Sparks Prioritizes Its Stormwater Improvements

In today’s difficult economic climate, local governments must ensure that every dollar they spend is allocated wisely. For the city of Sparks, NV, this maxim certainly applies as it seeks to improve its stormwater infrastructure. Therefore, the city is working with the engineering consulting firm PBS&J, an Atkins Company, to conduct detailed stormwater master planning analysis on several areas within Sparks. As part of a three-phase process that is nearly complete, PBS&J is developing a comprehensive list of recommended capital improvement projects (CIPs) that the city will use to determine how best to proceed as it upgrades its existing stormwater infrastructure.

Home to approximately 80,000 residents, Sparks is located immediately to the east of its larger neighbor, Reno, NV. Although localized flooding occurs throughout Sparks, most of the problem areas are located in the city’s older, urban core. Not only was this area constructed before much thought was given to stormwater management, but runoff from other areas in the city must drain through it to reach the Truckee River.

Over the years, the city has conducted several different stormwater master plans for various areas throughout Sparks. Because they were carried out at different times and with different methodologies, the various plans lacked consistency and failed to provide a unified, cohesive approach for tackling stormwater problems across Sparks. To rectify this situation, the city decided to conduct a new comprehensive master plan that would result in a uniform assessment of stormwater needs throughout Sparks, except for one area that had been constructed fairly recently using the most current stormwater management techniques. Ultimately, the city wanted to be certain that it had the best possible information on hand as it determined which improvements to make to its existing stormwater infrastructure. Armed with an up-to-date assessment of its stormwater needs, the city would be in a better position to ensure that the CIPs funded by its existing utility for sewer and storm drains would have the greatest effects in terms of protecting public safety, reducing flooding, and maximizing resources.

Conducting the Master Plan in Phases
In January 2009, the city hired PBS&J to analyze the stormwater needs for several areas within Sparks encompassing more than 8,900 acres. The purpose of the planning project was to complete a rain gage network analysis, inventory and evaluate the capacity of the existing stormwater conveyance systems, identify conveyance deficiencies in the current stormwater infrastructure, develop improvement alternatives to resolve the deficiencies, and prepare the comprehensive master plan with CIPs for improving identified stormwater conveyance deficiencies. The ultimate goal was to develop and rank the CIPs from the most to the least critical and beneficial, providing a critical planning tool for the city as it prioritizes and implements its stormwater drainage system CIP budget.

The project is being completed in three phases, the first two of which have been concluded. Begun in February 2009, phase I included the rainfall and gage network analyses, data collection for areas 1 through 7, hydrologic and hydraulic analysis for areas 1 through 3, and development of improvement alternatives for areas 1 through 3 (see Sparks Areas map). Phase II, which began in January 2010 and lasted through the fall, involved hydrologic and hydraulic analysis and development of improvement alternatives for areas 4 through 8. Phase III, which began in December 2010, will complete the master planning process for the core of the city of Sparks.

The underlying objective in preparing the master plan is to develop stormwater improvement alternatives that intercept as much of the runoff from a five-year storm as possible in the storm drain, while limiting the hydraulic grade line below an approximated top-of-curb elevation of 0.5 foot.

The city of Sparks provided the project team with 2-foot contour data and aerial photographs. Dating to 2004, the contour data encompassed the entire study area. Generally, the project team relied on the 2004 topography, as it appeared to be more detailed than the aerial photographs and because most areas had not experienced any significant regrading since then. Before the project began, GIS data were obtained from the geospatial technologies staff at the city of Sparks in the form of a disconnected editing session. Collected field data and survey data were added to the editing session and periodically reviewed, approved, and incorporated by the city’s geospatial technologies staff.

Evaluating Existing Conveyance Capacity
To perform hydrologic and hydraulic analyses, the project team selected XP Software’s XPSWMM (Version 2009), mainly because of its ability to model unsteady flow, link hydrologic and hydraulic analysis, interface with GIS databases, and model two-dimensional overland flow. Using the GIS data from the city, PBS&J extracted point and line features—representing manholes, drop inlets, and storm drains—from the geodatabase as appropriate to create GIS shapefiles representing the main lines of the storm drain system. The point-and-line shapefiles were then imported into the XPSWMM model as nodes and links, and appropriate GIS attributes were mapped and populated in XPSWMM using the available database connection tools.

After the storm drain network had been established in XPSWMM, another GIS shapefile was imported to represent the tributary catchments, or sub-basins, draining to the storm drain network. Runoff variables associated with the catchments, including area, time of concentration, and curve numbers, were mapped and populated during the import process. Catchments were then connected to appropriate nodes in the storm drain network. The model was set up to solve both hydrologic runoff and hydraulic conveyance simultaneously. Some of the main considerations related to the setup of the XPSWMM model addressed the use of multi-conduit links, modeling of overland routing links, and modeling of storm drain lines.

When it comes to the use of multi-conduit links, the intent of the model is to analyze surface flow and subsurface flow in the storm drain system. This was done using the multi-conduits feature in the XPSWMM model to analyze the interaction between underground storm drain conduits and overland sections, such as roadways and channels. Multi-conduits are a special type of link in XPSWMM that essentially allow multiple conduits between two nodes. For example, street flow can be modeled in one conduit and pipe flow in another, thereby creating a dual drainage system capable of modeling the interaction between the two.

Overlying roadway sections were modeled as natural sections that were assumed to be crowned with a 2% cross-slope and bounded by curbs and gutters with a curb height of 0.5 foot. At both edges of the street section right of way, vertical walls were extended up to ensure that the cross-sections were large enough to convey the peak flow. This methodology was used so that no flow would be lost from the system and so that the water that surcharges on the street from a pipe could be routed downstream on the street surface. The extended vertical walls also assume that the effective flow is largely limited to the right of way.

Most multi-conduit links consist of an underground conduit and the overlying roadway section as described above. However, in some instances where the surface above the underground conduit was not a roadway, an appropriate overlying geometric section was assumed. In areas where an obstruction, such as a roadway embankment or other impediment, was present on the surface above the underground conduit, a single conduit was modeled to represent the storm drain. This approach was also used for areas not requiring an overland conduit.

As for modeling overland routing links, runoff conveyance initially was modeled in a dual drainage system, where the overland section followed the path and direction of the underground storm drain conduit. However, in some instances this approach did not adequately represent surface drainage patterns, resulting in unreasonable flow depths. At these locations, overland routing links were modeled to convey flow from one area of the storm drain to another, typically via a standard roadway section. These select links were added to increase the detail of the master plan modeling efforts and better account for surface flow paths. However, these links do not represent all overland flow that will occur during a given event. To sufficiently model that interaction, a two-dimensional model would be required.

For modeling of storm drain lines, major storm drain lines were modeled to the terminal manhole and do not include the modeling of upstream drop inlets. Sub-basins were delineated to all major storm drain lines. In some instances, however, delineating sub-basins to the upstream end of a storm drain would have resulted in a very small basin, which could artificially increase flow within the system. In these locations, the sub-basin was not delineated, and a small portion of the storm drain was not modeled.

Generally, the modeling of runoff conveyed both in the storm drain and a representative roadway section provides a reasonable estimate of gutter-flow depth and time to peak during five-year storm events. However, the amount of flow intercepted by a storm drain at any one sub-basin location is highly dynamic and subject to variability. Some factors that affect storm drain capacity and/or conveyance time include storm drain maintenance conditions, roadway gutter conditions, split flow at intersections, specific roadway section widths, and topography outside the right of way. Because these additional factors are subject to a high degree of variability, simplifying assumptions were included in the hydrologic and hydraulic models. For the purpose of this master plan, storm drains and gutter sections were assumed to be free of sediment, locally representative roadway section widths were assumed, and flow conveyance was assumed to be limited to the right of way. Sufficient inlet capacity was also assumed, and no detailed drop inlet calculations were performed in this study.

Four separate hydrologic and hydraulic modeling scenarios were completed for each existing condition analysis. The scenarios, which were identical except for the use of different rainfall data and tailwater conditions, entailed analyses of a five-year storm, an actual storm that occurred on January 4, 2008, a two-year storm, and a 100-year storm. The five-year existing condition analysis was used to determine existing storm drain capacity and to develop improvement alternatives. The analysis of the January 4, 2008, storm incorporated NEXRAD rainfall data and was used to determine if the hydrologic parameters used for the five-year existing condition analysis were reasonable. The two-year analysis was developed to determine peak flows for future water-quality/environmental considerations. The 100-year analysis was primarily developed to determine overall general drainage patterns and runoff volumes from the watershed.

The local watershed hydrology was developed based on procedures outlined in the draft Truckee Meadows Regional Drainage Manual (TMRDM). Sub-basin areas were delineated in GIS using the 2-foot contour data, storm drain and inlet location, aerial photographs, and street alignments to determine the local drainage areas for each major lateral. Sub-basins were developed in GIS as topological elements to ensure that no gaps or overlaps appeared in the sub-basin boundaries. Nearly all sub-basins are less than 1 square mile and less than 10% slope. The SCS Curve Number Method was used to compute runoff.

Double-Checking Land Uses
Land-use data used in the hydrologic analysis is a crucial component of the overall plan. Land-use densities and the associated impervious areas directly affect the amount of runoff that will occur for a given area. Impervious area increases as land use densities increase, which, in turn, increases both runoff volume and rate. Because of the importance of categorizing land use accurately, PBS&J cross-checked two sources of land-use data—aerial photographs and satellite images—to confirm categorical assumptions regarding land use that had been made based on the aerial photos and the city’s existing land-use/zoning maps.

Land-use boundaries and categories were developed in a GIS polygon shapefile using the supplied 2008 aerial photos, knowledge of current development, and existing GIS data (land use and zoning layers) provided by the city. Typical impervious percentages and curve numbers were taken from the TMRDM and expanded upon where needed to adequately define the varying land uses within the master plan area. These land uses and associated impervious percentages were then compared in GIS to estimates of percent imperviousness of the land cover (for example, rooftops, roads, and parking lots) from the impervious surface data of the 2001 National Land Cover Database to evaluate the appropriateness of the assigned land use.

When the satellite images, aerial photos, and the assigned land-use category consistently identified similar levels of imperviousness for a given area, PBS&J assumed that it had the correct percentage of imperviousness. However, when the aerial photos and/or the satellite data revealed discrepancies in imperviousness percentage when compared to the assigned land-use category, PBS&J would double check those locations to verify actual level of imperviousness or modify the land-use category or boundaries as appropriate. Using several data sources ensured better estimates of hydrologic parameters used in the modeling effort.

Analyzing Rainfall Data and the City’s Rain Gage Network
Two different sources of rainfall data were used in the development of the master plan: NEXRAD gridded precipitation data for the storm event that occurred on January 4, 2008, and the rainfall tables from the TMRDM for regions 1 through 3 in the city of Sparks. The NEXRAD precipitation data were input into one of the XPSWMM simulations and used primarily for calibration purposes to ensure that the runoff model was providing reasonable results, as well as to fine-tune hydrologic parameters. The NEXRAD precipitation data were imported into GIS to determine 24-hour time-series distributions for each 1-kilometer-square grid, based on five-minute rainfall increments within that 24-hour period (January 4).

Before the development of the master plan, the rain gage network relied upon by the city of Sparks was rather sparse. As part of the process of developing the master plan for the city, PBS&J was asked to analyze the city’s existing network of rain gages and recommend how best to expand this network. The intent of the analysis was to improve the gage network for better resolution to capture localized thunderstorm events. To this end, PBS&J recommended adding rain gages at six new sites and adding both rain and water level gages at two new sites to measure rain and detention pond levels. Based on this analysis, the city installed the eight new rain gages and two new detention level gages during phase II of the master plan, significantly expanding the rain gage network to 10 sites that are used by the city and Washoe County. Adding detention basin sensors enables the city to inexpensively monitor its facilities remotely and reallocate resources to other critical activities related to storm response, as needed.

Identifying Problem Areas and Developing CIPs
To develop design flows affecting the stormwater system, the project team used the five-year-frequency storm event along with data regarding land use and soil type. Areas of likely flooding were identified on the basis of whether the combined conveyance capacity of the subsurface storm drain and the overland roadway section can convey the five-year design flows while still maintaining a computed hydraulic grade line or water surface elevation less than a standard gutter section depth of approximately 0.5 foot. Storm drains that resulted in flow depths within the roadway of greater than 0.5 foot were categorized as potential problem areas, and CIPs were developed.

Multiple alternatives have been investigated for each area. In some cases, a preferred alternative was developed for each area of this master plan and developed into a CIP. In other cases, multiple alternatives were developed into CIPs. The decisions of which alternatives to develop into CIPs were made primarily based on the results of each alternative and coordination with city staff. Probable cost estimates for each CIP were developed based on 2009–2010 dollars and are at a master plan level.

CIPs have been summarized into the following three general categories:

• Priority CIPs. Classified as the highest priority, these CIPs generally increase stormwater conveyance capacity significantly and reduce potential flooding areas. The master plan recommends that these CIPs be constructed as soon as possible.
• Opportunity CIPs. Classified as a mid-level priority, opportunity CIPs benefit stormwater conveyance, but their associated improvements are typically limited to smaller localized areas. The master plan recommends that these CIPs be constructed as opportunities arise in relation to coinciding street improvements, other construction projects, and the like.
• Convenience CIPs. Classified as lower priority, these CIPs generally improve factors pertaining to convenience, such as accessibility, maintenance, and potential for land development. Although convenience CIPs generally provide minimal increase in stormwater conveyance capacity, the master plan recommends that the city construct them at its discretion.

For each of the eight areas assessed by phases I and II of the master plan, PBS&J identified potential problem flooding areas based on the five-year existing condition analysis and summarized those locations where flows are expected to exceed gutter capacity. To date, PBS&J has recommended 45 CIPs. Of these, 17 are classified as priority CIPs, 21 are opportunity CIPs, and the remaining seven are convenience CIPs. With estimated costs ranging from $47,000 to $8.3 million, the recommended projects have an estimated total cost of approximately $70 million. Examples of recommended CIPs include expansion of existing storm drains, installation of new storm drains, construction of detention basins, rerouting of drop inlets, and installation of trash racks.

The intent of these recommendations is to assist city staff in developing their annual overall CIP list. Final determination of CIP priority will be made by city staff. Because the CIP recommendations in the master plan primarily encompass factors related to hydrology, hydraulics, and flooding, the recommendations may not reflect all considerations that the city may need to determine and schedule CIPs. Therefore, the systems identified and described in the master plan will be subject to further amendments and revisions during final design. At that time, detailed analyses and information will be required to, at a minimum, verify all utility locations and develop hydrological data for analyses of specific drop inlets. In addition, field verifications of inlet locations and storm-drain connectivity will be necessary, as might more accurate topography data. Storm drain locations and slopes also will need to be finalized.

Conclusion
Scheduled for completion this summer, the ongoing phase III of the master plan entails an evaluation of three other areas of Sparks not studied as part of the previous phases. Following the conclusion of the master plan, city staff will use the results to begin prioritizing the recommended solutions and developing a long-term, coherent approach to CIPs aimed at reducing flooding throughout Sparks as cost-effectively as possible.