves to settle out excess debris and sediment within the reservoir upstream, preventing them from entering and plugging up the outlet works. The debris pool is generally built at the start of a storm runoff event, and upon passing of the storm runoff event, the debris pool is completely drained. Discharge from the debris pool impoundment can be coordinated with OCWD to accommodate the downstream diversion capacity for their recharge facilities (up to 600 ft³/s).

5.6. Buffer Pool

The “buffer pool” is a defined temporary storage behind the dam for water conservation benefits. During the implementation of the current IWCP, the buffer pool elevation ranges from 490 ft, up to 505 ft. If runoff forecast indicates that an impending storm will bring significant inflow into the

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 16 reservoir, the Buffer Pool will be drained enough to account for the estimated inflow volume, or even drained entirely back down to the top of debris pool elevation of 490.0 ft, if needed. The Buffer Pool is within the District’s complete operational discretion to benefit flood risk management objectives at any time during the water year.

6. LOWER SANTA ANA RIVER REACHES

The Santa Ana River basin below Prado Dam comprises about 208 mi², excluding about 19 mi² tributary to Carbon Canyon Creek above Carbon Canyon Dam. The lower Santa Ana River flows about 30.5 miles from Prado Dam through Santa Ana Canyon and the cities of Yorba Linda, Anaheim, Orange, Santa Ana, Fountain Valley, Costa Mesa, and Huntington Beach before emptying into the Pacific Ocean.

The lower Santa Ana River channel can be divided into three geomorphic reaches. The first channel reach commences near Prado Dam and ends at North Weir Canyon Road. This reach is referred as canyon reach or Reach 9. The USACE is currently constructing bank protection along Reach 9.

The second geomorphic unit is the groundwater recharge reach, extending from North Weir Canyon Road to the Garden Grove Freeway (22 Freeway). This section has a natural bed, but its banks have been significantly modified. The river in this reach is contained within a regular, trapezoidal channel with a bottom width of about 325 ft (99.1 m). Several drop and grade control structures were constructed in this reach to help maintain the river slope and prevent undue degradation. Downstream of Imperial Highway, temporary levees are regularly constructed, maintained and modified by OCWD within the prismatic flood risk management channel to enhance infiltration of water into the river bed in the groundwater recharge reach. These temporary structures are designed to maximize wetted area in the river bed for percolation. These structures are designed to self-level during high flows, thereby preserving the full capacity of the SAR channel for flood risk management purposes. Flow diversion structures located in this river reach divert flows to adjacent infiltration ponds

The final geomorphic reach is located between the downstream end of the recharge reach (downstream of the Garden Grove Freeway) and the Pacific Ocean. This reach has a mild longitudinal slope and has been converted to a trapezoidal shape. A portion of the channel is concrete lined. The first stretch of this reach runs through a golf course and appears to have a somewhat natural shape. A hardscape (grouted stone) channel exists under the golf course fill. Below the golf course (just above 17th Street) the channel is fully concrete-lined to just above Adams Avenue. At the lower channel portion (below Adams Avenue) it has a natural river bed with concrete lined earthen levees and flood walls and is subjected to tidal fluctuations. It is therefore prone to periodic sediment deposition depending on flood flows and tidal patterns (Golder 2009).

The 8 miles immediately downstream of the dam (Reach 9) is included in the study area. The lower portion of the river has been modified for efficient flood risk management to the ocean.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 17 7. RESERVOIR SIMULATION

7.1. Representative Period of Record

Runoff records for inflow into Prado Dam exist for the period 1920-present. Determining a representative period of record facilitated data management and presented a more reasonable picture of current conditions. Establishing the representative period of record required the examination of long-term records of annual totals for precipitation gages and annual volumes for stream gages located in various areas above Prado Dam. In the 2004 Prado Water Conservation Study by COE, the representative period of record was selected using 1949 through 1988. This period includes both wet and dry weather conditions of the Prado Basin. In this study, the representative period of record was also selected using 1949 through 1988. It represents the characteristic inflow condition for the Prado Dam.

Average annual rainfall data from 1950 to 2014 for downtown Los Angeles is shown as Diagram A. This diagram demonstrates the rainfall pattern of the past 64 years in Southern California. As shown in this diagram, 1949 through 1988 can serve as a representative period. Additional data from 1989 to 2014 does not change the characteristics of the Prado dam inflow hydrograph.

Los Angeles Average Rainfall (1950-2014) ■ Jufy l-J u, e3 0Ave , .;• R.ain f 1

40

35

30

10 - I I I I - .I .I I. I .I .I I I

Diagram A

7.2. Flood Frequency

One of the major premises of the water conservation study is that modifications to the operating plan for water conservation at Prado Dam will not have any significant impact on flood risk management, i.e., will not significantly decrease the level-of-protection afforded by the dam. If the Buffer Pool is not evacuated prior to a major storm, there is minimal impact on flood risk management, as the entire buffer pool storage accounts for less than 10% of the available flood risk management storage at

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 18 Prado Dam. To evaluate this, the reservoir design flood (RDF), which is based on the Standard Project Flood (SPF), was routed through the reservoir under several different scenarios. The RDF for Prado Dam is equal to 92% of the SPF and is based on a 4-day general storm with a runoff volume of 415,800 ac-ft. The 2-year to 1,000-year frequency flood routings were also evaluated. One of the worst and most remote scenarios which could impact on flood risk management would be to have no advanced warning (no forecast) of a RDF at Prado Dam when water is being held to the top of the Buffer Pool. Flood risk management releases did not begin until the water surface exceeded the Buffer Pool elevation.

7.3. Data Collection

The inflow data downstream of Prado Dam and the evaporation data were collected from the COE 2005 Water Con Study database.

7.4. Inflow Adjustments and Projections

As mentioned in Section 7.1, the representative period of record used for the simulation analysis is from 1949 through 1988. The mean daily flows for the representative period of record required adjustments in order to properly simulate flows for Present (WY 2021) and Future (WY 2071) Conditions. The adjustments are mainly for the future land use and the wastewater discharge projections in the watershed. Wildermuth Environmental Inc. conducted the data adjustment and projections for OCWD and provided the daily inflow hydrographs for the study.

Table 3 presents the adjusted annual inflow of 39 years period of record adjusted to watershed for the years 2021 and 2071. The table shows the total inflow for the years 2021 and 2071 are 8,457,605 acre-ft and 6,681,353 acre-ft respectively. The projected year 2071 future condition inflow is much less than the year 2021 present condition inflow. The detailed presentation of the years 2021 and 2071 daily inflow analysis is shown in the Wildermuth Environmental Inc., report as attached in Attachment C.

7.5. Evaporation

Evaporation was computed by compiling average monthly pan evaporation at Lake Mathews and the average monthly precipitation at Corona from Riverside County's "Hydrologic Data for 1979-80, 1980-81, 1981-82 Seasons". Each station was selected for its proximity to Prado Dam. Using the net evaporation equation for reservoirs:

Net Reservoir Evaporation = 0.70 (EVAP - PRECIP)

where;

0.70 = the Pan Coefficient for this region; EVAP = Average Monthly Pan Evaporation; and, PRECIP = Average Monthly Precipitation All units presented are expressed in inches per month.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 19

The average monthly evaporation rates for Prado Dam used in the HEC-5 models are presented in Table 4.

7.6. Elevation-Area-Storage Capacity Curve.

The elevation-area-storage capacity curves were developed for the present and future conditions using the Prado Dam 2008 survey and the Empirical Area-Reduction Method in EM 1110-2-4000. This method was developed by Borland and Miller in 1958 for the Bureau of Reclamation. It is recommended for use under extreme reservoir operation conditions or when the reservoir shape is unusual. EM 1110-2-4000 does not provide a definition of “extreme operation conditions.” However, in comparison with many dams which have large conservation pools, any single purpose flood risk management reservoir which is nearly dry at the beginning of an event, then fills rapidly and then drawn down until it is nearly dry again as quickly as possible, could probably be characterized as having extreme operation conditions. Based upon this reason and that this method has been used in previous studies for Prado Dam, this method was selected to distribute sediment in the Prado Reservoir. The same assumptions utilized in the distribution of the 100-year sediment allowance for the Phase II GDM were used in the distribution computations of this study.

Using the 2008 survey, a sediment deposition volume of 9,100 ac-ft (700 ac-ft per year) for present condition and 44,100 ac-ft (700 ac-ft per year) for the future condition were distributed throughout the reservoir area up to the spillway crest, elevation 563 ft. The elevation-storage capacity data for Present conditions is shown in Table 5, and the elevation-storage capacity data for future condition is shown in Table 6.

7.7. Present Condition: Flood Season Water Conservation to Elevation 498.0 ft plus Non-Flood Season Water Conservation to Elevation 505.0 ft. (no deviation in place)

When inflow to Prado Dam is greater than the percolation capacity of the downstream spreading grounds, the existing Debris Pool (elev. 490.0 ft) can be used to provide incidental water conservation any time during the year. The existing operation schedule calls for controlled releases up to 600 cfs until the reservoir reaches elevation 490.0 ft. Also, during the flood season, it allows encroachment into the Flood-Control Pool up to elevation 498.0 ft (top of Buffer Pool) for water conservation purposes when weather conditions are favorable. When the threat of unfavorable weather is forecast, the reservoir will be drawn down enough (490.0 ft, if necessary) to accommodate the anticipated inflow volume from the storm(s) to ensure there is storage available for flood risk management operations.

During the non-flood season, water can be held up to elevation 505.0 ft (top of Seasonal Pool) for water conservation purposes. Beginning the 1st of March, the maximum allowable water surface elevation for conservation is increased from elevation 498.0 ft to elevation 505.0 ft on the 10th of March. The pool may be maintained as high as elevation 505.0 ft until the 30th of September. However, if maintenance is required, the reservoir must be evacuated before the 1st of September. If summer flood runoff occurs in the month of September, the dam can be operated for water

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 20 conservation up to elevation 505.0 ft, provided that the impoundment does not interfere with maintenance requirements.

Releases from Prado Dam, during water conservation operations, will be based on the estimated rate that the downstream spreading channel can percolate. If hydrologic forecasts and reservoir conditions indicate that the water surface elevation will exceed elevation 505.0 ft, the reservoir is put in full flood risk management mode and outflows can be made up to 10,000 ft3/s.

7.8. Alternative 4: Water Conservation to Elevation up to 505.0 ft during both flood and non- flood seasons.

As mentioned in the Introduction of this appendix, Alternative 4 of the USACE 2004 study was selected for further evaluation. When inflow to Prado Dam is greater than the percolation capacity of the downstream spreading grounds, the current Debris Pool (elev. 490.0 ft) can be used to provide incidental water conservation any time during the year. The existing operation schedule calls for controlled releases up to 600 cfs until the reservoir reaches elevation 490.0 ft. According to memorandum dated 5 February 2013 (Attachment B), OCWD stated that they conserve approximately 350 cfs in average.

Also, during the flood season allow maximizing the water conservation Buffer Pool up to elevation 505.0 ft for water conservation purposes when weather conditions are favorable. When the threat of unfavorable weather is forecast, the reservoir will be drawn down enough (490.0 ft, if necessary) to accommodate the anticipated inflow volume from the storm(s) to ensure there is storage available for flood risk management operations.

During the non-flood season, water can also be maximized up to elevation 505.0 ft for water conservation purposes. The pool may be maximized up to elevation 505.0 ft until the 31st of August. However, if maintenance is required, the reservoir must be evacuated by or before the 1st of September. If summer flood runoff occurs in the month of September, the dam can still be operated for water conservation up to elevation 505.0 ft, provided that the impoundment does not interfere with flood risk management objectives or maintenance requirements.

Releases from Prado Dam, during water conservation operations, considers OCWD’s diversion capacity into their spreading grounds. If hydrologic forecasts and reservoir conditions indicate that the water surface elevation will exceed elevation 505.0 ft, the reservoir regulation transitions into flood risk management mode and outflows can be made up to the downstream channel capacity of the Santa Ana River (up to 10,000 ft³/s).

7.9. Sediment

The average annual sediment deposition rate assumed for this study is based on the rate determined for the modified Prado Dam. As documented in Santa Ana River, Design Memorandum No 1, Phase II GDM on the Santa Ana River Mainstem including Santiago Creek, Volume 2 – Prado Dam, August 1988, the sediment allocation for the authorized plan of the modified Prado Dam is 70,000 acre-ft. This allocation corresponds to a design annual deposition rate of 700 acre-ft per year based

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 21 on the project life being 100 years. This rate was determined by multiplying the average annual sediment yield of 0.75 acre-ft per year per square mile by the future sediment-producing area of 935 square miles, which takes into account the future urbanization of the basin and Seven Oaks Dam.

7.10. Analysis Overview

The operation for Prado Dam was simulated using the HEC-5 computer program (Hydrologic Engineering Center’s Simulation of Flood Risk Management and Conservation Systems). The current approved operation plan for the reservoir was modeled and modified for Alternative 4 to account for the improved outlets and downstream channel. The 2008 survey was used to estimate available storage and area. Inflows for a representative period of record were adjusted to account for urbanization and wastewater effluent. The HEC-5 model was used to estimate the water conservation yields at the downstream spreading facilities.

The model also includes local inflow from the area between Prado Dam and the downstream spreading facilities. These flows were also established for the representative period. There were no adjustments to Present Conditions inflows, including the area downstream from Prado. The Future Condition flow required adjustments for urbanization, effluent, and sediment deposition.

7.11. Results

The Present Conditions average annual yield for Prado Dam which was used for alternative comparison is 151,000 ac-ft. This is an estimate of the volume of surface water spread in the downstream OCWD spreading grounds. It includes releases from Prado Dam and local flow from the intervening area between Prado Dam and the downstream spreading facilities. The average volume of water “lost” to the Pacific Ocean for Present Conditions on an annual basis is approximately 73,200 ac-ft. These results are presented in Table 7. The Future without project Conditions yield is 93,400 ac-ft with 86,200 ac-ft “lost” to the ocean. These results are presented in Table 9. The present with project simulation results and the future with project simulation results are presented in Table 8 and Table 10 respectively. Table 11 presents the comparisons between with and without project condition results. For the present condition, the with-project condition will yield 6,000 acre-ft per year more than the without-project condition. For the future condition, the with- project condition will yield 11,600 acre-ft per year more than the without-project condition.

Frequency-based days of inundation (duration) were estimated for selected pool elevations. Frequency-duration pairs were computed using the following process:

. HEC-5 computed reservoir elevations (elevations) were sorted from highest to lowest for each water year. . 20 duration-elevation pairs were sampled from each water year ranging from one to 360 days of inundation. . The sampled duration-elevation pairs were transposed and the elevations were ranked from highest to lowest for each duration followed by the assignment of an exceedance value based on the Chegodayev plotting position formula. Given that there are only 39 years of data, the

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 22 lowest exceedance value attained is 0.018, which corresponds to a 56-year event. . Curve-fitting was applied to known frequency-elevation pairs for the purpose of extrapolating to the 100-year event for each sampled duration. Due to the magnitude of the extrapolation (from a 56- to 100-year event), the extension of the fitted curves produced results that have a high degree of uncertainty and were not always consistent with the expected trend of decreasing pool elevations, which generally decrease with increasing duration. To limit the ambiguity of this process, the same curve type was applied to each duration and for each alternative. Where an extrapolated elevation for a given duration was not physically possible, interpolation was used. While applying consistent curve fitting across all durations for all alternatives does not reduce the amount of uncertainty linked to the results of an alternative by itself, it does lend more meaning to the comparative results between two alternatives.

Table 12 presents the inundation duration analysis results for the present without-project condition. Table 13 presents the inundation duration analysis results for the present with-project condition and the increase of inundation from the present without-project condition. Table 14 presents the inundation duration analysis results for the future without-project condition. Table 15 presents the inundation duration analysis results for the future with-project condition and the increase of inundation from the future without-project condition.

7.12. Sensitivity Analysis

The water conservation simulation modeling with HEC-5 model presented above was assumed with fixed 350 cfs recharge and 350 cfs minimum release from Prado Dam. However, both of the minimum Prado Dam release and the spreading ground recharge rate may be variable. To evaluate the effect from different Prado Dam minimum release rate and the downstream spreading ground recharge rate, sensitivity model runs were conducted and are presented in this section.

In the sensitivity model runs, the minimum Prado Dam release is 450 cfs and the spreading ground recharge rate is as follows.

Month Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Recharge(cfs) 450 450 450 450 450 420 400 350 350 450 450 450

The simulation results are presented in Tables 16 through 24. These tables are in the same format as Tables 7 through 15.

Table 11 presents the benefit of the project from 350 cfs fixed recharge. Table 20 presents the benefit of the project from 450 cfs and variable recharge. A comparison of these two tables shows that the slightly higher recharge rate produces a slightly higher benefit.

7.13. Flood Risk Due To Project Alternative

The current Prado Dam operation is to keep water level at 498 ft during the flood season and to raise to elevation 505 ft for the non-flood season. The project alternative is to keep water level at 505 ft year round. The increased volume for the project condition is 10,500 acre-ft. The current dam

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 23 capacity at spillway crest is 186,700 acre-ft. Therefore, the increased volume for the project condition is 5.6% of the total reservoir capacity. After Prado Dam modification per the Phase II GDM is completed, the dam capacity will be about 335,409 ac-ft (based on 2008 survey) at spillway crest. Then, the increased volume for the project condition will be only 3.1% of the total reservoir capacity. The current Prado Reservoir operation is that when inclement weather is forecast, the reservoir will be drawn down to make space for the anticipated inflow, ensuring there is adequate storage available for the flood risk management operations. In addition, ResReg Section will increase monitoring to ensure safety due to the deviation. With the facts presented above, we conclude that the upstream and downstream increase in flood risk to operate the conservation pool at elevation 505 year round is very minimal.

8. SEDIMENT TRANSPORT – UPPER SANTA ANA RIVER (USAR)

Sediment transport modeling of Upper Santa Ana River (USAR) was performed by Golder Associates Inc. (Golder) using the HEC-RAS model (version 4.1) for the following two purposes:

. To analyze the effects of creating a sediment trap and increasing the slope of the channel into the sediment trap using a transition channel on sediment deposition/scour and bed grain size.

. To evaluate the effects of increasing the reservoir water conservation elevation during flood season (October through February) from 498 feet NGVD 29 to 505 feet NGVD 29 on sediment deposition/scour and bed grain size.

The extent of the sediment transport modeling is from the San Bernardino County/Riverside County line on the upstream end and Prado Basin on the downstream end for approximately 20 miles (See Figure 4). The downstream end of the model corresponds to the downstream end of the transition channel.

8.1 Effects of Sediment trap and transition channel

To analyze the effects of creating a sediment trap and increasing the slope of the channel into the trap using a transition channel, two scenarios were modeled for comparison; 1) an existing conditions scenario (“base case scenario”) and 2) a scenario with a dredged sediment trap and realigned transition channel into the trap (“trap scenario”). The model was run for a 10-year period for both base case and trap scenarios.

Various sediment trap scenarios were considered and modeled for the feasibility project. The current trap scenario modeled is shown in Figure 4. The transition channel modeled has a slope of 0.33 percent, 80-foot bottom width, 4H:1V side slopes, and an 18.5-foot depth. The design has since been changed to include 5H:1V side slopes. However, the change in side slopes does not seem to have an appreciable effect on the results within the transition channel. The model does not include the full

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 24 sediment trap. Instead, considering that the sediment trap bottom elevation is sufficiently low as to be influenced by the reservoir water surface elevation, the reservoir water surface elevation based on the current dam operation was used as the downstream boundary condition of the trap scenario; i.e., 498 feet from September through February and 505 feet from March through August. Under the base case scenario, the downstream end of the model is not influenced by the reservoir water surface elevation; therefore, a normal depth is used as the downstream boundary condition.

Information about data and assumptions used in the development of the sediment transport model are detailed in the technical memorandum by Golder (Golder 2015a).

8.1.1 Modeling Results

Figures 5 and 6 show the river bed profile change after 10-year simulation for the base case and trap scenarios, respectively. The river bed profile change for the base case scenario (Figure 5) shows areas of deposition present immediately downstream of Hamner Avenue and River Road Bridge, and immediately upstream of I-15 with a maximum deposition of about 10 feet. Areas of scour are present throughout the reach with a maximum scour of about 15 feet, between Van Buren Blvd. and I-15. The river bed profile change for the trap scenario (Figure 6) is similar to the base case scenario in the upstream portions of the modeled reach. The main differences are in the area of the transition channel and sediment trap as well as upstream for some distance. Figure 6 clearly shows that river bed degradation starting at the transition channel propagates upstream until reaching a point immediately downstream of Hamner Avenue.

Figure 7 shows the average channel slope change comparison of the base case and trap scenarios after 10-year simulation over selected four segments. The modeled reach was broken into four segments based generally on grade breaks present under existing conditions. The figure shows that the slope of the trap scenario stays fairly consistent in the upstream two segments. However, the downstream-most segment (i.e., transition channel and trap) significantly flattens, whereas the slope of the next upstream segment increases. Steepening of the next upstream segment is mainly due to an abrupt slope change from the flattened transition channel downstream to the milder river bed upstream.

Figure 8 shows the average D50 change comparison of the base case and trap scenarios after 10-year simulation over selected six segments. The six segments are the same segments that were used to input varying particle size distributions during the model setup. The figure shows an overall coarsening of the downstream four segments under the trap scenario after 10-year simulation. The base case scenario also shows a coarsening of the downstream five segments, but its degree of coarsening is not as much as the trap scenario in the downstream three segments.

8.1.2 Conclusions

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 25

Per the modeling results, the sediment trap and the transition channel will degrade and coarsen the river bed of the transition channel as well as upstream for some distance. It is predicted that the river bed degradation and coarsening stop downstream of Hamner Avenue after 10 years. However, common geomorphology knowledge dictates that the degradation should continue upstream. The rate at which it continues upstream is difficult to predict but it is possible that the degradation could eventually adversely affect piers of the upstream bridges. It is recommended that the potential for upstream migration be investigated further prior to implementation of the trap and transition channel.

8.2 Effects of Increasing Reservoir Water Conservation Elevation

To evaluate the effects of increasing the reservoir elevation during flood season, two scenarios were modeled for comparison; 1) a scenario representing the existing reservoir operations - water conservation level of 505 feet from March through August and 498 feet from September through February (“1st scenario”) and 2) a scenario representing the proposed modification of the water conservation elevation - water conservation level of 505 feet from October through August and 498 feet during September for maintenance purposes (“2nd scenario”). The model was run for a 10-year period for both scenarios.

The 1st and 2nd scenarios are different only by the downstream boundary condition within the reservoir. The reservoir response to the modeled incoming flow (October 1, 1975 to September 30, 1985) was modeled by OCWD in their recharge facilities model with a varied water conservation level (498 or 505 feet) and the resulting reservoir elevation was used as the downstream boundary condition for both scenarios.

Information about data and assumptions used in the development of the sediment transport model are detailed in the technical memorandum by Golder (Golder 2015b).

8.2.1 Modeling Results

Figures 9 and 10 show the river bed profile change after 10-year simulation for the 1st and 2nd scenarios, respectively. The river bed profile change for the 1st scenario (Figure 9) shows that areas of deposition are present upstream of Market Street, upstream of the railroad bridge, and consistently between River Road and I-15 Bridge. These areas exhibit aggradation after 10-year simulation ranging between 5 and 10 feet. The river bed profile change for the 2nd scenario (Figure 10) is very similar to the 1st scenario. Because the downstream end of the reach is only seldom influenced by the change in reservoir operation rules, the difference between the 1st and 2nd scenarios in river bed profile is minimal.

8.2.2 Conclusions

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 26

The modeling results show that increasing the reservoir water conservation elevation does not significantly affect deposition or scour pattern along the modeling reach. Deposition occurs at the downstream end of the reach for both scenarios, when higher reservoir water surface elevations are encountered. When water surface elevation decreases, the deposition is reduced for both scenarios.

9. SEDIMENT TRANSPORT – LOWER SANTA ANA RIVER (LSAR)

The removed sediment from upstream of Prado dam as a result of creating sediment traps would be eventually re-entrained in the flow released from Prado dam and transported to the Lower Santa Ana River (LSAR). Golder (Golder 2017) studied the effects of this sediment re-entrainment on the long- term sediment transport through Reach 9 and entire reach of the LSAR.

Golder initially performed a slope analysis to estimate the slope change that will develop in Reach 9 under varied flow conditions, re-entrainment concentration, and re-entrained sediment size. Following the slope analysis, 10 years of sediment re-entrainment was simulated using the HEC- RAS model (version 5.0.3) over the entire reach of the LSAR from Prado Dam to the Pacific Ocean. Following 10 years of re-entrainment, an additional 5-year period was modeled with only the existing sediment load through the dam to simulate how the deposited sediment changes if re-entrainment stops.

9.1 Re-entrainment Slope Analysis in Reach 9

Two analyses were performed to determine the effects of the sediment re-entrainment on the bed slope in Reach 9. A reach-wide, incremental flow analysis using manual calculations with Meyer- Peter Muller equation was conducted to calculate average deposition or scour based on a particular sediment gradation and concentration by weight over a range of expected flows. The results of this analysis were compared to those from the HEC-RAS model of the reach simulating the same conditions.

Figures 11 through 14 show the re-entrained sediment, deposited sediment estimated using the HEC- RAS model, and deposition manually calculated using the Meyer-Peter Muller equation for Reach 9. In the modeling, it was assumed that the maximum re-entrainment sediment load was 1 yd3/sec (or 27 cfs). Therefore, these figures show that deposited sediment generally increases with the amount being re-entrained until the re-entrainment sediment load becomes 27 cfs.

9.2 Long-term Sediment Transport Modeling in LSAR

The HEC-RAS model was used to simulate the long-term sediment re-entrainment through the LSAR. The model was set up to simulate 10 years of re-entrainment for both fine and medium sand

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 27 gradations at a 1% re-entrainment concentration by weight. Base case scenario was run without re- entrainment using the existing sediment load from the dam. Following the 10-year simulation, an additional 5-year period was simulated with no re-entrainment to study the effects of stopping re- entrainment on the deposition that had occurred. These “clear water” runs only have sediment inflow that is being released from Prado Dam. The reservoir discharge sediment load is estimated by the USACE in the GDM (USACE 1988). These simulations are summarized as follows.

Simulation Model Incoming Sediment Load Incoming Sediment Scenarios Duration Gradation

Existing flow from dam Assumed Gradation: Base Fine Sand 10 Years Fine Sand

Existing flow from dam and 1% Assumed Gradation: concentration by weight re- Fine Sand 10 + 5 1% Fine Sand entrainment load for 10 years; Years and existing flow from dam only for 5 years

Existing flow from dam and 1% Assumed Gradation: concentration by weight re- Medium Sand 10 + 5 1% Medium Sand entrainment load for years; and years existing flow from dam only for 5 years

Information about data and assumptions used in the development of the long-term sediment re- entrainment HEC-RAS model are detailed in Golder report (Golder 2017). LSAR river bed profile changes over time throughout re-entrainment are presented in Figures 15 through 17. Figure 15 (base fine sand) shows continued degradation upstream of Gypsum Canyon Road, minor aggradation/degradation changes throughout the majority of the remainder of the modeled reach, and some aggradation downstream of a grade change near the Adams Street Bridge. Figure 16 (1% fine sand) shows aggradation immediately downstream of the dam, minor aggradation between Gypsum Canyon Road and the 91 Freeway Bridge, and more pronounced deposition between the 405 Freeway Bridge and the ocean. Figure 17 (1% medium sand) shows greater deposition just downstream of the dam than even the fine sand scenario (Figure 16), little change in river profile between Gypsum Canyon Road and the 405 Freeway Bridge, and deposition between the 405 Freeway Bridge and the ocean. The deposition between the 405 Freeway Bridge and the ocean is noticeably less for the 1% medium sand scenario (Figure 17) when compared to the fine sand scenario (Figure 16) which is most likely due to the fact that more sediment had dropped out just downstream of the dam than with the fine sand simulation.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 28 Figures 18 and 19 show the results of the clear water analysis compared against the results after 10 years for 1% find sand and 1% medium sand scenarios, respectively. The change in river bed profile, which is degradation only, in 5 years of clear water flows is generally limited to the areas just downstream of the dam that showed large deposition during re-entrainment. The remainder of the modeled reach remains generally unchanged from year 10 to year 15.

9.3 Conclusions

Reversing of channel incision within the reach downstream of Prado Dam can be accomplished by re-entraining fine or medium sand downstream of Prado Dam based on the following analysis and modeling results:

. Per the re-entrainment slope analysis, deposited sediment generally increases with the amount being re-entrained as long as sediment concentration is maintained during the sediment re- entrainment.

. Per the long-term sediment transport modeling by re-entraining fine or medium sand at a 1% concentration by weight downstream of Prado Dam, the potential exists for areas of excess deposition to occur in some locations, notably just downstream of the dam. Therefore, based on these modeling results, the project team needs to adjust the amount of re-entrainment to minimize excessive sediment deposition and reduce the flood risk. Additional modeling may need to be performed to vary the re-entrainment schedule or concentration to mitigate against the excessive deposition. It should be noted that higher sediment concentration than 1% may need to be considered during higher flows to mimic the natural system.

. It has been shown that stopping re-entrainment for a period of time can reverse the effects of large amounts of deposition, even though clear water modeling for 5 years after 10-year re- entrainment shows that the change in profile (degradation) is generally limited to the areas just downstream of the dam.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 29 10. REFERENCES

CDM Smith. 2016. Santa Ana River Watershed Bacteria Monitoring Program Monitoring Plan.

Crockett, J. L., & Westerling, A. L. 2018. Greater temperature and precipitation extremes intensify Western US droughts, wildfire severity, and Sierra Nevada tree mortality. Journal of Climate, 31(1), 341-354.

Easterling, D.R., K.E. Kunkel, J.R. Arnold, T. Knutson, A.N. LeGrande, L.R. Leung, R.S. Vose, D.E. Waliser, and M.F. Wehner, 2017: Precipitation change in the United States. In: Climate Science Special Report: Fourth National Climate Assessment, Volume I [Wuebbles, D.J., D.W. Fahey, K.A. Hibbard, D.J. Dokken, B.C. Stewart, and T.K. Maycock (eds.)]. U.S. Global Change Research Program, Washington, DC, USA, pp. 207-230, doi: 10.7930/J0H993CC.

Garfin, G., G. Franco, H. Blanco, A. Comrie, P. Gonzalez, T. Piechota, R. Smyth, and R. Waskom. 2014. Ch. 20: Southwest. Climate Change Impacts in the United States: The Third National Climate Assessment, J. M. Melillo, Terese (T.C.) Richmond, and G. W. Yohe, Eds., U.S. Global Change Research Program, 462-486.

Golder Associates, Inc. 2009. Santa Ana River Bed Sediment Gradation Characterization Study: Phase 3. Prepared for the OCWD by Golder Associates, Inc., Lakewood, CO. November 30, 2009.

Golder Associates, Inc. 2015a. Technical Memorandum - Prado Feasibility Study Project – Upper Santa Ana River Initial Sediment Transport Modeling Results. Prepared for the OCWD by Golder Associates, Inc., Lakewood, CO. February 18, 2015.

Golder Associates, Inc. 2015b. Technical Memorandum - Prado Feasibility Study Project – Prado Water Level Analysis Sediment Transport Modeling Results. Prepared for the OCWD by Golder Associates, Inc., Lakewood, CO. March 26, 2015.

Golder Associates, Inc. 2017. Prado Feasibility Study – Lower Santa Ana River Sediment Transport Analysis. Prepared for the OCWD by Golder Associates, Inc., Lakewood, CO. September 27, 2017.

Grundstein A. 2009. Evaluation of Climate Change over the Continental United States Using a Moisture Index. Climatic Change 93:103-115. 2009.

Melillo, J.M., Richmond, Terese (T.C.), Yohe, G.W. eds. 2014. Climate Change Impacts in the United States: The Third National Climate Assessment. U.S. Global Change Research Program, pp. 462-486. 2014.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 30 Hoerling MP, Dettinger M, Wolter K, Lukas J, Eischeid J, Nemani R, Liebmann B, Kunkel KE. 2013. Chapter 5: Present weather and Climate: Evolving Conditions. In Garfin G JA, Merideth R, Black M, LeRoy S (eds.), National Climate Assessment (ed.). Assessment of climate change in southwest United States: a report prepared for the National Climate Assessment. Island Press, Washington, D.C., pp. pp. 74-100. 2013.

Kunkel KE, Stevens LE, Stevens SE, Sun L, Janssen E, Wuebbles D, Redmond KT, Dobson JG. 2013. NOAA Technical Report NESDIS 142-5: Regional Climate Trends and Scenarios for the U.S. National Climate Assessment. Part 5. Climate of the Southwest. U.S. National Oceanic and Atmospheric Administration (NOAA), National Environmental Satellite Data and Information Service (NESDIS), pp. 1-79. 2013.

NOAA. 2013. Regional Climate Trends and Scenarios for the U.S. National Climate Assessment: Part 5. Climate of the Southwest U.S. 2013.

Scheevel Engineering. 2015. Prado Basin Feasibility Study Reach 9 – Projected Sedimentation Trends (Final). Prepared for the OCWD by Scheevel Engineering, Anaheim, CA. January 21, 2015.

Tebaldi C. 2006. Going to the Extremes: An Intercomparison of Model-Simulated Historical and Future Changes in Extreme Events. Climate Change 79:185-211. 2006.

Tebaldi C, Adams-Smith D, Heller N. 2012. The Heat is On: U.S. Temperature Trends. Climate Central, Palo Alto, California. 2012.

U.S. Army Corps of Engineers (USACE). 1988. Santa Ana River Design Memorandum No. 1: Phase II GDM on the Santa Ana River Mainstem Including Santiago Creek, Volume 3: Lower Santa Ana River (Prado Dam to Pacific Ocean). August, 1988.

USACE. 2015. Recent US Climate Change and Hydrology Literature Applicable to US Army Corps of Engineers Missions – Water Resources Region 18, California. Civil Works Technical Report. July, 2015.

USACE. 2018. ECB No. 2018-14 Guidance for Incorporating Climate Change Impacts to Inland Hydrology in Civil Works Studies, Designs, and Projects. September, 2018.

USACE. 2017. Climate Preparedness and Resilience CoP Applications Portal: Climate Hydrology Assessment Tool, Nonstationarity Tool, and Vulnerability Assessment. URL: https://maps.crrel.usace.army.mil/projects/rcc/portal.html. Tools and Data Accessed: 8 November, 2017.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 31 Walsh J, Wuebble D, Hayhoe K, Kossin J, Kunkel K, Stephens G, Thorne P, Vose R, Wehner M, Willis J, Anderson D, Kharin V, Knutson T, Landerer F, Lenton T, Kennedy J, Somerville R. 2014. Ch 2: Our Changing Climate. Climate Change Impacts in the United States: The Third National Climate Assessment, Melillo, J.M., Richmond, Terese (T.C.), Yohe, G.W. eds, U.S.Global Change Research Program, pp. 19-67. 2014.

Wang J, Zhang X. 2008. Downscaling and projection of winter extreme daily precipitation over North America. Journal of Climate 21:923-937. 2008.

Wildermuth Environmental, Inc. 2009. 2008 Santa Ana River Wasteload Allocation Model Report. Lake Forest, CA. 2009.

Vose, R.S., D.R. Easterling, K.E. Kunkel, A.N. LeGrande, and M.F. Wehner, 2017: Temperature changes in the United States. In: Climate Science Special Report: Fourth National Climate Assessment, Volume I [Wuebbles, D.J., D.W. Fahey, K.A. Hibbard, D.J. Dokken, B.C. Stewart, and T.K. Maycock (eds.)]. U.S. Global Change Research Program, Washington, DC, USA, pp. 185-206, doi: 10.7930/J0N29V45.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 32 TABLE 1: PERTINENT DATA - WITH PHASE II MODIFICATIONS PRADO DAM AND RESERVOIR

Reservoir Elevation Debris Pool 490.0 ft Top of Water Conservation Pool 505.0 ft Flood risk managementRisk Management Pool (spillway crest) 563.0 ft Top of Dam 594.4 ft Reservoir Gross Capacity Debris Pool 4,688 ac-ft Top of Water Conservation Pool 19,500 ac-ft Spillway Crest 351,700 ac-ft Allowance for Sediment (50-year) 35,000 ac-ft Allowance for Sediment (100-year) 70,000 ac-ft Reservoir Area Debris Pool 768 ac Top of Water Conservation Pool 1,760 ac Spillway Crest 10,280 ac Dam – Type Rolled Earthfill Height above Original Streambed 134 ft Top Length 3,050 ft Top Width 30 ft Freeboard 4.5 ft Main Regulating Outlets Type of Gates Vertical Lift Number and Size of Gates 6 - 9.75' W x 14.75' H Invert Elevation 470.0 ft Max Regulated Outflow at Spillway Crest 30,000 ft3/s Low-Flow Outlets Knife Gate Valve Type of Gates 2 – 3 ft diameter Invert Elevation 470.0 ft Max Regulated Outflow at Spillway Crest 600 ft3/s Spillway Type Overflow Concrete Crest Length 1,000 ft Design Discharge 481,000 ft3/s Includes Phase II GDM modifications to Prado Dam

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 33 TABLE 2: IDENTIFIED CLIMATE RISKS ASSOCIATED WITH MEASURES

Qualitative Measures Trigger Hazard Harm Likelihood

Ecosystem Restoration Plan Measures

Reduced reservoir Increased extreme Increased sediment capacity (minor); Sediment Management precipitation following Likely loads in runoff Potential embedment drought of sensitive flora

Inability to reliably Native Plantings and Runoff variations Annual Variation in support native Riparian Edge creates variable Highly Likely Runoff vegetation during Management streamflows drought periods

Conditions less Runoff variations favorable to native Non-native aquatic Annual Variation in create variable fauna allowing out- Likely fauna management Runoff streamflows competing non-native fauna to re-establish

Water Conservation Plan Measures

Unreliable water Annual and Monthly Inability to maintain supply outside of Highly Likely Variation in Runoff conservation pool Water Conservation at rainy season 505’ WSE year round without additional Reduction of Increased extreme increased sediment sediment removal conservation storage in precipitation loads to reservoir Likely lower pool due to following droughts following drought sediment

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 34 TABLE 3: PROJECTED ANNUAL INFLOW FOR 2021 AND 2071

Inflow-2021 Inflow-2071 Year (acre-ft) (acre-ft) 1950 158031 110468 1951 132793 84895 1952 294218 250176 1953 146898 100463 1954 203256 150415 1955 156158 111243 1956 211313 166570 1957 164854 121377 1958 287917 243665 1959 128759 80022 1960 146352 101144 1961 116152 65728 1962 192853 147758 1963 165498 121316 1964 140727 95711 1965 165545 123199 1966 247658 200881 1967 296873 257355 1968 163480 117290 1969 688821 647592 1970 157234 108722 1971 150390 102192 1972 150982 102531 1973 218100 176057 1974 192296 136774 1975 163456 118182 1976 156016 109769 1977 167130 120665 1978 442187 406418 1979 251909 205139 1980 590663 547853 1981 153159 104857 1982 233848 191431 1983 409874 365979 1984 147910 101597 1985 159205 112061 1986 201271 158325 1987 136631 89623 1988 167190 125913 Total 8457605 6681353

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 35 TABLE 4: MONTHLY EVAPORATION RATES FOR PRADO DAM USED IN HEC-5 MODEL

Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep I4.00 I2.00 I1.06 I0.30 I0.64 I1.93 I3.28 I4.96 I6.03 I7.48 I7.48 I5.41 I Monthly evaporation in inches.

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 36 TABLE 5: ELEVATION-AREA-CAPACITY RELATIONSHIP PRADO DAM PRESENT CONDITIONS – YEAR 2021

Elevation Area Storage Elevation Area Storage (ft) (ac) (ac-ft) (ft) (ac) (ac-ft) 470 0 0 509 2146 24699 471 0 0 510 2224 26892 490 437 1266 511 2316 29174 491 513 1747 512 2412 31549 494 730 3646 513 2524 34027 495 802 4419 514 2643 36908 496 869 5260 515 2770 39352 498 1015 7159 516 2885 42188 499 1105 8245 517 3015 45191 500 1197 9411 518 3123 48273 501 1327 10753 519 3228 51457 502 1411 12138 520 3350 54788 503 1491 13599 525 3908 73008 504 1630 15194 530 4626 94444 505 1733 16885 540 6004 148081 506 1830 18685 550 7448 215492 507 1940 20584 560 9057 298081 508 2039 22582 563 9634 326283 Elevations in ft MSL; 1929 datum Debris Pool = 490 ft Area and Storages based on 2008 Survey Spillway Crest = 563.0 ft 2021 volumes adjusted for sedimentation Top of Dam = 594.4 ft

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 37 TABLE 6: ELEVATION-AREA-CAPACITY RELATIONSHIP PRADO DAM FUTURE CONDITIONS – YEAR 2071

Elevation Area Storage Elevation Area Storage (ft) (ac) (ac-ft) (ft) (ac) (ac-ft) 470 0 0 509 1628 14089 471 0 0 510 1705 15764 490 437 1266 511 1793 17525 491 60 25 512 1887 19375 494 260 541 513 1998 21328 495 320 838 514 2116 23683 496 380 1193 515 2243 25599 498 525 2114 516 2353 27906 499 612 2708 517 2479 30375 500 703 3380 518 2588 32922 501 828 4226 519 2695 35572 502 904 5109 520 2818 38370 503 982 6062 525 3381 53941 504 1120 7148 530 4107 72760 505 1223 8328 540 5518 121368 506 1319 9618 550 7032 184249 507 1427 11005 560 8807 263513 508 1524 12489 563 9634 291341 Elevations in ft MSL; 1929 datum Debris Pool = 490 ft Area and Storages based on 2008 Survey Spillway Crest = 563.0 ft 2071 volumes adjusted for sedimentation Top of Dam = 594.4 ft

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 38 TABLE 7: HEC-5 RESULTS - WITHOUT PROJECT – PRESENT CONDITIONS (ALT2-2021) 350 cfs Recharge Rate

Prado Dam Total Local Water Water Water Max Max Prado Inflow "Lost" Year Spread WSE Release Inflow (ac- (ac-ft) (ac-ft) (ft) (cfs) ft) (ac-ft) 1950 499.2 2810 158000 5050 147000 16300 1951 494.7 350 133000 1980 135000 0 1952 512.1 5340 294000 21200 170000 144000 1953 498.4 1430 147000 4040 145000 5760 1954 506.4 5000 203000 7840 149000 61100 1955 498.6 2300 156000 7770 150000 14000 1956 521.3 10650 211000 365 129000 82700 1957 505.9 5000 165000 4070 142000 26400 1958 508.5 5000 288000 11900 176000 122000 1959 498.4 1390 129000 0 124000 4880 1960 498.4 1650 146000 3800 139000 11300 1961 496.6 350 116000 9150 124000 1690 1962 500.1 3970 193000 7030 143000 56500 1963 504 5000 165000 3340 143000 23200 1964 499.8 650 141000 4290 146000 1270 1965 505.5 5000 166000 4380 148000 21100 1966 518.5 8850 248000 1610 142000 107000 1967 518.8 10700 297000 11400 177000 130000 1968 503.9 2240 163000 9850 154000 18900 1969 531.2 19830 689000 21700 183000 525000 1970 501.2 4720 157000 3210 139000 21200 1971 499.1 2740 150000 3420 135000 18200 1972 504.3 5000 151000 3970 122000 32700 1973 506.9 5000 218000 4000 163000 58300 1974 508.4 5000 192000 9340 144000 56900 1975 505 1390 163000 2250 159000 5370 1976 500.6 4080 156000 3650 139000 20400 1977 501.9 2970 167000 1270 151000 16900 1978 516.7 7670 442000 31900 174000 299000 1979 507.3 5000 252000 4650 173000 82700 1980 522.3 13640 591000 22200 178000 432000 1981 500.1 2800 153000 4350 139000 18500 1982 508.7 5000 234000 6840 164000 75800 1983 516 8470 410000 41300 205000 242000 1984 498.8 2190 148000 710 140000 10700 1985 500.5 3230 159000 1840 140000 20900 1986 505 5000 201000 12600 164000 48800 1987 498.7 1950 137000 4040 135000 6050 1988 498.9 2370 167000 5450 159000 12800 Average: 505.7 4760 217000 7890 151000 73200

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 39 TABLE 8: HEC-5 RESULTS – ALTERNATIVE 4 – PRESENT CONDITIONS (ALT4-2021) 350 cfs Recharge Rate

Prado Dam Total Local Water Water Water Max Max Prado Inflow "Lost" Year Spread WSE Release Inflow (ac-ft) (ac-ft) (ft) (cfs) (ac-ft) (ac-ft) 1950 505.5 1800 158000 5050 156000 6470 1951 494.7 350 133000 1980 135000 0 1952 515 6730 294000 21200 172000 142000 1953 502.5 350 147000 4040 151000 18 1954 510.9 5050 203000 7840 150000 59800 1955 505.1 500 156000 7770 159000 4170 1956 521.3 10650 211000 365 138000 73100 1957 506 2620 165000 4070 151000 17200 1958 511.7 5220 288000 11900 181000 117000 1959 502 350 129000 0 129000 0 1960 501.1 620 146000 3800 148000 1480 1961 496.6 350 116000 9150 124000 1690 1962 506 3820 193000 7030 156000 43000 1963 505.5 1800 165000 3.34 152000 5750 1964 500 350 141000 4290 154000 121 1965 507.2 3940 166000 4380 148000 21100 1966 519.3 9390 248000 1610 151000 97600 1967 521.3 12370 297000 11400 182000 124000 1968 505.3 1440 163000 9850 164000 8560 1969 532.1 20980 689000 21700 183000 525000 1970 505.9 2910 157000 3210 145000 15400 1971 505.6 2430 150000 3420 145000 8,430 1972 506.4 4070 151000 3970 132000 23000 1973 511.3 5170 218000 4000 164000 57400 1974 511.2 5190 192000 9340 152000 49000 1975 505.3 1010 163000 2250 162000 2430 1976 505.7 2130 156000 3650 148000 11300 1977 505.6 2460 167000 1270 160000 7180 1978 518.2 8890 442000 31900 174000 299000 1979 511.7 5250 252000 4650 173000 82500 1980 522.7 13950 591000 22200 178000 432000 1981 505.6 2990 153000 4350 146000 11100 1982 512.9 5680 234000 6840 164000 75700 1983 516.9 9030 410000 41300 205000 241000 1984 504.6 350 148000 710 150000 0 1985 505.8 2660 159000 1840 149000 11100 1986 508.2 4290 101000 12600 167000 45700 1987 502.8 350 137000 4040 140000 0 1988 505 350 167000 5450 170000 1660 Average: 508.8 4300 217000 7890 157000 67200

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 40 TABLE 9: HEC-5 RESULTS - WITHOUT PROJECT – FUTURE CONDITIONS (ALT2-2071) 350 cfs Recharge Rate

Prado Dam Total Local Water Water Water Max Max Prado Inflow "Lost" Year Spread WSE Release Inflow (ac-ft) (ac-ft) (ft) (cfs) (ac-ft) (ac-ft) 1950 502.2 3600 110000 552 86700 29200 1951 500.8 990 84900 2180 82400 4520 1952 514.6 6370 250000 23 111000 162000 1953 501.2 1900 100000 4420 91800 12900 1954 509.3 5000 150000 8540 86200 72500 1955 500.7 2640 111000 8400 92300 27200 1956 524.5 13740 167000 403 75300 91500 1957 510.2 5010 121000 4430 87900 37800 1958 511.4 5150 244000 12900 119000 137000 1959 498.9 2240 80000 0 65700 14300 1960 499.2 2220 101000 4130 78900 26300 1961 498.2 910 65700 9910 68800 6790 1962 503.1 4250 148000 7640 85800 69500 1963 508.7 5000 121000 3640 87500 37200 1964 505 2610 95700 4690 84900 15300 1965 506.1 5000 123000 4790 88500 39100 1966 521.6 10740 201000 1770 83700 119000 1967 521.7 12540 257000 12400 113000 156000 1968 508.5 4320 117000 10700 90300 37500 1969 532.7 21720 648000 23500 141000 529000 1970 502.7 5000 109000 3540 75700 36500 1971 500.2 3290 102000 3760 78900 27000 1972 506.8 5000 103000 4370 66500 40300 1973 509.9 5000 176000 4410 101000 78700 1974 508.4 5000 137000 10100 81600 65100 1975 505 2800 118000 2480 93300 27100 1976 502.5 4150 110000 4010 75700 38000 1977 505 3010 121000 1420 92700 28800 1978 518.8 9040 406000 34500 126000 314000 1979 510.9 5050 205000 5140 116000 94300 1980 523.5 14430 548000 23800 132000 438000 1981 500.5 3720 105000 4780 82100 27400 1982 513.2 5160 191000 7490 101000 97100 1983 517.4 9320 366000 44300 156000 251000 1984 501.4 2840 102000 789 78200 26200 1985 501.1 4230 112000 2030 84800 29200 1986 507 5000 158000 13600 103000 68900 1987 498.9 3350 89600 4430 79700 14300 1988 505 3400 126000 5920 97800 33700 Average: 508.1 5520 171000 8560 93400 86200

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 41 TABLE 10: HEC-5 RESULTS – ALTERNATIVE 4 – FUTURE CONDITIONS (ALT4-2071) 350 cfs Recharge Rate

Prado Dam Total Local Water Water Water Max Max Prado Inflow "Lost" Year Spread WSE Release Inflow (ac-ft) (ac-ft) (ft) (cfs) (ac-ft) (ac-ft) 1950 505.4 2060 110000 5520 107000 8580 1951 500.9 350 84900 2180 86900 0 1952 516.6 7390 250000 23 121000 152000 1953 505.3 950 100000 4420 103000 1580 1954 511.5 5160 150000 8540 98400 60300 1955 505.7 2540 111000 8400 106000 13000 1956 524.7 13960 167000 403 81500 85300 1957 512.9 5650 121000 4430 99500 26000 1958 513.8 6020 244000 12900 130000 126000 1959 505.3 1270 80000 0 76400 3550 1960 505.4 1560 101000 4130 97700 7350 1961 502.3 350 65700 9910 73200 2350 1962 506.3 4050 148000 7640 102000 53600 1963 510.6 5030 121000 3640 97600 24700 1964 505.3 1080 95700 4690 99500 2840 1965 501.9 4020 123000 4790 94100 33500 1966 523.2 11830 201000 1770 101000 101000 1967 523.4 13630 257000 12400 128000 141000 1968 508.5 3470 117000 10700 103000 24700 1969 533.6 22830 648000 23500 142000 528000 1970 505.9 4150 109000 3540 88900 23200 1971 505.9 2760 102000 3760 88600 17300 1972 511.3 5120 103000 4370 72800 34100 1973 512.7 5700 176000 4410 117000 63000 1974 511.9 5560 137000 10100 90800 55900 1975 505.6 2270 118000 2480 106000 13900 1976 501.9 4020 110000 4010 92300 21100 1977 505.6 3000 121000 1420 99700 21800 1978 519.4 9940 406000 34500 132000 308000 1979 512.8 5660 205000 5140 129000 80600 1980 523.8 14620 548000 23800 136000 435000 1981 505.8 3040 105000 4780 91300 18100 1982 514.2 6540 191000 7490 118000 80500 1983 517.7 9480 366000 44300 173000 233 1984 505.5 1610 102000 789 96900 7440 1985 507.3 3350 112000 2030 93500 20500 1986 511.2 l,100 158000 13600 116000 55100 1987 505.8 2190 89600 4430 87500 6510 1988 505.8 2190 126000 5920 120000 11800 Average: 510.9 5370 171000 8560 105000 74400

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 42 TABLE 11: Project Benefit Analysis

350 cfs Fixed Recharge Rate

Water Scenario Condition Recharge Lost Alt 2 (Current Condition) 2021 151,000 73,200 Alt 4 (505 Yr Round) 2021 157,000 67,200 Benefit 6,000 (6,000) Alt 2 (Current Condition) 2071 93,400 86,200 Alt 4 (505 Yr Round) 2071 105,000 74,400 Benefit 11,600 (11,800)

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 43 TABLE 12: INUNDATION DURATIONS FOR PRADO DAM

Without Project – Present Conditions - 350 cfs Recharge Rate Flood Season Water Conservation with Forecasting to 498.0 ft + Non-Flood Season Water Conservation to 505.0 ft

Frequency Duration of Inundation (Days) at various elevations ALT2_2021 (Years) 470 480 490 494 498 500 505 510 520 530 540 2 124 112 100 65 20 6 0 0 0 0 0 5 191 168 152 129 71 57 l 3 0 0 0 10 199 188 176 156 117 93 10 l 0 0 0 25 220 190 178 166 142 119 20 10 2 0 0 50 269 256 220 188 155 139 20 14 7 0 0 70 270 259 240 215 156 141 20 14 7 2 0 100 272 263 254 251 236 227 20 12 7 4 0 October 3 2 1 0 0 0 0 0 0 0 0 November 12 10 8 4 0 0 0 0 0 0 0 December 16 14 13 9 1 0 0 0 0 0 0 January 23 22 20 14 2 1 1 0 0 0 0

February 22 21 19 15 2 1 1 0 0 0 0 March 1-14 10 10 8 6 4 3 0 0 0 0 0 March 15-31 12 11 10 8 6 6 0 0 0 0 0 April 17 16 15 13 11 10 0 0 0 0 0 May 12 11 10 8 5 4 0 0 0 0 0 Average Monthly Average Monthly June 3 3 3 2 2 1 0 0 0 0 0 July 1 1 1 0 0 0 0 0 0 0 0 August 1 1 1 0 0 0 0 0 0 0 0 September 2 1 1 0 0 0 0 0 0 0 0 Average Annual 133 123 110 81 35 26 2 1 0 0 0

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 44 TABLE 13: INUNDATION DURATIONS FOR PRADO DAM

Present Conditions - 350 cfs Recharge Rate Water Conservation to Elevation 505.0 ft Year Round

Duration of Inundation (Days) at various elevations ALT4_2021 Frequency (Years) (Increase from Alt. 2 Present Conditions) 470 480 490 494 498 500 505 510 520 530 540 148 137 126 110 85 65 5 0 0 0 0 2 24 25 26 45 65 59 5 0 0 0 0 196 183 163 151 128 110 20 3 0 0 0 5 5 15 12 23 57 53 15 1 0 0 0 215 192 178 168 155 146 30 6 1 0 0 10 16 5 2 12 38 54 20 1 1 0 0 233 212 195 184 171 157 50 12 3 0 0 25 13 22 17 18 30 38 30 2 1 0 0 270 260 247 223 206 195 50 16 7 2 0 50 1 3 27 35 51 56 30 2 1 2 0 271 261 252 238 214 203 50 16 8 3 0 70 1 2 12 23 58 62 30 2 1 1 0 272 265 257 254 251 250 50 17 9 5 0 100 1 2 3 4 15 23 30 5 1 1 0 3 3 2 1 0 0 0 0 0 0 0 October 1 1 1 0 0 0 0 0 0 0 0 12 10 8 4 1 1 0 0 0 0 0 November 0 0 0 0 1 1 0 0 0 0 0 16 15 14 11 8 6 1 0 0 0 0 December 0 0 l 2 7 6 0 0 0 0 0 25 24 22 20 17 14 2 0 0 0 0 January 1 2 3 6 15 13 1 0 0 0 0 26 26 25 23 19 17 2 1 0 0 0 February

4 5 6 8 16 16 1 0 0 0 0 12 12 12 11 9 8 1 0 0 0 0 March 1-14 2 3 3 5 5 5 1 0 0 0 0 14 14 13 12 10 9 1 0 0 0 0 March 15-31 3 3 3 4 4 3 1 0 0 0 0 21 20 19 17 13 11 0 0 0 0 0 April 4 4 4 4 2 1 0 0 0 0 0 Average Monthly Monthly Average 13 12 11 8 6 4 0 0 0 0 0 May 1 1 1 0 0 0 0 0 0 0 0 3 3 3 2 2 1 0 0 0 0 0 June 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 July 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 August 0 0 0 0 0 0 0 0 0 0 0 2 1 1 1 0 0 0 0 0 0 0 September 0 0 0 0 0 0 0 0 0 0 0 149 141 131 109 86 71 7 2 0 0 0 Average Annual 16 18 22 29 51 45 5 0 0 0 0

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 45 TABLE 14: INUNDATION DURATIONS FOR PRADO DAM

Without Project – Future Conditions - 350 cfs Recharge Rate Flood Season Water Conservation with Forecasting to 498.0 ft + Non-Flood Season Water Conservation to 505.0 ft

Frequency Duration of Inundation (Days) at various elevations ALT2_2071 (Years) 470 480 490 494 498 500 505 510 520 530 540 2 72 63 54 51 27 16 2 0 0 0 0 5 121 109 94 81 50 37 10 3 0 0 0 10 146 137 128 122 97 84 10 5 1 0 0 25 174 163 153 143 125 108 20 10 3 0 0 50 197 187 177 166 138 117 20 14 8 2 0 70 198 189 180 176 150 119 20 14 8 3 0 100 199 192 185 182 179 178 20 15 9 5 0 October 2 2 2 1 0 0 0 0 0 0 0 November 8 8 8 6 1 0 0 0 0 0 0 December 9 9 8 7 1 0 0 0 0 0 0 January 13 13 13 11 3 1 1 0 0 0 0

February 12 12 12 10 3 1 1 0 0 0 0 March 1-14 7 7 7 6 4 3 0 0 0 0 0 March 15-31 9 9 9 8 7 6 0 0 0 0 0 April 13 13 13 11 9 7 0 0 0 0 0 May 5 5 5 5 4 3 0 0 0 0 0 Average Monthly Average Monthly June 1 1 1 1 1 1 0 0 0 0 0 July 0 0 0 0 0 0 0 0 0 0 0 August 1 1 1 1 0 0 0 0 0 0 0 September 1 1 1 1 0 0 0 0 0 0 0 Average Annual 79 79 79 66 32 23 3 1 0 0 0

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 46 TABLE 15: INUNDATION DURATIONS FOR PRADO DAM

Future Conditions - 350 cfs Recharge Rate Water Conservation to Elevation 505.0 ft Year Round

Duration of Inundation (Days) at various elevations ALT4_2071 Frequency (Years) (Increase from Alt. 2 Future Conditions) 470 480 490 494 498 500 505 510 520 530 540 97 89 81 78 71 54 5 0 0 0 0 2 25 26 27 27 44 38 3 0 0 0 0 146 138 129 126 108 95 20 4 0 0 0 5 25 28 35 45 58 58 10 1 0 0 0 170 162 154 151 136 127 30 7 2 0 0 10 24 25 26 29 39 43 20 1 0 0 0 175 166 157 153 148 139 50 13 3 0 0 25 1 3 4 10 23 30 30 3 0 0 0 243 234 223 208 175 164 50 16 8 3 0 50 46 47 46 42 38 47 30 2 0 1 0 244 236 227 220 205 194 50 16 8 4 0 56 47 47 48 44 55 74 30 2 0 1 0 246 239 233 230 227 226 50 18 9 6 0 100 47 47 48 48 48 48 30 2 0 1 0 2 2 2 2 1 1 0 0 0 0 0 October 1 1 1 1 1 0 0 0 0 0 0 10 10 10 9 6 4 0 0 0 0 0 November 2 2 3 3 5 4 0 0 0 0 0 14 14 14 13 11 9 1 0 0 0 0 December 5 5 5 6 9 8 1 0 0 0 0 20 20 20 18 15 13 2 1 0 0 0 January 7 7 7 8 12 12 1 0 0 0 0 19 19 19 18 15 13 2 1 0 0 0 February

7 7 7 8 13 12 1 0 0 0 0 9 9 9 8 6 6 1 0 0 0 0 March 1-14 2 2 2 2 3 3 1 0 0 0 0 10 10 10 9 8 7 1 0 0 0 0 March 15-31 1 1 1 1 1 1 1 0 0 0 0 13 13 13 11 9 7 0 0 0 0 0 April 0 0 0 0 0 0 0 0 0 0 0 Average Monthly Monthly Average 5 5 5 5 4 3 0 0 0 0 0 May 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 June 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 July 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 0 0 0 0 0 August 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 0 0 0 0 0 September 0 0 0 1 1 0 0 0 0 0 0 105 105 105 95 76 63 8 2 0 0 0 Average Annual 26 26 26 30 44 40 5 1 0 0 0

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 47 TABLE 16: HEC-5 RESULTS - WITHOUT PROJECT – PRESENT CONDITIONS (ALT2-2021) 450 cfs Recharge Rate

Prado Dam Total Local Water Water Water Max Max Prado Inflow "Lost" Year Spread WSE Release Inflow (ac- (ac-ft) (ac-ft) (ft) (cfs) ft) (ac-ft) 1950 498.4 2210 158000 5050 155000 7970 1951 493.4 450 133000 1980 135000 0 1952 512.1 5330 294000 21200 180000 135000 1953 497.6 450 147000 4040 151000 0 1954 505 4990 203000 7840 159000 51200 1955 498.6 2290 156000 7770 153000 10600 1956 521.3 10640 211000 365 130000 81400 1957 505.1 5000 165000 4070 149000 19300 1958 508.5 5000 288000 11900 193000 106000 1959 498.2 1120 129000 0 127000 1920 1960 498.4 1430 146000 3800 147000 2990 1961 496 450 116000 9150 125000 411 1962 499.3 3920 193000 7030 154000 46100 1963 503.8 4910 165000 3340 144000 24400 1964 498.2 450 141000 4290 145000 0 1965 505.5 5000 166000 4380 151000 18700 1966 518.3 8740 248000 1610 154000 94700 1967 518.8 10680 297000 11400 182000 126000 1968 503.8 2110 163000 9850 161000 11700 1969 531.1 19760 689000 21700 206000 502000 1970 501.2 3980 157000 3210 143000 17300 1971 498.8 2380 150000 3420 141000 12700 1972 504.5 5000 151000 3970 124000 30900 1973 501 5000 218000 4000 177000 44700 1974 508.1 5000 192000 9340 147000 54800 1975 505 1370 163000 2250 163000 2120 1976 500.6 4070 156000 3650 140000 19200 1977 501.5 2980 167000 1270 149000 18500 1978 516.6 7600 442000 31900 197000 275000 1979 501.5 5000 252000 4650 194000 62100 1980 522.3 13640 591000 22200 201000 410000 1981 499.4 2440 153000 4350 146000 11700 1982 509.3 5000 234000 6840 179000 60700 1983 515.8 8360 410000 41300 234000 212000 1984 498.3 1160 148000 710 146000 4100 1985 499.8 2870 159000 1840 143000 17600 1986 505 5000 201000 12600 176000 37700 1987 498.6 1820 137000 4040 135000 5170 1988 498.9 2370 167000 5450 163000 8890 Average: 505.4 4610 217000 7890 159000 65300

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 48 TABLE 17: HEC-5 RESULTS – ALTERNATIVE 4 – PRESENT CONDITIONS

(ALT4-2021) 450 cfs Recharge Rate

Prado Dam Total Local Water Water Water Max Max Prado Inflow "Lost" Year Spread WSE Release Inflow (ac- (ac-ft) (ac-ft) (ft) (cfs) ft) (ac-ft) 1950 503.1 450 158000 5050 163000 0 1951 493.4 450 133000 1980 135000 0 1952 514.9 6710 294000 21200 189000 126000 1953 497.6 450 147000 4040 151000 0 1954 509.4 4610 203000 7840 166000 44400 1955 503.9 450 156000 7770 161000 2670 1956 521.3 10640 211000 365 140000 72000 1957 505.7 2130 165000 4070 160000 8670 1958 510.3 5010 288000 11900 198000 101000 1959 499.8 450 129000 0 129000 0 1960 500.3 450 146000 3800 150000 419 1961 496 450 116000 9150 125000 411 1962 506.1 3780 193000 7030 165000 34.9 1963 505.4 1750 165000 3.34 153000 7450 1964 498.2 450 141000 4290 152000 0 1965 507.3 3720 166000 4380 151000 18700 1966 519.5 9510 248000 1610 169000 80200 1967 521.3 12380 297000 11400 201000 107000 1968 505.2 1100 163000 9850 169000 3970 1969 532.1 20970 689000 21700 206000 502000 1970 505.9 2800 157000 3210 149000 11300 1971 505.4 1450 150000 3420 151000 2900 1972 506 3780 151000 3970 134000 21300 1973 511.l 5200 218000 4000 177000 44400 1974 511.3 5230 192000 9340 156000 45200 1975 505 450 163000 2250 165000 0 1976 505.6 1970 156000 3650 150000 8930 1977 505.7 2160 167000 1270 159000 8770 1978 518.2 8860 442000 31900 197000 275000 1979 511.7 5240 252000 4650 194000 61900 1980 522.7 13950 591000 22200 201000 410000 1981 505.4 1400 153000 4350 154000 3550 1982 512 5290 234000 6840 183000 56800 1983 516.8 8980 410000 41300 243000 204000 1984 499.6 450 148000 710 150000 0 1985 505.8 2380 159000 1840 153000 7850 1986 506.8 3180 201000 12600 176000 36700 1987 502.2 450 137000 4040 141000 25 1988 502.5 450 167000 5450 171000 1180 Average: 508 4090 217000 7890 165000 59200

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 49 TABLE 18: HEC-5 RESULTS - WITHOUT PROJECT – FUTURE CONDITIONS (ALT2-2071) 450 cfs Recharge Rate

Prado Dam Total Local Water Water Water Max Max Prado Inflow "Lost" Year Spread WSE Release Inflow (ac- (ac-ft) (ac-ft) (ft) (cfs) ft) (ac-ft) 1950 502.2 3590 110000 552 90600 25300 1951 499.4 950 84900 2180 84500 2470 1952 514.6 6360 250000 23000 119000 154000 1953 501.1 1910 100000 4420 94600 10200 1954 509.3 5000 150000 8540 92200 66600 1955 500.2 2640 111000 8400 96000 23500 1956 524.5 13740 167000 403 76800 90100 1957 510.2 5010 121000 4430 93500 32200 1958 511.4 5150 244000 12900 132000 124000 1959 498.9 2240 80000 0 68100 11900 1960 498.9 2200 101000 4130 83300 21900 1961 4982 920 65700 9910 71000 4650 1962 502.8 4120 148000 7640 91800 63500 1963 508.7 5000 121000 3640 88100 36700 1964 505 2580 95700 4690 87400 12800 1965 506.1 5000 123000 4790 92300 35500 1966 521.5 10610 201000 1770 90500 112000 1967 521.7 12510 257000 12400 121000 149000 1968 508.4 4, 180 117000 10700 93600 34300 1969 532.7 21790 648000 23500 165000 505000 1970 502.6 5000 109000 3540 79100 33100 1971 500.1 3430 102000 3760 81700 24200 1972 506.8 5000 103000 4370 68100 38800 1973 510 5000 176000 4410 111000 68900 1974 508.4 5000 137000 10100 86100 60700 1975 505.0 2730 118000 2480 97300 23200 1976 502.7 4020 110000 4010 77500 36200 1977 505 3020 121000 1420 93800 27900 1978 518.8 9030 406000 34500 142000 298000 1979 510.7 5030 205000 5140 125000 85400 1980 523.5 14450 548000 23800 152000 419000 1981 5002 3450 105000 4780 85900 23700 1982 512.8 5630 191000 7490 110000 88900 1983 517.4 9320 366000 44300 176000 231000 1984 5012 2820 102000 789 81200 23200 1985 501.3 4140 112000 2030 87600 26500 1986 507.1 5000 158000 13600 109000 63000 1987 498.8 3340 89600 4430 81300 12700 1988 504.4 3380 126000 5920 103000 29000 Average: 508.0 5490 171000 8560 99400 80200

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 50 TABLE 19: HEC-5 RESULTS – ALTERNATIVE 4 – FUTURE CONDITIONS

(ALT4-2071) 450 cfs Recharge Rate

Prado Dam Total Local Water Water Water Max Max Prado Inflow "Lost" Year Spread WSE Release Inflow (ac-ft) (ac-ft) (ft) (cfs) (ac-ft) (ac-ft) 1950 505.4 2000 110000 5522 110000 5870 1951 500.0 450 84900 2182 87000 0 1952 516.5 7330 250000 23014 129000 144000 1953 503.4 450 100000 4419 105000 0 1954 510.8 4780 150000 8537 105000 53300 1955 505.4 2250 111000 8404 110000 9910 1956 524.7 13900 167000 403 82700 84100 1957 512.3 5420 121000 4429 105000 20300 1958 514.0 6100 244000 12916 146000 110000 1959 505.2 989 80000 0 78800 1210 1960 505.4 1460 101000 4132 102000 2870 1961 501.9 450 65700 9911 74800 829 1962 506.4 3940 148000 7640 110000 44800 1963 510.5 5020 121000 3638 98500 25800 1964 505.2 981 95700 4687 99500 1200 1965 508.1 3870 123000 4792 96600 31100 1966 523.1 11700 201000 1769 111000 91100 1967 523.4 13600 257000 12436 136000 134000 1968 508.4 3370 117000 10725 109000 18600 1969 533.6 22800 648000 23536 165000 505000 1970 506.0 3950 109000 3544 91200 21000 1971 505.8 2380 102000 3759 94100 11800 1972 511.4 5140 103000 4372 74300 32500 1973 512.3 5510 176000 4409 130000 50200 1974 512.0 5600 137000 10120 94300 52500 1975 505.5 2180 118000 2483 109000 11800 1976 508.0 3990 110000 4007 93000 20500 1977 505.6 3000 121000 1418 101000 21000 1978 519.6 10100 406000 34465 153000 287000 1979 511.8 5260 205000 5139 147000 62500 1980 523.8 14600 548000 23843 158000 413000 1981 506.0 2640 105000 4951 97600 11900 1982 514.0 6320 191000 7494 131000 67100 1983 517.6 9420 366000 44537 198000 208000 1984 505.1 848 102000 789 101000 2880 1985 506.6 2990 112000 2033 96300 17700 1986 511.8 5240 158000 13595 126000 46000 1987 505.7 2100 89600 4433 88300 5680 1988 505.8 2220 126000 5923 123000 8940 Average: 510.7 5240 171279 8574 111974 67589

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 51 TABLE 20: Project Benefit Analysis

450 cfs Maximum Recharge Rate

Water Scenario Condition Recharge Lost Alt 2 (Current Condition) 2021 158,282 65,969 Alt 4 (505 Yr Round) 2021 165,051 59,220 Benefit 6,769 (6,749) Alt 2 (Current Condition) 2071 98,890 80,760 Alt 4 (505 Yr Round) 2071 111,974 67,589 Benefit 13,084 (13,171)

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 52 TABLE 21: INUNDATION DURATIONS FOR PRADO DAM

Without Project - Present Conditions – 450 cfs Recharge Rate Flood Season Water Conservation with Forecasting to 498.0 ft + Non-Flood Season Water Conservation to 505.0 ft

Frequency Duration of Inundation (Days) at various elevations ALT2_2021 (Years) 470 480 490 494 498 500 505 510 520 530 540 2 94 75 55 42 12 4 0 0 0 0 0 5 145 130 108 91 49 37 8 3 0 0 0 10 173 163 154 148 104 79 10 5 0 0 0 25 196 184 167 156 126 110 20 14 3 0 0 50 236 211 184 170 144 118 30 14 6 0 0 70 240 222 204 180 150 121 30 14 7 2 0 100 242 225 208 201 154 131 60 15 8 3 0

TABLE 22: INUNDATION DURATIONS FOR PRADO DAM

Alternative 4 – Present Conditions - 450 cfs Recharge Rate Water Conservation to Elevation 505.0 ft Year Round

Duration of Inundation (Days) at various elevations ALT4_2021 Frequency (Increase from Alt. 2 Present Conditions) (Years) 470 480 490 494 498 500 505 510 520 530 540 117 95 77 64 47 36 5 0 0 0 0 2 23 20 22 22 35 32 5 0 0 0 0 169 154 131 118 99 84 19 4 0 0 0 5 24 24 23 27 50 47 11 1 0 0 0 173 164 155 151 141 133 51 7 1 0 0 10 0 1 1 3 37 54 41 2 1 0 0 218 205 181 169 157 151 79 16 3 0 0 25 22 21 14 13 31 41 59 2 0 0 0 244 231 212 202 182 168 86 17 7 0 0 50 8 20 28 32 38 50 56 3 1 0 0 246 237 228 219 190 177 89 17 8 3 0 70 6 15 24 39 40 56 59 3 1 1 0 246 238 229 226 200 180 100 17 8 4 0 100 4 13 21 25 46 49 40 2 0 1 0

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 53 TABLE 23: INUNDATION DURATIONS FOR PRADO DAM

Without Project – Future Conditions - 450 cfs Recharge Rate Flood Season Water Conservation with Forecasting to 498.0 ft + Non-Flood Season Water Conservation to 505.0 ft

Frequency Duration of Inundation (Days) at various elevations ALT2_2071 (Years) 470 480 490 494 498 500 505 510 520 530 540 2 70 61 52 41 20 10 2 0 0 0 0 5 96 87 77 68 44 27 9 3 0 0 0 10 144 133 117 106 80 65 19 7 2 0 0 25 167 155 139 132 119 99 20 14 4 0 0 50 171 163 154 151 126 106 20 14 7 1 0 70 172 164 155 152 126 106 20 15 8 3 0 100 172 164 156 152 131 110 30 15 8 5 0

TABLE 24: INUNDATION DURATIONS FOR PRADO DAM

Alternative 4 – Future Conditions - 450 cfs Recharge Rate Water Conservation to Elevation 505.0 ft Year Round

Duration of Inundation (Days) at various elevations ALT4_2071 Frequency (Increase from Alt. 2 Future Conditions) (Years) 470 480 490 494 498 500 505 510 520 530 540 95 85 76 66 52 40 7 0 0 0 0 2 25 24 24 25 32 30 5 0 0 0 0 121 113 105 102 89 77 19 5 0 0 0 5 25 26 28 34 45 50 10 2 0 0 0 147 139 132 129 126 115 50 8 2 0 0 10 3 6 15 23 46 50 31 1 0 0 0 171 162 153 149 137 131 75 16 4 0 0 25 4 7 14 17 18 32 55 2 0 0 0 216 205 172 163 153 145 76 17 8 2 0 50 45 42 18 12 27 39 56 3 1 1 0 219 211 202 195 177 161 77 17 8 4 0 70 47 47 47 43 51 55 57 2 0 1 0 219 212 204 201 188 179 86 17 8 5 0 100 47 48 48 49 57 69 56 2 0 0 0

Prado Basin Feasibility Study February 2021 Water Conservation & Sediment Transport Analysis Report 54 N 0 10 20 end * Stream Gage (Gage Index - Table 5-2) SANTAANA RIVER BASlN, CALfF'ORNlA JNTERTh.:f WATER CONTROL MANUAL • Rain Gage (Gage Index - Table 5-2) .A. Reservoir Pool Gage (Gage Index - Table 5-2) GOES AND ALERT - ---- Water Course STREAM AND RESERVOIR Santa Ana River Watershed GAGING STATIONS LJ Apri 2016

11 7•Q'(1'W 11s·o-r:rvv U.S. .Afil.fY CORPS OF ENGINEERS LOSANGfil,ES DISTRICT Figure 1: Santa Ana River Drainage Area Basin Map

Figure 2: Prado Basin Study Area Map

SEASO,N------~ Prado Dam Release Ranges - Using New Outlet Works I EL 5 4.4 65 4 I Pertinent EL 594.4 ~ TOP OF DAM Pool Elevations Recommended Plan of Discharge Description ft NGVD29 I I At the start of a runoff event, a debris pool is built to settle out floating debris to limit their DEBRIS POOL being drawn into the outlet works. Discharge from within this pool range varies (0-600 cfs), I 470 _ 490 but generally accommodates the OCWD's dm.vnstream spreading capacity without waste to I the Pacific Ocean. Flow may also be shut off temporarily to facilitate construction I BUFFER POOL Discharge range (0 - 5000 cfs). The buffer pool range is borrowed flood risk management 4 6 r-Fl-oo_d__ ~N-o-n--F-lo_o_d--1 Q > 10,000 cfs:(Uncontrolled Spillway Flow) ' ~i:;:gt~~e~~~:~~~i:t~~:::i~~ ~;\:~~r4~o8~s~~::~:~- d~~~~o~:::;:~~e~o~~~::~~~on I Season* Season** period, this elevation can be maximized up to 505. Upon completion of the Prado Basin EL 490-498 EL 490-505 Feasibility Study and approval of water conservation alternative, the top of buffer pool will I /505 remain at EL 505 durin both flood and non-flood seasons I FLOOD RISK I MANAGEMENT POOL Discharge ( :S 10,000 cfs). The resulting maximum reservoir release will depend on I Flood Season Non-Flood anticipated inflow and dm.vnstream conditions. Maximum discharge is maintained lllltil the ilhv,ay ,Surcharge. ______!. ______.ELS45J1--JLilll.-l-J-Q_ EL 498/505- Season pool elevation recedes back do\Vll to/below the top of buffer pool elevation . . . Q ~ 10,0001cfs (Outll.'t Works & Spillway Flow) u 520 EL 505-520 Discharge ( :S 10,000 cfs). The resulting maximum reservoir release will depend on FLOOD RISK anticipated inflow and do\Vllstream conditions. Maximum discharge is maintained until the I MANAGEMENT POOL pool elevation recedes back do\Vll to/below the top of buffer pool elevation. Ifhydrologic I 520 - 543 *** conditions warrant, or if dam safety is a concern, discharge through the outlet works may Approved Deviation 1 exceed 10,000 cfs 4 6 Discharge ( > 10,000 cfs) Combined controlled discharge through outlet works and Top of Buffer Pool I Q :s_ 10,000 cfs ' SPILLWAY SURCHARGE POOL uncontrolled discharge over the spillway. Existing dam safety concern with the unmodified during Flood Season H : EL 520 spillway. Controlled outlet discharge may be maximized ( > 10,000 cfs) to minimize the 543 - 545.10 *** duration of spillway flow. UNCONTROLLED Discharge ( > 10,000 cfs) Combined controlled discharge through outlet works and uncontrolled discharge over the spillway. Existing dam safety concern with the unmodified SPILLWAY DISCHARG spillway. Controlled outlet discharge may be maximized ( > 10,000 cfs) to minimize the 545.10 - 594.4 *** I~ ::~,;,~Nm::: .,__..cU+L.L..L duration of spillway flow. r Footnotes 1 3 5 8 9 f------1L.L.J..i~-ER POOL ' ' ' ' EL4 8 9 369 Q := 0 to 5,000 cfs

I 3 5 D~ RIS POQL ' Q = 0 to 600 cfs I I OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP 3

NOTES: PRADO DAM 1. The top of buffer pool in during flood season is normally up to elevation 498 ft and at 505 ft during non-flood season. Beginning 1 March, the reservoir pool is gradually increased to transition from EL 498 ft SANTAANARIVERBASIN, CALIFORNIA to EL 505 ft by March 10. Discharge from Prado to accommodate OCWD's spreading capacity varies, maximizing up to 600 cfs during both flood and non-flood seasons. INTERJM WATER CONTROL MANUAL 2. Storage Data based on Reservoir Surveys completed in 2008, 2013, and 2015. Area and Storage Capacity Tables provided in ExhibitB of the IWCM. 3. Maintenance activities typically scheduled during summer months (i.e., July, August, September). Reservoir should be drained by end of August, if possible. INTERIM 4. Maximum outlet discharge capability is up to 30,000 cfs. Above elevation 520 ft, releases may be increased to the maximum possible discharge to prevent/minimize spillway flow, or due to dam safety concerns. 5. During non-flood season, discharge is subject to minimum release requirements (running average 350 cfs) while the pool elevation remains above elevation 498. WATER CONTROL PLAN 6. If discharge will exceed 10,000 cfs, a dam safety team must be on site to evaluate impacts to the dam and outlet works. Discharge greater than 10,000 cfs will be also be coordinated with SPD 7. All elevations cited are in feet, NGVD 1929 (IWCP - August 2020) 8. The l\•l ult i-Year De\"ialion Request changes ON LY t.lle top of lm fTer pool elevation frnm 498 to 505 during Oood seasons. All otl1er operating eondilions specified fo1· the IWCP remai11s lhe same. 9. Upon comp letion of lhe Prado Basin Feasibi lily Srudy and approva l of Ill• water consel"'·otion alleruolive, lhe •11pro"ed de,•lation plan will be permanently lmplemeillecl as parr of rills {WCP. U.S. ARMY CORPS OF ENGINEERS Should lhc devialion pe.-iod expire plior lo cornplet.ing tile Feasibility Sludy, tben the maximum lop of flood season buITe1· pool will re.main. a t elevation 498. LOS ANGELES DISTRJCT

This Diagram Supersedes th e 2003 IWCP Diagram (Plan A) Plate 7-01

Figure 3: Interim Water Control Plan – August 2020

Figure 4: Modeling Extent and Trap Scenario Modeled (Golder, 2015a)

Base Case Scenario - Channel Bed Profile Change Oct. 1975 - Oct. 1985

725 --Begin Elev

---- End Elev

- Market St - 675 ~ - 60Fwy C: ·;;0 - M ission Blvd "'> QJ - RR Bridge i:ij 625 - MWD Crossing - Va n Buren Blvd

- 1-15 575 - Hamner Ave

- River Rd Bridge

525

0 20000 40000 60000 80000 100000 120000 Distance From End of Reach (ft)

Figure 5: River Bed Profile Change after 10 Years for Base Case Scenario (Golder, 2015a)

Trap Scenario - Channel Bed Profile Change Oct. 1975 - Oct. 1985

775 --Begin Elev

---- End Elev

725 - Market St - 60 Fwy

- M ission Blvd ~ 675 - RR Bridge C: 0 ·.;::; - MWD Crossing > .!!"" w 625 - Van Bu ren Blvd

- 1-15

- Hamner Ave 575 - Upstream End of Transition Channel - River Rd Bridge 525 - Upstream End of Trap

0 20000 40000 60000 80000 100000 120000 Dista nce From End of Reach (ft)

Figure 6: River Bed Profile Change after 10 Years for Trap Scenario (Golder, 2015a)

se Case vs . Trap- Channel Slope After 10 Years (Oct 1985)

0 .008 ..------

---- - Yr 10 Base Case 0 .007 --Begin Trap

--Yr lO Trap 0 .006 +-+-- -+-+------+-- 1--+-----+-- o+------M arket St

- 60 Fwy - 0 .005 - M ission Blvd i.. - RR Bridge Cl. ~ 0 .004 - M WD Crossing ai C ..C - van Buren Blvd s:; u 0 .003 +------======------1-15

- Hamner Ave

0 .002 -1------u pst ream End of Transition Channel - River Rd Bridge

0 .001 -1------u pst ream End of Trap

0 -1------~-----~-----~------~-----~-----~ 0 20000 40000 60000 80000 100000 120000 Distance From End of Reach (ft)

Figure 7: Comparison of Channel Slope Change after 10 Years (Golder, 2015a)

Base Case vs. Trap - Average 050 After 10 Years {1985)

100.00

Coarse Gravel

------D50 Trap Avg 10.00 r1ne 1.:1rave1 --D50 Base Case Avg -- - Begin of Run ----- Coarse Sand - Market St ------60Fwy E - Mission Blvd 1.00 0 ----- Medium Sand - RR Bridge lS - MWD Crossing - van Buren Blvd Fi ne Sand - 1-15 - Hamner Ave 0.10 - Upstream End of Transition Channel

- River Rd Bridge

- Upstream End of Trap

Si lt and Clay

0.01 ' . ' ' ' . 0 20000 40000 60000 80000 100000 120000 Distance From End of Reach (ft)

Figure 8: Comparison of Average D50 Change after 10 Years (Golder, 2015a)

1st Scenario - Channel Bed Profile Change Oct. 1975 - Oct. 1985

825

775 --Begin Elev

---- End Elev

725 - Market St

- 60 Fwy

- M ission Blvd :!:,_ 675 - RR Bridge C: .....S! - MWD Crossing > .!!"" w 625 - Van Buren Blvd

- 1-15

- Hamner Ave 575 - River Rd Bridge

- Upst ream End of Transition Channel 525

475 0 20000 40000 60000 80000 100000 120000 Dist ance From End of Reach (ft)

Figure 9: River Bed Profile Change after 10 Years for 1st Scenario (Golder, 2015b)

2nd Scenario Channel Bed Profile Change Oct. 1975 - Oct. 1985

775 --Begin Elev

---- End Elev 725 - M arket St

- 60 Fwy

- M iss ion Blvd ~ 675

C: 0 - RR Bridge ·.;::; > .!!"" - MWD Crossing w 625 - Van Buren Blvd

- 1-15

575 - Hamner Ave

- River Rd Bridge

- Upstream End of Transition 525 Ch annel

0 20000 40000 60000 80000 100000 120000 Distance From End of Reach (ft)

Figure 10: River Bed Profile Change after 10 Years for 2nd Scenario (Golder, 2015b)

1 % Fine Sand Deposited and Re-entrained Sediment RAS Model Results

40.00 ~------

30.00 +------

20.00 +------<------

- sediment Deposited

- sediment Re-ent ra ined 25000 - MPM Deposition

-30.00 +------

-40.00 ..______River Flow (cfs)

Figure 11: Deposited and Re-entrained Sediment – 1% Fine Sand (Golder, 2017)

1 % Medium Sand Deposited and Re-entrained Sediment RAS Model Results

40.00 ..------

30.00 +------

- sediment Deposited

- sediment Re-ent ra ined 25000 - MPM Deposition

-20.00 +------

-30.00 +------=------

-40.00 ~------River Flow (cfs)

Figure 12: Deposited and Re-entrained Sediment – 1% Medium Sand (Golder, 2017)

5% Fine Sand Deposited and Re-entrained Sediment RAS Model Results

40.00 ~------

30.00 +------

20.00 +-+,------

10.00

-~ - sediment Dep osit ed -.,C: 0.00 E - sediment Re-ent rained 'o., 15000 20000 25000 V, - MPM Dep osition

-10.00

-40.00 ~------River Flow (cfs)

Figure 13: Deposited and Re-entrained Sediment – 5% Fine Sand (Golder, 2017)

5 % Medium Sand Deposited and Re-entrained Sediment RAS Model Results

40.00 ~------

30.00 +------

- sediment Deposited

- sediment Re-ent ra ined

- MPM Deposition

-30.00 +------

-40.00 ..______River Flow (cfs)

Figure 14: Deposited and Re-entrained Sediment – 5% Medium Sand (Golder, 2017)

~ ~ -g a::i V') en ru II ru CD :: - ccS .... Q. 0 ci::: II) <( V') ~ _, 0 cu ru CD ...... 0 ~ ' {\ ', ~ ' j ' "' O'l 0 ~ ' \ , F .. O'l 0 · ail Ball Road \_ Rd G, < "' 0 N I 3ridge psu ~ G 57 racks Frwy ' \ F m (iJ) rwy Ca IV U0! 'Veir .. 0 N I hyon \. Bridge cFadden Bicycle ridge l EM13 ~ IC .s 'oi, Q. D D CD CD CD e ',. 00 C: C: QJ Can 'I · ~ Rd > > 11') H , V, V, ' "' "' J QJ QJ 4 1 I I "' .... 0 11pr ~ dam Bridge rdige 91 JS on rbor Hamilt .... 0 Frwy F Rd Hw, s rwy Blv Br Li .... 0 .. lridge dge Jn \_ Bridge I coin Bridge PCH Bridge Victo1 Av• Bri _ '\ "' 0 a ge Bridge Bridge \ 0 .. \ I 1 >- >- >- >- 1 ':' 0 § § § .. g g § ...... g 0 0 0 0 0 00 0 0 0 0 0 0 ~ N N .. :;:: . g "' C: - "' 0 C: QJ 2: ~

Figure 15: River Bed Profiles after 5 and 10 Years – Base Fine Sand (Golder, 2017)

·- ~ ~ ""'~ LL V") m C CIJ s:::: 0 ·- ..! o?J ~ i ""' ..... Ln 0. 0 a:: O ~ y c le C "C "C L LL LL "'1 ->- >- l Vl Vl . L.. L.. : C: C: l (ll (ll J QJ QJ ~ ~ I .-! a-, I rdige Bridge 91 Ham ...... 0 I Frwv ➔ .-! 0 'Sj" I l I I I I i I I , 1 Bridge Bridge - PC _ "J H _ B - ri 0 a-, t +- I g B _ e ridg __ " e _ 0 q- I I I I 11 I f ~ " ,-1 0 I I ' ' i ~ ...... l.O ,-1 ,-1 0 0 0 0 0 0 0 O 0 0 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 'OJ' 'Sj" N . · E ,.,., ..., p a:: ... ? 0 C: Q) ('Cl

Figure 16: River Bed Profiles after 5 and 10 Years – 1% Fine Sand (Golder, 2017)

·- "'1J "'1J ~ .... ~ V) E ru ::::J cu C - - ; ~ ..., .... Q. 0 ci:::: II) <( <( ~ V) +- ru 0 cu cu ...... 0 .. .. I I I I "' 0 (71 "' 0 .. I I Ball ii \J Road Rd 9 N 0 (71 Tracks G C:7 ridge Frwy l'.:run, (iJ) UO .. N 0 ' cFadden ll ridge ! l rirln EM13 IC .!: ' Q. CD i,o 00 0 C C QJ QJ ~ o 0 -0 -0 > > . 1/) 0 -0 -0 ) V) V) ! :!? :!? H~rbor E E I (I) (I) 2 C C J QJ QJ I ...... 0 (71 I l Bridge rdige 91 on 5 .... 0 ::, I Frwy r Frwy Rd Blv~ Li 0 .. I I I , Bridge coin ridge PCH II Brid ~ ¥1 .... Av g Bri '-VI e 0 (71 I I I u - ge Bridge .... , .... ' 0 \I ... 0 .. I I 1 I 1 I \ \. ~ I I I I. I I > I I ' I ~ ~ ' ':' 0 I I l i i .. g ~ § ...... o N 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 O 00 . ·.;:; g 0: V) ~ (I) 0 C QJ 2! ~

Figure 17: River Bed Profiles after 5 and 10 Years – 1% Medium Sand (Golder, 2017)

1% Fine Sand LSAR Profi le After 15 Years (Years 11-15 Clear Water)

440 +------

390

340 :::0w-- to ~ :::0 0 ::::0 Q.) a. G) a. 290 "Cl 9 - V, ... Q.) C n 7' 3 V, n -C 240 Q.) 0 n Q?. ... :.:; 'T1 < nl QJ n ci ::::!. 0 > a. < Cltl a. :::, (I) a. ro Cltl a. (I) :::0 ---- Beg inning Profi le w (I) (I) a. 190 :::, to ... \J) 0 --Fine Sa nd Yr 10 a. I-" :::, :I: ::::!. citi' QJ a. 'Tl, ::::0 Cltl (I) .... 3 (I) a. --Fine Sa nd Yr 15 ;::. ~ 140 o,, ,, OJ ... to ::::t -, n ci a. :I: Cltl ... a. ro ci Cltl OQ co Cltl (I) (I) .... ro 90 OQ QJ (I) to ::::!. a. OQ 40 (I) -- - -10 - 0 20000 40000 60000 80000 100000 120000 140000 160000 River Stat ion (ft)

Figure 18: River Bed Profiles after 15 Years – 1% Fine Sand (Golder, 2017)

1% Medium Sand LSAR Profile After 15 Years (Years 11-15 Clear Water)

... ;:;· a: ... -< OQ a: Q. Cl) OQ Cl) Cl) Q ---- Beginning Profile :, 190 +----+,!e<'---+-i".----...... ,- +=------+----11-,----+-cc------Medium Sand Yr 10

-10 0 20000 40000 60000 80000 100000 120000 140000 160000 River Station (ft)

Figure 19: River Bed Profiles after 15 Years – 1% Medium Sand (Golder, 2017)

Annual Maximum Projected Annual Max Monthly Mean Projected Annual Max M... Huc-4 Reference Map

1) Choose a HUC-4 2) Click Map Location or Name to Select Stream 1 807-Southern California Coastal Gage

Site Number Search tor Gage within HUC-4 by Name 11075755 11074500 11059300 SANTA ANA RAE ST NR SAN BERNARDINO CA 3) Include Only Years (It Desired) 11075600 1967 2016 11066440 0 D 11066460 11067890 11066500 Annual Peak Instantaneous Streamflow, SANTAANA RAE ST NR SAN BERNARDINO CA Selected (Hover Over Trend Line For Significance (p) Value) Climate Hydrology Assessment Tool v.1 .0 Analysis: 2/812021 10:49 AM

35K •

30K • • Tile p-va/ue is for the 25K /,near regression fit drawn; a smaller p-va/ue \Vou/d indicate greater statistical • significance. There 1s no recommended threshold for statistical significance, but typically 0.05 1s used as • • this 1s associated with a 5% • • risk of a Type I error or false 10K • • SK --~·~----•!______--4 • • ..._1eate•----,~~---=--=----• •• • • •• • • • • • • • • •• • • OK • • •• • •••• • • 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 Water Year

Prado Basin San Bernardino County California

Value = 60.1296*Water Year + -112993 ANNUAL PEAK INSTANTANEOUS STREAMFLOW R-Squared: 0.0120548 P-value: 0.462542 SANTA ANA RIVER AT E STREET U. Ar:my Corps, of Enginee:rs

L0:s Angel1es Distl:'i.d :

Plate 1 1) Choose a HUC-4 2, Click Map Location or Harne to Select Stream 1'807-So uthem C alifornia Coas tal Gage Site Num ber Search tor Gage wilhin HUC-4 by Name 11108145 All 11073360 CHINO CA SCHAEF ER AVENUE NR CHINO CA 11073500 11055800 3) Include Only Years ("Desire d) 11055801 1757 t.o 20 16 11106400 11090700 11090500

A.nnual Peak Instantaneous Streamflow, CHINO CA SCHAEFER AVENUE NR CHINO CA Selected rHover Over Trend Line For Signjficance (p) Value) ::limate Hydrology Assessment Tool v .1 0 An alysis: 11/812017 1:39 PM

• 12K

10K The p-va/ue is for the linear regression fit drawn; a smaller p -value would in • indicate greater statistical LL 8K 8 significance. There is no recomm ended threshold for ~ statistical significance , E • but typically 0. 05 is used as :!'"' 6K this is asrociated with a 5% 1ii • • risk of a Type I error or false • •• positive . 4K •• • •

• ••• • • 2K • • • • • • ••• • • • • •• • • • OK •••••••• • • 1970 1975 1980 19S5 1990 1995 2000 2005 20 10 2015 Water Ye ar Prado Basin San Bernardino County California

Value = 11.0115*Water Year + -19027.9 ANNUAL PEAK INSTANTANEOUS STREAMFLOW R-Squared: 0.0035992 P- value: 0.692064 CHINO CREEK AT SCHAEFER AVENUE

1 .S Anny Cmrp,s of Engineers 1.,os Ang,des D isturkt

Plate 2 1) Choose a HUC-4 2) Click Map Location or Name to Select Stream 1S07-Soutllem C alifornia Coastal Gage

Site Number Search for Gage wirhin HUC-4 by Name 11071900 All 11072100 TE.M ESCAL CAB MAIN ST A CORONA CA 11072000 11-18 11013000 3) Include Only Years ("Desired) 11013500 1757 to 2016 11104000 11094000 11094500 fl.nnua l Peak Instantaneous Streamflow, TEMESCAL CAB MAIN ST A CORONA CA Selected rHover Over Trend Line For Significance (p) Value) ~limate Hydrology Assessment Tool v.1.0 Analysis: 11/812017 1:44 PM

SK • •

4K The p-value is for the • • linear regression fit drawn; a smaller p-value would vi • • indicate greater statistical u... • g_ significance. There is no 3K recommended threshold for ~ statistical significance, E.. but typically 0. 05 is used as ~ this: is associated with a 5% iii • • risk of a Type t error or false 2K • • positive. e • • • • 1K • • • • • • • • • • • • • • • • • OK 19SO 19S2 19S4 1966 1966 1990 1992 1994 1996 1996 2000 2002 2004 2006 2006 2010 2012 2014 2016 Water Year

Prado Basin Riverside County California

Value = 25.3171*Water Year + -48725.8 ANNUAL PEAK INSTANTANEOUS STREAMFLOW R-Squared: 0.0356223 TEMESCAL CREEK ABOVE MAIN STREET P-value: 0.292843 1 . ' Ar:ID}r Cmrp!S, of :Engine,,ers Lo!S, Ang,elies D isfiri.ct

Plate 3 1) Choose a HUC-4 2) Click Map Location or Name to Select Stream 1807.Southem C alifornia Coastal Gage

Site Number Search for Gage wirhin HUC-4 by Name f·iti8000 All 11 117600 11046360 1118 11046350 3) Include Only Years /If Desired} 11073495 CUCAMONGA C NR MIRA LOMA CA 1757 to 2016 11073470 11086000 11067000 110.A~Ann f>.nnua l Peak Instantaneous Streamflow, CUCAMONGA C NR MIRA LOMA CA Selected (Hover Over Trend Line For Significance (p) Value) ~limate Hydrology Assessment Toolv.1.0 Analysis: 11/8/2017 1:S0 PM • • 15K • The p.value is for the • linear regression fit drawn; a smaller p-value would iii • indicate greater statistical ~ 10K • significance. There is no recommended threshold for ~ • • statistic.al signif,cance, e • • but typically 0.05 is used as .. • this associated w;th 5% ~ is a iii • risk of a Type I error or false positive. SK • • • •• • • • • • • • •• •

1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 Water Year

Value = 162.808*Water Year + -319368 Prado Basin Riverside County California R-Squared: 0.230916 ANNUAL PEAK INSTANTANEOUS STREAMFLOW P-value: 0.0008348 CUCAMONGA CREEK NEAR MIRA LOMA, CA U.S Army Co1·ps of Engineers Los Angeles District

Plate 4 Nonstationarities Detected using Maximum Annual Flow/Height Pareme!&r h t&ctlon @1n$bnt.,neou:iPciik.Slf~ S"9• s.tta SeJ&etlon Select a state 20K CA

Select a site

11059300 • SANTAANA RAE ST NR $.A.H • • • '°" n merreme se1&et1 0n 1967 1994 0-0

81:n u tlvr!)o: Fe.rur,e,c," 1990 ~ .e!\Se'Y.~ p...-=.,,-e ,:'e~ ~:!!~ n:.,.-:,,:;•,11t =~ J:.¢~iSfe.)l..'t"r-'$fff.!O'l-c'e.~ll\';:.,.,a~= .,,-e ; e.'el'.fe=:). Thi:; gage h:is ,1 d1mr1119e lll"Cll cl 541.0 :;qu~c: miles. L:rg;:ir~ v.Q~!n fitN«~olk'lt'o11'll~O(lMJIIO. ViARUING: Tit~ pe:IO!J o(RC011'.i W~!ed IS !00 Sl!Oft rr.e:e aR ;.;«: nri'al tsSl.i':S WMU!~ a.:,-;g:y®!S ~recrea. CPM fl&thoCl e. Bum•ln P&IIOO _J_Od'olllt:.201._ 20 The USGS :ilfe,1m8ow g,:igc siles :,v.:iilllble for ,1:;se:i:imenl willwl tii:i :ipplic~ indude loc;,lion:; \ffle--e the--e :ire di:icortWlla.'lic:sin USGS pc::ik low 0 < > d;,t, ccleclion llvoughout ltte period or record imd 9:,9~ with :ihort reccn:l:i. EnginccrSlg judgment should be exen::i$ed when tlll'f}'ingout :in.:ily:ii:i \mer e ltterc: ,11e signik:.t1I d;,t, g.,p:i. CPM fl &thoCl e. In gcner:.I. :, 11W1imum or30 ye:,r:; or conlinuo~ :itte:.ml'low me,.,:iurc:men~ mu'l'-1 be .1v:tibbae before lhi:i :ippk:.tion :ihould be u:ied la d~ed sane.lflvty t1011?.:,!ior.., rilies S'I flow record:;. i0.\11.A. ~.OC:O) _ 1,000 Heatmap . Graphical Representation of Statistical Results 0 < > Cramer-Von-1/ises (CPI.I{) Kotnx,gorov.Smrnov (CPM} Bayee.lan sane.tflvty -~olllt:.051 1.ePtlCt (CPU) 0.5 Etl8'!1f C'ivi$i,.,9 Mtlllod 0 < > Lomt:-ard 'Mlcoo:on Penrtl Energy OM91ve fl&thod Sena1ttvty Mann-Whr.ne)' (CPl,1) -~olllt:.051 0.5 8ay~an - 0 < > Lomb.vdMood - Mood(CPM)

Smooth Lombard Wilcoxon Smooth LOrntlard lilOOCI Lom D:J ra Smooth fl&thoCl e. SSne.tfMty 1965 1970 1975 19'0 1985 1990 1995 (OQ".-11A:_0.~) 0.05 Legena • !Y.P.:e or s e cany: SJgn1ncan c n:ange b&Jng oetactad 0 < > ■ V,1n..,noc: ■ 11,c::.n ■ Smoo~ Pettttt Sena1ttv1ty Mean and Variance Between All Nons tationarities Detected IOQ".-11A:0.~) 0.05 °' 0 < > Sc:9(11Cnt Me11t1 "' (CFS) 2K 6K. ~~hUSJllrrrt~«~ff.fot Sc:9f11Cnl Smnd:.td Oevi:ilion 4K O"(d.,dnglnlst0"1$t;J.<)n,VA',4Cli:.:L4nb31M-l)olrto1Ni' (CFS) 2K ~5hdio'!l:lllt~.¥1d,-,:.fi;nc:,,;o;ancllm:,i;,g il ~.w.11;m11. ----JOIA ------Sc:9(11Cnt \/;,rmnc;e ,.,,, (CFS Squ:irc:d) 10M

1965 1970 1975 1980 1990 1995

Prado Basin San Bernardino County California NONSTATIONARYITY ANALYSIS OF MAXIMUM ANNUAL FLOW

SANTA ANA RIVER

f.iiiir.il U.S Ar Octobermy Co 2020rps of Engin•••·• ~ Los Angeles District

Plate 5 Nonstationarities Detected using Maximum Annual Flow/Height Parameter Selection ------@ Instantaneous Peak Streamflow 0 Stage

Site Selection 10K- Select a state CA

Select a site 11073360 - CHINO CA SCHAEFER AVENUE N ..

SK - Timetrame selection 1969 to 2065

Sens it iv ity Parameter s (~ nsnivity parameters are described in the manual. 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 Engineering judgment is required if non-defautt parameters are selected). Water Year Larger Values will Result in fewer Nonslationartties Detect e

CPM Methods Burn-In Period 20 The USGS streamnow gage sites available for assessment within this application include locations where there are discontinuities in USGS peak flow data collection throughout the period of record and gages with short records. Engineering judgment should be exercised when carrying out CPM Methods Sensit ivty analysis where there are stgnificant data gaps. 1,000 In general, a minimum of 30 years of continuous streamflow measurements must be available before this application should be used to detect nonstationarities in now records Bayesian Sensitivty 0.5 Heatmap - Graphical Representation of Statistical Results Cramer-Von-Mises (CPM) I I Kolrnogorov-Smimov (CPM) I Le Page (CPM) I I Energy Divisive Method Sensitivty Energy Divisive Method 0.5 Lombard Wilcoxon I Pettitt I Mann-Wllltney (CPM) I I Smaller Values will Result in Bayesian More Nonstationarities Detected - Lombard MOOd Mood (CPM) Lombard Smooth Methods Sensitivity 0.01

Smooth Lombard Wilcoxon

Sroooth Lombard Mood Pettitt Sensitivity 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 0.05 Legend • !Yoe of Statistic3IIY Sianific3nt Change being Detected I■ Distribution ■ Variance ■ Mean ■ smooth Mean and V ariance Between All Nonstationarities Detected

3K Non-default value select ed for one or more sensitivity parameters. segment Mean 2K- (CFS) 1K

2K- Segment Standard Deviation (CFS) 1K OK 6M Please acknowledge the us Army Corps of segment Variance 4M - Engineers for producing this nonstationartty (CFS Squared) detection tool as part of their progress in climate 2M - preparedness ancl resilience and making it freely available 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

Prado Basin San Bernardino County California NONSTATIONARYITY ANALYSIS OF MAXIMUM ANNUAL FLOW

CHINO CREEK

U.S A Octoberrmy C o12020·ps of Engineers Los Angeles District

Plate 6 Noostationarity Det ector Trend Analysjs Method Explorer

P~re.m ~t!.1 ~~:eol!OJI Nons b tionarities Detected using M axim um Annu-31 Flow/Heig'.ht @ lt!st.:w1Wlllflll$ ~ Slrfl¥nlfl• 7 ""'

CA

11072t00- TEI.IESCALC ASUll.l'f ST A •••

Tlnletraint h lt otlon

2015 IK 0--0

~.. , .11 ,,.,.,, ...... ,,.. ,,., (l.. • ~ l'l<'l)',:;o.-.,..,,.:o -. ... :'u~·.:• i e i,o.e:s.,:co(. "'' ; , ;-..e.- e., ; .r.:.:\;"'.. ... ,i ""~".. ..6 ,,, ,,..... !J :.1"~ ··"'•:-...... :-• :t•:;I.

CJ'lil, l.'.: ltloa, S lim ~ P<: rk14 _.l~Jll• 20 ThoUS GS lil,_-,,....g.:,g11 ~ 4"4~ 11 l'Ot 4U4liStNM •'llllh ll'le.. .!ll)l)llc4o.tln lndl.$1Ge41loM wl>IIIO r.ota .:irft Cllsellnl~4li h US GS pi:i,~ Jaw 0 ~~ <1'114 c.of4<:IIM 11'>~ ll'lll l)IIN)d o,,.oo,o .1M911911t.•th $IIClt'I ro,:O)l'(IS,. e~~fll SllOIMOII ~-a vd'.On c.vr-,lng 0111 4!'!,!ll~lt, •ll-ll'lll!'O-liqt,l'l!'.¥11<1'1149/ll)li. CJ'll lh lhOClt. h nct~My 1t1911Mr.:it,.i...,;ml.l'!t0130)"04f't.ol((Wll~li!,_llowtn44-m.l'tOO;,,,~~lnlt.4At)le41lMSIIO.MOOu 544IO» (Cl"l.l) - Sl"-«1111 Lcm1;1;11"(S wrcoxon ~--•--"-''" Sl1'lOOl1I LC/llb9fCU04!tJ ----,. Lo nt.)t--­t4 ~moolh Metllocr~ h n lMM tt ;90S .,,. 0 ~~

Pettitt h n UIM ty M ean and Variance SetY'Jee.n Al N-onstationari1ies Detected - -=•,~=·-- """~ 0 ~~ "'"" •---~u.. uil,.,...,°""""~i.,, &cqNM S l4fld4!'0 011~n """ ----..,. , ___ _ d u .., (CFS! 5~ I - ••------'""'""'°MIid 150CK➔ &cqNM V~o> 1(00t( -- (CFS Sqr..~ SOCK ,... l~ 1005 '°"

Prado Basin Riverside County California NONSTATIONARYITY ANALYSIS OF MAXIMUM ANNUAL FLOW

TEMESCAL CREEK

U.S Ar my Corps of EnginoHs lr.l"r.l October 2020 Los A~ele> Dfabid ~

Plate 7 Noostationarity Detector Trend Analysjs Method Explorer Nons t3tionarities Detected using M aximum Annu-31F low/Heig'.ht P,11.r ,11.m, ti.1 ~,ll~ OIIO!I @ 11'15t.41WIIIOU$ ~ ~ao¥1'doa

M 7 t; "" - • CA • 1 i ,;;• '"' ~ 11073-'!$ • CUc.AMOttGA C NR MlRAlO• •• l • Tlnletra~ e h lt otlc111 f< "' 1979 2015

~.. , .,,.. .., I"\,...... , ... . 2000 (l-.•.si.'><'l)',:o,,...,,.:o-. . ..iu ~·.:• i i,, i , • .,.o.,:cot ;,;-..e.- ~,; .r.:.:\;"'"'"''~~ "... ..6 ..,, ,,,.__..,,,.,.~ ~ -· "'• : ...... -.:1 .,.i. t.._-•---..1 .. t ____ CPM lo'.: ltlo,a S t.rm .. P,11 rlo<: _..l~Jll• 20 ThoUS GS li!,_..lbw9.:,g11 ~ .i-,;11&1,1:1111 t'Ot ;llUllliStNnl • i ll>h IM .!ll)l)llc4wln lnCll.delGei\lloM wl>ll~ r.ota .:irft Cllsellnlln.•llli h US GS pi:i,~ Jaw 0 ~~ Clol!;ll «dllCUM l'I~1!'1111111....,., o,,.!)0(0 ~9/19llli•· $11(1t'!tO(O)l'(IS,. e~~fll SllOIMOII ~lld vd'.On c.vr-,lng 0111 ;lll'!,!ll~lli •h-1!'111!'0- liqw'l~Clol!;ll ~- CPl.o\11.elhod ~ h ai ct~My ll'l{llll"#/11,.t...,.ml.mOI 30~liol~ J;!-'iowtn4;ll-1'!11.1'tOO;,,,,~~ lnlli ~ l ( i\llcwt SIIO.MOOu 54CIIO a , yn l~/1 t,Mt~My -~ "~' l ~l"i~(CrM; .. E««,y Otv~ tt,: 1,Mllutl 0 ~~ lOl'nbiW'1'.l \\'IWCCn ~II EL-U Q)' Ol'fl ~ t Met!IOCI h ai u llvty ,.,.._..,W.'ltm:v(CPI.C) 1\.W.<11 ~~· ~ ,::111m -r .. t 0 ID@ lomb:11'11 Unnd A.b:tcl (CP >.1) - sr-.oom Lcmb;lf'(S Vll'Cl))'Ul l-1,-"'l.oo• ..."-1 1o, Sfflool1I LC/llbllfC UoetJ LOR.) ~tC ~m oo lh Mef!lcd ~ h 11UIMty ~ 7 --­ 1900 193; 1990 ,.,, 200, 2010 201;

0 ~~

... um o.. , ,.. u ~u-, M ean and V3ri3noe Setwee.n Al Nonst3tion3ri1ies Detected --=•·~=·-- 0 ~~

'""-•odl~u.. us,.,...,o..:-ood~lto S<:qNnt Sl4nc1;:iro D1t.-..:a60n -e....-100.... , .,_..IWl•S.oo'ldU"'• - ••-P'-----•os...,-u1111 (CFS! =t ~ OM S<:qNnc v:ir.:inc:~ iOt.c -- (CFS ~

~· ~,.., ,.., l!l;)S 2(11 3 "" ""'

Prado Basin Riverside County California NONSTATIONARYITY ANALYSIS OF MAXIMUM ANNUAL FLOW

CUCAMONGA CREEK

n s A.rmy Corp~ of F.ngtnPPr-s r.iiif.il October 2020 Los Ang t"l<'s District l=i

Plate 8 Range of 93 Climate-Changed Hydrology Models of HUC 1807-Southern California Coastal 1) Choose a HUC-4 1807-Southem California 300K Proj@a ed Routed Runoff not biased comm@d. Not for use in quantitativ@assessm @nts. Coastal 2) Change Displayed Date Range of Modeled Data (ff Desired} 250K 2018 to 2099

Legend in u. ■ Mean of 93 Projections 8. ~ 200K ~ Range of Projections i ~ iii

SOK

CMIP-5 Data, Downscaled to HUC-4 leve/ via BCSD Method, Based on 93 combinations of GCMIRCP model projections Climate Hydrology Assessment Tool v.1.0 Analysis: 11/8/2017 2:25 PM

Prado Basin Riverside County California RANGE IN THE PROJECTED ANNUAL MAXIMUM MONTHLY FLOWS, HUC 1807 U .S Ar·my Corps of Engineers Los Angeles Di.~trict

Plate 9 Trends in Mean of 93 Climate-Changed Hydrology Models of HUC 1807-Southern California Coastal 1) Choose a HUC-4 1807-Southem C alifornia (Hover Over Trend Line For Significance (p) Value) C:.o;i~t;:il Projected Rout ed Runoff not biased corrected. Not for use in quantitative assessments. 2) Select Year Dividing "Earlier" and "Later" Periods 25K 2000 Date Period Later iii' ■ g~ ---0------3) Change Displayed ~ 20K Date Range of ~.. Modeled Data ~ (ff Desired) Iii ;,. 2018 to 2099 =C 0 Trend line ::; § 15K -E~ ::;

..~ C C < ] Trend line: .!l_ 10K 0 Annual Max. Monthly Flow = 5.80485*Year of Water Year + 9864..29 a: R-Squared: 0.0029382 The p.value is tor the 'o linear regressk>n fit drawn; a C P-value: 0.62862 smaller p-va/ue would .. indicate greater statistical ::; " significance. There is no recommended threshold for SK statistic.al signif,cance, but typically 0.05 is used as this is associated with a 5% risk of a Type I error or false positive.

OK : Year = 2000 2000 2010 2020 203) 2040 2050 2060 2070 2080 2090 2100 Water Year CMIP-5 Data, Downscaled to HUC-4 level via BCSD Method, Based on 93 combinations ofGCMIRCP model projections Climate Hydrology Assessment Tool v .1.0 Analysis: 11/8/2017 2:26 PM

Prado Basin Riverside County California MEAN PROJECTED ANNUAL MAXIMUM

MONTHLY STREAM FLOW, HUC 1807

U.S Army Corps of Engineers Los Angeles Dish'ict

Plate 10 ■ Observed ■ HadGEM2-ES (Warm/Drier) ■ CNRM-CMS (Cooler/Wetter) ■ CanESM2 (Average) ■ MIROCS (Complement) 95% Confidence Intervals

J

"in'., 6 .s= .s0 .s 5 C ~ .2 • ·a. 4 ·o /r • • .; 3 • .,"' • • _J • • • C 5 2 .; a:

oJ Oct 1961 - Sep 1990 Oct 2035 - Sep 2064 Oct 2070 - Sep 2099 Historical Mid 4 Century Late 4 Century

Source: Cal-Adapt. Data: LOCA Downscaled Climate Projections (Scripps Institution of Oceanography), Gridded Historical Observed MEteorological and Hydrological Data (University of Colorado, Boulder). Four models have been selected by California's Climate Action Team Research Working Group as priority models for research contributing to California's Fourth Cl mate Change Assessment. Projected future climate from these four models can be described as producing: • A warm/dry simulation (HadGEM2-ES) • A cooler/wetter simulation (CNRM-CMS) o An average simulation (CanESM2) • The model simulation that is most unlike the first three for the best coverage of different possibilities (MIROCS)

Prado Basin Riverside County California PROJECTED 100-YEAR 24 HR Rainfall RCP8.5

~ U.S Army Corps of Engineers ~ Los A.ug eles District

Plate 11 Maximum Temperature

Maximum daily temperature which typically occurs in the early afternoon.

92 OBSERVED HISTORICAL 2051 <- Extended Drought Period (20 years)-+ 2(170 91 1961- 1990 Average t, 90 89 "'5 88 78.9 °F .; 87 ~ Q. 86 E DROUGHT SC ENARIO 85 ~ 2051 - 2070 Average E 84 :, E 83 ·;a 82 86.9 °F ~ 81 80

2045 2050 2055 2060 2065 2070 2075

Water Year (Oct -Sep)

Minimum Temperature Minimum daily temperature which typically occurs in the early morning before sunrise.

61 OBSERVED HISTORICAL 2051 <- Extended Drought Period (20 years)-+ 2c,10 60 1961- 1990 Average t, 59 ~ 58 49.6 °F ~ ~ 57 "'Q. E 56 DROUGHT SC ENARIO >-"' 55 2051 - 2070 Average E :, 54 E ·c 53 57.6 °F ~ 52 2045 2050 2055 2060 2065 2070 2075 Water Year (Oct - Sep)

Prado Basin Riverside County California PROJECTED TEMPERATURE RISE AND LATE

CENTRUY DROUGHT

lr.Pir.il U.S Army Corps of Engineers ~ Los An.geles District

Plate 12 Summary of HUC Results Select a HUC or HUC.s to show the districts in each HUC and a summary of the vul rn erable HUC:s and Clirnale Oal.i ln1egra1ed Business Line Threshold ORness indicator contributions to those HUC:s .• Source Analysis Type /Jataset 212016 - data update for selected jndir:a/ors Flood Risk Redllclion CM IP-5 (2014) EACH 20% 0.70 WOWA Score Dry Wet l 5.66 76 .07 Dry We! r c;, LJ"J 1 I-I UC(s} selecled 1 I-I UC(s) selected 0 N 1 HUC(s) vu lnerable 1 HUC(s) vulne Colora , LJ"J cO 1 I-I UC(s} selecled 1 I-I UC(s) selected 1 HUC(s) vulnerable 1 HUC(s) vu lnerable 0 ·alitorr, ·aliforr, ~ an 0 N Ory Wei New New 0 Mexico Vlex1co LJ"J 0 N

LJ"J cO c::> N Ba fl Ba1a ~ Indicator ■ 175C Annual Cov ■ 590 500YR Urban Floodplain l ■ 277 Runoff Precipitation Utcli U 2r 568C Flood Magnification Colorad ■ ■ 568L Flood Magnification an ·alitorr, ·alitorr, 00 0 HUC Dislricl National N 1807 SPL Standard New New Setting~? Mexico Vlexico Yes

Prado Basin Riverside County California

PROJECTED VULNERABILITY FLOOD RISK REDUCTION BUSINESS LINE FOR HUC-1807 .' , Ar:mr1r CorpiS, of Engineers L os. Ang,eles Disflrid

Plate 13 Summary of HUC Results Sel'.ect a HUC or HUC:s to show the districts in each HUC and a summary of the vulnerable HUCs and Climate Data tntegra1ed Business Une Thr,eshold ORness indicator contributions to tho,se HUC.s .. Source Analysis Type Dataset 2/'2016 - data update for selected indicators E cosystem Restoration CMIP,5 (2014) EACH 20% 0.70 WOWA Score Dry Wet I I 65.36 79.65 Dry Wet

f 0 1 1-1 UO(s) selected 1 I-I UO(s) selected 0 "'N 1 HUO(s) vu lnerable 1 HUO(s) vu lnerable u .-.ti Color ad Utc'l Colora , "'a:, 1 I-I UC(s) selected 1 I-I UC(s) selected 0 1 HUO(s) vu lnerable 1 HUO(s) vu lnerable 0 ·alitorr. ·a itorr, N in 0 N Dry Wel New New 0 Mexico Mexico Lt, 0 ~ N "'a:, 0 N B i sa·a ~ l'ndicator 70 Low Flow 8 At Risk Frshwtr Plant 277 Runoff Precip. ■ ■ ■ Reduction f ■ 65 Mean Ann Runoff ■ 297 Macroinvertebrate .-.ti Utc'l 156 Sediment 568C Flood Magnification u Colorad Colorad ■ ■ 221C Monthly Cov 568L Flood Magnification ■ ■ in '"a itorr, 'il iforr, CX) 0 HUC District National N 1807 SPL Standard New New Settings? Mexico Mexico Yes

"

Prado Basin Riverside County California PROJECTED VULNERABILITY ECOSYSTEM RESTORATION BUSINESS LINE FOR HUC-1807

1 • , A:r:my Corp.s of :Engineer-s Los Angel1es Distcri.ct:

Plate 14 Summary of HUC Results Select a HUC or HUCs to show the districts in each HUC and a summary of the vulnerable HUCs and Climate Data Integrated Business Line Threshold ORness indicator contributions to those HUCs. Source Analysis Type Dataset: 212016 - data update for selected indicators Water Supply (selected HUCs) CMIP-5 (2014) EACH 20% 0.70 WOWA Score

Dry Wet 68.382 70.816

Dry Wet

:is 1 HUC(s) selected 1 HUC(s) selected ~ 1 HUC(s) vulnerable 1 HUC(s) vulnerable

:lJ 1 HUC(s) selected 1 HUC(s) selected ~ 1 HUC(s) vulnerable 1 HUC(s) vulnerable 0 It) 0 N Dry Wet

0 0 "'N

Indicator ■ 95_DROUGHT_S .. ■ 277_RUNOFF_P.. ■ 156_SEDIMENT ■ 175C_ANNUAL_. . ■ 221C_MONTHL¥ ..

It) 00 0 National N Standard Settings? Yes

© OpenStreetMap contributors

Prado Basin Riverside County California PROJECTED VULNERABILITY WATER SUPPLY BUSINESS LINE FOR HUC-1807 U .S Army Corps of Engine-ers Los Angeles District

Plate 15

Attachment A

Prado Basin, California, Water Conservation Memorandum of Agreement – 7 July 2006

Prado Basin Feasibility Study May 2017 Water Conservation & Sediment Transport Analysis Report 89

Attachment B

Evaluation of Suitable Discharge Rate from Prado Dam to Surface Water Recharge System – February 5, 2013, OCWD

Prado Basin Feasibility Study May 2017 Water Conservation & Sediment Transport Analysis Report 90

Attachment C

Prado Basin Daily Discharge Estimate for 2021 and 2071 Using the Wasteload Allocation Model - December 2013, Wildermuth Environmental, Inc.

Prado Basin Feasibility Study May 2017 Water Conservation & Sediment Transport Analysis Report 91