Milliken Stormwater Master Plan

1.1 Authorization

This report was authorized by the Town of Milliken under the April 2014 agreement regarding "Storm Drainage Master Plan, Utility Fee Assessment, and Storm Drainage Criteria Manual Update". Notice to proceed was provided by the Town of Milliken on April 16, 2014.

1.2 Purpose & Scope

ICON Engineering, Inc. has been placed under contract by the Town of Milliken to prepare an update to the Town's 1982 drainage Master Plan. The purpose of this report is to evaluate the watershed tributary to the Town, identify existing system deficiencies, and to develop an alternatives plan for mitigation of flood problems, stormwater conveyance, and water quality improvement. Methodology and calculations for this analysis focus on the 100-year storm event, although calculations for the 2-, 5-, 10-, 25-, 50-, and 500-year return periods were also developed.

The following is a summary of the scope of work completed for this study:

1. Coordination and meetins with the Town and project stakeholders;
2. Hydrologic Analysis —
1. Obtain and review previously published hydrologic information within the Milliken watershed;
2. Review and provide adjustments to land use parameters and subwatershed boundaries;
3. Utilize CUHP v1.4.3 and EPA-SWMM v5.0.22 to complete the hydrologic analysis;
4. Develop hydrologic discharges and hydrographs for the 2-,5-,10-,25-,50-,100-, and 500- year recurrence intervals for both existing and future development conditions;
3. Hydraulic Analysis —
1. Evaluate the hydraulic capacity of existing channels, conduits, streets, and culverts;
2. Evaluate impacts of residual flooding throughout the Town;
4. Alternatives Analysis —
1. Creation of baseline hydrology;
2. Development of project alternatives considering conveyance, detention, water quality, stream stabilization, and conduits;
3. Development of alternative cost estimates;
4. Recommend an alternative plan for conceptual design.

Hydrologic information presented in this report is based on current (existing) and projected (future) land use conditions. Prior to the use of this report for planning or design purposes, the user is advised to contact the Town to determine if the information presented in this report has been amended.

1.3 Planning Process

One hydrologic study has been previously developed for the Town of Milliken, as part of a report entitled "Flood Plain Information, Drainage Plan and Street Improvement Plan at Milliken, Colorado" 1, dated October 1982. Additionally, in 1979, the Federal Emergency Management Agency (FEMA) completed a Flood Insurance Study (FIS) for the Milliken area, however, this study focuses on the Little Thompson River, and not areas tributary to downtown Milliken.

Progress meetings were held to discuss project status and technical approach for the project. Philosophical goals for alternatives were discussed at periodic progress meetings.

Within the Town of Milliken, the goals for the project include:

1. The reduction of the threat to human life and property caused by localized flooding;
2. To provide a basis for the management of future development within the Town;
3. To determine the most economically feasible flood control alternative for local drainage problem areas;
4. Increasing detention to mitigate hydrologic changes due to land use;
5. Increased conveyance capacity at canal and railroad crossings;
6. Use of regional detention to minimize required conveyance

A kickoff meeting was held on May 8, 2014 to introduce project sponsors and understand the goals for this Master Plan. Subsequent meetings were held regularly to discuss hydrologic and hydraulic results, alternative plans, and to review comments.

This master plan, like its predecessor, is developed in close coordination with the greater flood recovery planning process. The flood event on September 12, 2013 triggered the immediate and long term responses to increasing risk awareness and resiliency. The Colorado Water Conservation Board has a full history of the 2013 flood event.

The Town’s master plan process follows closely along with the local, regional, state, and federal recovery, planning, and mitigation processes to ensure a comprehensive and coordinated plan for the Town’s resiliency to flooding and other related natural disasters. At the time of this report, there are still many technical and administrative details related to the flood recovery that are undetermined. For example, flood studies on the Big Thompson and Little Thompson Rivers have not published revised hydrologic estimates for the respective basins. Therefore, all flood recovery mapping and related work on these flood sources has used the effective FEMA hydrology. Notably, the flood recovery mapping provided by the Colorado Water Conservation Board, on the Little Thompson through the Town of Milliken is based on FEMA effective flows and fully blocked crossings. As additional details are provided, these studies may be revised. It is possible revised base flood event hydrology for the Little Thompson River could be released in 2015. Other mitigation grants and programs also remain in progress.

There are a number of grants and programs that were coordinated with the development of this master plan. The Department of Labor and Affairs (DOLA), Hazard Mitigation Grant Program (HMGP), and Community Development Block Grant – Disaster Recovery (CDBG-DR) were all programs the Town Staff and associated partners have spent considerable time through 2013 and 2014 to leverage the best possible outcome for the Town and its residents.

A unique and noteworthy point for this master plan is the coordinated public outreach between the State Flood Recovery mapping project, grant programs, and the master plan. Summarized by the Town’s public meeting records, nearly every public meeting on any one of the aforementioned projects or plans related to the Town’s master planning process. As drainage knows no administrative boundaries, stormwater master planning involves local, regional, state and federal efforts to ensure a coordinated approach.

As of the date of this report, the Floodplain Recovery Mapping is complete – the Town’s adoption of the floodplain was recorded in March 2014. The CDBG Traunch 1 application was submitted, funded, and implemented. The CDBG Traunch 2 application has been submitted and is awaiting award notice. The HMGP application was submitted in August and is awaiting award notice.

1.4 Mapping & Surveys

LiDAR mapping, completed in January 2014, was obtained from the United States Geological Survey (USGS) for use on this project. The LiDAR dataset included 0.35-meter ground sampling distance, provided on NAVD 88 vertical datum and the NAD 83 horizontal datum. Supplemental ground survey was completed by Lamp, Rynearson & Associates in summer of 2014. Additional mapping for existing utilities was provided by the Town of Milliken in CAD format.

1.5 Data Collection

Multiple drainage studies and as-built construction documents were collected from the Town of Milliken. The primary references used for this study are as follows:

• Town of Milliken Comprehensive Plan Framework Map, 2010
• Town of Milliken Zoning Map, from Town Website, circa 2012
• Town of Milliken Estimated Growth Areas, from Town Website, created circa 2011-2012
• Tetra Tech RMC, January 2004, Final Drainage Report for Pheasant Hills
• Rocky Mountain Consultants, (RMC), July 2000, Final Drainage Report for Centennial Farms Subdivision
• RMC, April 2000, Supplemental Drainage Study for Centennial and Pheasant Hills Master Plan
• RMC, August 2000, Preliminary Drainage Report for Settlers Village of the Centennial Development
• RMC, July 2000, Final Drainage Report for the Colony Pointe Subdivision
• Loonan & Associates, November 1999, Quentine Avenue Storm Sewer As-Built Drawings

Acknowledgements

This report was prepared with the cooperation of the Town of Milliken. The representatives who were involved with this study are listed in the following table:

2.1 Project Area

The watershed area studied for this Master Plan update is approximately 19.8 square miles in size, divided into five main flow paths, designated herein as the Western Tributary, the Lola Park Tributary, the Eastern Tributary, the Big Thompson Tributary, and the Northern Tributary. The Lola Park and Western tributaries discharge into the Little Thompson River, while the Northern, Eastern, and Big Thompson tributaries drain to the Big Thompson River.

The Northern Tributary is located north of the Big Thompson River, and has been annexed by the Town since the publication of the 1982 stormwater master plan. As with the other tributaries, it consists mainly of open space rangeland, with some single family residential development centered around the Mad Russian Golf Course. The Northern Tributaries are generally long, thin watersheds, each discharging to a separate point along the Big Thompson River. Although some of these areas are outside of the Town limits, they were included in order to properly account for any future changes that may occur related to the Town’s road right-of-way in that basin. The Northern Tributary includes several watersheds directly tributary to the Big Thompson River, downstream of the confluence with the Little Thompson River. These watersheds discharge across a portion of Town property which connects the future Homestead at Ashton subdivision to the Old Town area. The Northern Tributary ranges in elevation from a high of 5050’ to a low of 4710’ along the Big Thompson River. Soils are predominately type B, with a small amount of type C soil located in the higher portion of the watershed. The Northern Tributary is steeper than the other tributaries, with an average slope of approximately 2.2%.

The Big Thompson Tributary is located east of the Union Pacific Railroad (UPRR), south of the Big Thompson River. The Settler’s Village development is located in the Big Thompson Tributary. It also contains isolated pockets of industrial development, as well as the Town water treatment plant. This tributary is relatively flat river bottom area, with elevations ranging from 4748’ to 4718’. The soils are mostly type B, with a small amount of type D soil in the south. Average slopes across the Big Thompson Tributary are approximately 0.4%.

Runoff generated by the five tributaries is conveyed through both open and closed conduit systems, depending on location. In undeveloped areas, conveyance is almost exclusively via open channels and roadway crossings. Storm sewers and roadways (including borrow ditches) are used to convey stormwater runoff in developed areas.

A total of five existing formal detention facilities were considered in this report. They include Ehrlich Lake, the Settler’s Village Pond at County Road 46 and CR25, Centennial Lake, located north of the Settler’s Village Pond on State Highway 60, and the Centennial Farms Pond at the intersection of Alice Avenue and County Road 46. The Pheasant Hills Sump, located on CR21 at the Lola Park Tributary crossing, was also included in this study, despite its lack of a formal outlet structure. Alternatives for additional detention facilities, such as ponds on the downstream end of Lola Tributary, are discussed later in this report.

2.2 Land Use

Zoning data was provided by the Town of Milliken for use on this project. Zoning maps were provided in pdf format and digitized for use in GIS software. Table 2.1 below lists the impervious values listed for various land uses. Land use values were applied consistently with other planning studies, specifically the Hydrologic Analysis of the Box Elder Creek/Cooper Slough Watershed.2 Detailed watershed parameters are discussed in Section 3 of this report.

Existing land uses were based on current zoning and development levels. Currently developed areas consist mainly of medium density, single-family residential areas in addition to some estate lot residential housing. Commercial uses are mainly limited to areas within one block of Broad Street, with the exception of a commercial strip mall along Alice Avenue just south of SH60. Some industrial uses exist, mainly in the Big Thompson Tributary. The majority of the areas within the Town boundary are currently undeveloped, existing as open rangeland or irrigated cropland.

Future imperviousness values were determined based on zoning, and a map entitled “Estimated Milliken Growth”, provided by the Town (included in Appendix D). This map shows areas within the Town boundary that are currently under construction, have pending development applications, expected future annexations, and land expected to be developed. For the future land use conditions, impervious values anticipate development based on land use and zoning information collected from the Town. Inside the Old Town area, the watershed is primarily developed, resulting in similar land use for both existing and future conditions. South of Old Town, higher impervious values result from the planned development areas.

2.3 Outfall & Tributary Description

Due to the various outfall locations along the Little Thompson and Big Thompson Rivers, drainage areas tributary to the Town have been broken into 5 tributaries, the Western Tributary, the Lola Park Tributary, the Eastern Tributary, the Big Thompson Tributary, and the Northern Tributary. Generally, stormwater runoff is conveyed by open channels higher in the basin. Where runoff enters the developed portion of the Town, it is generally split between open channel (streets and drainage channels) and storm sewer conveyance.

2.4 Flood History

The 1982 stormwater master plan references several past flooding events affecting the Town. These include a rainfall event in June of 1981 which resulted in 2 to 2.5 inches of rainfall over the watershed. This event caused localized, limited damages, including several roadway washouts, but no major structural damage reported. Another event was noted in June 1982, when up to 3 inches of rain fell in one hour. This storm damaged approximately 50 structures, and overtopped the Union Pacific railroad on the Eastern Tributary, in addition to extensive private and public damages along County Road 23.

Milliken was also severely impacted by the flood of September 2013. Between September 10 and September 14, 2013, a rain gage in the Lola Park Tributary recorded a total of 4.28” of rain, 2.2” of which occurred on September 12. Flooding below Broad Street was widespread, due to overflows from the Little Thompson River. These flows caused structural damage to the UPRR crossing at the Little Thompson River, as well as the spur line that parallels Broad Street. Flooding from the Little Thompson also caused the intersection of Broad Street and Quentine Avenue to flood (photos 4-7).

3.1 Overview

This study will update the Town of Milliken hydrology for the 2-, 5-, 10-, 25-, 50-, 100-, and 500-year storm frequencies under both existing and future development conditions. The Colorado Urban Hydrograph Procedure (CUHP) version 1.4.3 was used to develop runoff hydrographs from individual sub-watersheds within each watershed. The EPA Stormwater Management Model (SWMM) version 5.0.22 was used to route the hydrographs through the watershed.

The study area for Milliken encompasses 19.8 square miles, divided into five main flow paths. A total of 81 subwatersheds are included in the hydrologic analysis. The average watershed size is approximately 156 acres. Due to the size of the watersheds, a 5-minute unit hydrograph was used in CUHP, per recommendations in the Urban Storm Drainage Criteria Manual (USDCM).4 For the current update, the overall watershed boundary was determined to be very similar to the watershed boundary used for the 1982 stormwater master plan, although the subwatershed delineations were slightly different.

3.2 Design Rainfall

3.3 Subwatershed Characteristics

For the current study, subwatershed boundaries were re-drawn based on the LiDAR topography described above. Initially, the subwatersheds were delineated based on 10-foot contour mapping, and the 2- or 1-foot mapping was reviewed as necessary to adjust the boundaries. The subwatersheds range from 4.8 acres to 1,114 acres, with an average size of 156 acres. Generally, subwatersheds in developed areas of Town are delineated on a finer basis than the undeveloped areas. Areas outside of the Town boundary directly tributary to the Big or Little Thompson Rivers were excluded from this study. Additionally, the canals within the study area were assumed to be flowing full during a rainfall event, as was done with the 1982 stormwater master plan. Therefore, their influence on basin delineation was ignored.

Imperviousness for each subwatershed was based on the updated land use and zoning information described in Section 2.2. For the future conditions, percent impervious values were generated to reflect anticipated future development based on land use and zoning information from the Town of Milliken. Impervious percentages were applied from recommendations presented in the USDCM, or taken directly from drainage reports, as applicable. For developed areas, impervious values were generally similar to the values used in the 1982 stormwater master plan. However, the current study reduced the impervious values of the undeveloped areas, from 7% to 2%, based on UDFCD recommendations and current practices. Modeling parameters such as subwatershed length, distance to centroid, and slope were calculated using a GIS interface. These parameters are summarized in Table A.2, located in Appendix A.

Several differences were noted in the hydrologic parameters between the 1982 stormwater master plan and current recommendations from the USDCM. Specifically, the 1982 stormwater master plan used infiltration rates significantly higher than recommended by current practice, and decay rates significantly lower than current practice. These parameters were revised to ensure consistency with current UDFCD recommendations. Additionally, the depression losses were decreased from 0.4” to 0.35”, per current recommendations.

3.4 Hydrograph Routing

Individual hydrographs developed for each subwatershed were routed throughout the basin using the EPA-SWMM program. Routing elements in SWMM consist of a system of channels, pipes, and storage elements. Conduit dimensions were determined through a GIS interface and current topographic mapping, in addition to field surveys. Roughness coefficients were selected in accordance with recommendations presented in the USDCM.

Five detention facilities were included in the SWMM modeling. These ponds include the Centennial Farms pond, Centennial Lake, Ehrlich Lake, the Settler’s Village south pond, and the Pheasant Hills sump. Storage volumes for all detention areas were taken from the LiDAR contour mapping. For the Centennial Farms and Settler’s Village ponds, and Centennial Lake, the outflow characteristics were taken from their individual drainage reports. All three of these ponds limit their 100-year release to a maximum of 6 cfs. For the Pheasant Hills sump and Ehrlich Lake, the outflow characteristics were calculated based on field surveys and outlet dimensions. The Pheasant Hills sump does not have a controlled outlet, but rather fills to an elevation of approximately 4793’ before spilling to the north across the Hillsboro Canal.

A total of two detention basins were included in the 1982 stormwater master plan. These include two small ponds on the Lola Park tributary. Based on current UDFCD eligibility criteria, neither pond was included for the current hydrology update.

3.5 Previous Studies and Results of Analysis

A summary of the hydrologic analysis can be found in Appendix A. The 1982 stormwater master plan is the only previous source of hydrologic results from the area, so comparisons are made at select design points between the current and previous studies. It was found that the current update results in higher peak discharges in all areas than the 1982 stormwater master plan. The 1982 stormwater master plan used SWMM to generate runoff hydrographs, whereas the current update uses CUHP, which can introduce differences in the peak runoff rates due to differences in the software. The current update also uses a lower infiltration rate than previously estimated, causing additional runoff to occur.

Changes were also made to the runoff routing for the current update. As mentioned above, the Lola Park Tributary is diverted near S. Irene Avenue and County Road 46. The majority of the 100-year discharge is routed east towards Alice Avenue. This diversion was not modeled in the 1982 stormwater master plan, causing discharges along Alice Avenue to increase significantly. Increased development in the Western and Lola Park Tributaries have also caused an increase in runoff rates. Table 3.2 includes a brief flow comparison, selected at locations that were directly comparable.

Table 3.3 also includes a summary of existing and future discharges across all return periods at selected design points in the watershed. Note that the Northern and Big Thompson tributaries are not included in this table, since they mostly consist of independent subwatersheds, rather than a specific drainageway.

4.1 Evaluation of Existing Facilities

Hydraulic analysis of existing infrastructure within the Town of Milliken was performed in order to determine the capacity and approximate return periods of conveyance systems and detention ponds. Several modeling programs were used to complete this analysis, most notably EPA-SWMM 5.0.22, FLO-2D Pro, and FlowMaster. The capacity of existing closed conduit systems was estimated from SWMM computations generated as part of the hydrologic analysis of this report. Inlet control nomographs were used with the project mapping to estimate culvert capacities where necessary. Detention ponds were modeling using SWMM as part of the hydrologic analysis.

4.2 Existing Drainage Facilities

Existing drainage facilities throughout the Town vary in size, shape and material. Capacity of the existing infrastructure also fluctuates and is generally undersized compared to the future conditions discharges. In the undeveloped portions of Town, there is a near-complete lack of infrastructure entirely. For example, ‘texas crossings’ at several gravel roads where no culverts have been installed at the low point of the creek or swale. Table 4.1 includes an inventory of the existing drainage infrastructure in the Town, as well as its approximate capacity and return period.

4.3 Flood Hazards

The majority of the flood hazards within the study area have resulted from the elimination of natural flow paths and/or insufficient conveyance capacity of the existing drainage facilities. As the Town has developed over time, the natural flow paths, such as along the Lola Park Tributary, have been encroached or blocked by railroad or roadway embankments. In most cases, cross-drainage is insufficient or non-existent, leading to residual backwater ponding, causing flooding throughout the Town. Inundation areas are mostly limited to residential areas, although commercial property along Broad Street is also at risk. The results of this report generally verify the flood hazards presented in the 1982 stormwater master plan. An inventory of the existing drainage facilities, and their approximate return period prior to overtopping (based on existing conditions hydrology) is shown in Table 4.1. These areas are also identified in Appendix B, and discussed on a tributary-by-tributary basis below.

4.4 Previous Analysis

This report incorporates the results of the 1982 stormwater master plan. Information from the previous study was used to identify flood hazards and evaluate the impacts of the updated future conditions hydrology to determine problem locations.

The 2013 Flood Recovery Floodplain and the effective FEMA Flood Insurance Study were also considered in this analysis. However, the hydraulic analyses on the Big Thompson and Little Thompson rivers within those State and Federal products are not re-evaluated in the local stormwater master plan. Those studies are referenced and used to define the boundary condition between local and regional flood hazards. This study benefited from the currency of the flood hazard determinations on the regional flood sources. Typically local or regional data is temporally superior to the other.

5.1 Alternative Development Process

Once the drainage problem areas were identified, alternatives were developed in order to mitigate flooding potential and implement the project details set forth by the Town of Milliken. These goals include:

• 100-year conveyance of stormwater, including at roadway crossings
• Increased detention to mitigate changes in hydrology
• Improved water quality
• Recreational & maintenance access

Several broad alternative categories were studied in order to minimize the drainage problem areas within the Town. Not all categories were considered for every reach, due to conditions specific to each tributary. Categories include:

• 100-year conveyance with detention
• 100-year conveyance without detention
• Floodplain preservation
• No improvements recommended

Generally, the 100-year conveyance with detention alternative favors the use of regional detention ponds to minimize required stormwater conveyance, whereas 100-year conveyance without detention alternative relies on larger storm sewer outfall systems to convey discharges. Water quality treatment is included in each detention pond. Property acquisition is generally included on an as-needed basis, either through easements or full acquisition in order to accommodate required stormwater infrastructure. Floodplain preservation was considered in areas where future development may encroach into existing floodplains or streams.

The design and evaluation of the conceptual alternatives was based on criteria set forth in the UDFCD Urban Storm Drainage Criteria Manual and the Flood Plain Information, Drainage Plan and Street Improvement Plan at Milliken, Colorado.

5.2 Alternative Hydrolugy & Hydraulics

As shown in Table 3.3, the future conditions hydrologic analysis resulted in large increases in the 100-year discharge throughout Town, due to additional development expected south of Town. In order to mitigate these changes in the hydrology, a baseline hydrologic model was created. Based on current development practices, it is expected that as these areas develop, they will be required to provide detention areas sufficient to reduce developed discharges to a level at or below the historic major and minor storm discharge. The baseline hydrologic model includes detention areas sized to reduce future discharges near the existing level on a watershed-by-watershed basis. These ponds are inclusive to each alternative studied, and may not represent the total detention volume required to completely offset the increase in runoff volume or downstream runoff rates expected with future development.

5.4 Alternative Plans

Alternative plans were developed to address problem areas depicted in Section 4. Alternative plan drawings are included in Appendix C and in the interactive web map. The following is a summary of the alternatives evaluated on a reach by reach basis, along with a discussion of their feasibility and project cost estimates.

6.2 Water Quality Impacts

Since there are currently few stormwater BMP’s in the Town of Milliken, the recommended plan will help increase water quality. As the upper portions of the watershed develop, the detention areas shown in Table 5.1 will treat runoff from new developments. The implementation of these detention areas will ensure that the majority of the Lola Park Tributary is treated for water quality, and much of the Western and Eastern tributaries will also be treated. The recommended plan generally favors open channel flows, for instance, the Western Tributary cutoff ditch, which also help increase water quality.

6.3 Operations and Maintenance

Maintenance and inspection is an essential part of all studied alternatives. In order for the proposed alternative to function properly, conveyance capacities must be maintained. The deposition of sediment in pipes, open channels, and detention areas reduces the effectiveness of the improvements and increases the probability of flood damages throughout the study area. Yearly and long-term maintenance costs have been included in each cost estimate.

Since there is currently relatively little drainage infrastructure in the Town of Milliken, it is likely that the additional regional improvements recommended herein will increase required maintenance within the Town. However, these facilities will reduce flooding and flood-related damages in the Town, reducing maintenance and clean-up activities associated with flooding.

It is recommended that future channel alignments include maintenance access trails incorporated into the cross-section. These trails greatly increase maintenance access while providing multi-use benefits for residents of the Town. Equipment, manpower, and permit authority are other considerations of the operation and maintenance of the recommended improvements. Existing equipment and manpower should be evaluated as recommended improvements are designed. For instance, is the maintenance access width compatible with the Town’s existing equipment? Permit authority is an administrative process to ensure Town work is compliant with building, environmental, and floodplain ordinances. The existing Town ordinances provide for emergency operation and maintenance, but should be evaluated for allowance of routine maintenance.

6.4 Environmental and Safety Assessment

Implementation of the selected plan will reduce peak flow rates, reduce flood hazards and flooded areas, and increase emergency access through reduced street flooding. The construction of open channels will also improve environmental and recreational conditions throughout the watershed, thereby enhancing neighborhoods.

1. Simons, Li & Associates, Inc, Flood Plain Information, Drainage Plan and Street Improvement Plan at Milliken, Colorado, October 1982
2. Icon Engineering, Inc., Hydrologic Analysis of the Box Elder Creek/Cooper Slough Watershed, April 2014
3. RMC, July 2000, Final Drainage Report for Centennial Farms Subdivision
4. Urban Drainage & Flood Control District, Urban Storm Drainage Criteria Manual, April 2008
5. National Oceanic and Atmospheric Administration, NOAA Atlas 2 Precipitation-Frequency Atlas of the Western United States, Volume III – Colorado, 1973
6. Urban Drainage and Flood Control District, UDFCD Position on the NOAA Atlas 14 Precipitation-Frequency Atlas, Volume 8, September 2013
7. Urban Drainage and Flood Control District, Colorado Urban Hydrograph Procedure 2005, computer program, version 1.4.3
8. RMC, July 2000, Final Drainage Report for the Colony Pointe Subdivision at Milliken, Colorado

• Appendix A - Hydrologic Analysis

• Appendix B - Hydraulic Analysis

• Appendix C - Alternatives Analysis

• Appendix D - Recommended Plan

• Appendix E - Relevant Excerpts

Download: Full Report (pdf)