Riverbank_Stability_Assessment_City_of_Pittsburgh_-Draft.pdf

i

Riverbank Stability Assessment and Applied Best Practices Guide

Pittsburgh, PA

Prepared for: City of Pittsburgh 414 Grant St.

Pittsburgh, PA 15219

Prepared by: Planning, Programs, & Project Management Division U.S. Army Corps of Engineers, Pittsburgh District 1000 Liberty Avenue Pittsburgh, Pennsylvania 15222

April 2021 ii

Executive Summary

Introduction There is growing consensus over the importance of considering social, political, and environmental factors together when making decisions regarding the rehabilitation and stabilization of riverbanks within urban settings. Toward this end, the City of Pittsburgh requested assistance from the U.S.

Army Corps of Engineers (USACE), Pittsburgh District with developing a series of tools to facilitate effective management of approximately 43 miles of riverbank along the Allegheny, Monongahela, and Ohio Rivers. The specific objectives of this study were to:

1. Assess and map the current condition of riverbanks within City limits

2. Develop an applied best practices document that guides future remediation of unstable riverbanks within the context of the planning goals established by the City and its partners.

The current study was conducted under Section 22 of the Water Resources Development Act of 1974, as amended.

Riverbank Condition Assessment & Mapping The study team characterized and evaluated critical attributes controlling riverbank erosion and failure throughout the study area, including bank material and condition, slope, the presence and type of vegetation, extent of impervious surfaces within the riparian area, and the presence and/or extent of existing features that can further contribute to bank instability (e.g., existing erosion and/or undercutting). Bank material, bank condition, and slope were used to establish riverbank segments and define 31 distinct riverbank typologies. Each riverbank segment was given an overall condition score of either poor, fair, or favorable, depending on its underlying typology. The presence and extent of erosion and undercutting and extent of adjacent impervious surface was then used to further refine each riverbank segment’s overall condition and stability rating. A total of 11.9 miles of riverbank received a ‘favorable’ condition assessment rating, 24.1 miles received a ‘fair’ rating, and 6.7 miles received a ‘poor’ rating (Figure ES1).

Assessment data for each reach, along with photographs depicting the characteristics of each segment were published to ArcGIS Online and an application was created to display the characteristics. The Web Map Application can be used to visually assess riverbank conditions throughout the study area and facilitate planning and management decisions, such as prioritizing areas for future restoration actions. This interface is live and accessible at the following location:

https://lrp.maps.arcgis.com/apps/webappviewer/index.html?id=87208aa3ff9f4cd7a2fdeaf3841d6ddc

Riverbank Stabilization Techniques and Applied Best Practices Results of the riverbank assessment were incorporated into a best practices guide designed to assist with identifying the most effective riverbank stabilization and rehabilitation techniques for each riverbank typology. Riverbank stabilization approaches included within this report fall within the broad categories of bioengineering (i.e., engineering techniques that utilize nature-based materials and processes to encourage riverbank stabilization and ecological diversity), erosion and sediment control, stormwater management, and geotechnical techniques.

iii

Figure ES1. Map of final riverbank condition assessment scores for the study area.

A brief, non-topology specific overview of commonly applied bioengineering, erosion and sediment control, and stormwater management are first provided, largely pulling from existing reports. Although bioengineering, erosion and sediment control, and stormwater management techniques may be effective at addressing bank instability in certain locations throughout the study area, they will likely be most effective when implemented in combination with various geotechnical approaches to provide a more holistic and long-term solution. Thus, topology-specific recommendations are provided for geotechnical approaches, with additional discussion of bioengineering, erosion and sediment control, and stormwater management approaches that may be paired with each geotechnical approach. Recommendations were also made within the context of planning priorities as identified by the City of Pittsburgh and the Riverbank Stability Working Group. In all instances, a licensed engineer should be consulted for design and implementation.

The report also details likely permitting and coordination requirements for stabilization work.

Conclusions The condition assessment and associated Web Map Application, along with the best practices guide and associated recommendations can be leveraged by the City of Pittsburgh and its partnering stakeholders to identify and prioritize portions of riverbank for stabilization and other management and development actions that maximally meet the identified planning objectives.

iv

Table of Contents

Executive Summary ...................................................................................................................... ii

Abbreviations and Acronyms ...................................................................................................... v List of Tables ................................................................................................................................ vi List of Figures .............................................................................................................................. vii List of Appendices ........................................................................................................................ ix

1 Introduction

1.1 Background & Objectives

1.2 Coordination

1.3 Study Area

1.4 Brief History of Pittsburgh Riverbanks

2 Riverbank Condition Assessment & Mapping

2.1 Methods

2.1.1 Criteria Development

2.1.2 Field Assessment

2.1.3 GIS Assessment

2.1.4 Condition Assignment

2.2 Results and Existing Conditions

2.2.1 Field Assessment

2.2.2 GIS Assessment

2.2.3 Condition Assignment

2.3 Published Map: Web Map Application

2.3.1 Using the Web Map Application

3 Riverbank Stabilization Techniques and Applied Best Practices

3.1 Bioengineering Techniques for Riverbank Stabilization

3.2 Erosion and Sediment Control

3.2.1 Vegetation for Erosion and Sediment Control and Riverbank Stabilization

3.3 Stormwater Management

3.4 Geotechnical Techniques for Riverbank Stabilization

3.4.1 Stabilization Techniques for Naturalized Riverbank Typologies

3.4.2 Stabilization Techniques for Manmade Riverbank Typologies

3.5 Riverbank Stabilization and Multi-Priority Planning

4 Permitting Process and Next Steps

4.1 Federal Permitting

4.2 State of Pennsylvania Permitting

4.3 Allegheny County and City of Pittsburgh Permitting

5 Summary

6 Glossary of Terms

7 Literature Cited v

Abbreviations and Acronyms ACCD ............................................................................. Allegheny County Conservation District ACPW ........................................................................... Allegheny County office of Public Works AGC ........................................................ U.S. Army Corps of Engineers Army Geospatial Center ALCOSAN ........................................................................... Allegheny County Sanitary Authority BMP ...................................................................................................... Best Management Practice CoP DCP ............................................................................. City of Pittsburgh Department of City CWA ..................................................................................................................... Clean Water Act ESRI ............................................................................. Environmental Systems Research Institute E&S ............................................................................................................... Erosion and Sediment GIS .............................................................................................. Geographic Information Systems GNSS ...................................................................................... Global Navigation Satellite System IENC ...................................................................................... Inland Electronic Navigation Charts IPaC.............................................................................. Information for Planning and Consultation NPDES .................................................................... National Point Discharge Elimination System MWMO ............................................................ Mississippi Watershed Management Organization NRCS ................................................................................ Natural Resource Conservation Service PADCNR .................................. Pennsylvania Department of Conservation of Natural Resources PADEP ..................................................... Pennsylvania Department of Environmental Protection PAFBC ........................................................................... Pennsylvania Fish and Boat Commission PAHMC .......................................................... Pennsylvania Historical and Museum Commission PAPUC ............................................................................ Pennsylvania Public Utility Commission PASPGP-4................................................... Pennsylvania State Programmatic General Permit # 4 PennDOT ................................................................... Pennsylvania Department of Transportation PHMC ............................................................. Pennsylvania Historical and Museum Commission PNDI ............................................................................. Pennsylvania Natural Diversity Inventory PNHP ............................................................................... Pennsylvania Natural Heritage Program SHPO ......................................................................................... State Historic Preservation Office TRM ........................................................................................................ Turf Reinforcement Mats USACE ........................................................................................... U.S. Army Corps of Engineers USEPA .............................................................................. U.S. Environmental Protection Agency USFWS ............................................................................................ U.S Fish and Wildlife Service vi

List of Tables

Table 1. Finalized list of characteristics included in the riverbank assessment. Attributes are categorized based on assessment methods (i.e., those assessed in the field and those assessed using existing data within GIS), as well as the type of feature each attribute represents (i.e., continuous or point)

Table 2. Definitions and key characteristics for the three natural and six manmade riverbank types

Table 3. Definitions and key characteristics of bank condition for natural and manmade riverbanks

Table 4. Slope chart showing the relationship between the horizontal (H) and vertical (V) components of slope as represented by ratio, percent grade, and slope angle

Table 5. Definitions of slope assessment criteria for natural and manmade riverbanks Table 6. Notation used to define riverbank typologies based on continuously assessed field attributes. Notations are broken out for natural and manmade riverbanks Table 7. The length and overall percentage of the 31 riverbank typologies encountered throughout the study area. Typologies are defined by the three ongoing attributes collected in the field (i.e., bank material, material condition, and slope. Typology notations reflect those presented in Figures 8 through 12

Table 8. Summary of bioengineering riverbank stabilization techniques Table 9. Selected Erosion and Sediment Control Best Management Practices suitable for riverbank stabilization techniques Table 10. Invasive species prolific along riverbanks in Pittsburgh, PA Table 11. General Practices of Invasive Plant Management Table 12. Selected recommended seed mixes based on site condition, modified from PADEP

Erosion and Sediment Control Manual Table 13. Recommendations based on river zone Table 14. Recommended species based on riverbank flooding frequency and elevation for

Pittsburgh region Table 15. Stormwater Management Techniques Table 16. Typologies applicable for flattening the slope Table 17. Typologies applicable for terracing the slope Table 18. Typologies applicable for excavation and replacement Table 19. Typologies applicable for buttressing a slope Table 20. Typologies applicable for installing a retaining wall Table 21. Manmade typologies applicable for flattening a slope Table 22. Manmade typologies applicable for terracing a slope Table 23. Manmade typologies applicable for excavating and replacing Table 24. Manmade typologies applicable for installing a retaining wall Table 25. Manmade typologies applicable when repairing an existing structure Table 26. Manmade typologies applicable when replacing an existing structure Table 27. Example recommendations for stabilization actions that would help meet each of the four planning priorities Table 28. List of possible agency coordination vii

List of Figures

Figure 1. Extent of assessed riverbanks within the City of Pittsburgh and adjacent municipalities. Location of the study area within Allegheny County is also shown Figure 2. Photographs showing a variety of riverbank land use and a barge carrying coal Figure 3. Photographs showing a naturalized riverbank and manmade concrete wall Figure 4. Photographs showing a deteriorating manmade slag and naturalized vegetated slope ...7 Figure 5. Horizontal (H) and vertical (V) relationships represented as ratios, percent grade, and slope angle Figure 6. Photographs showing of examples of undercutting and erosion point features Figure 7. Photographs showing examples of culvert point location and impervious surface Figure 8. Decision tree used to define and assign bank condition scores to natural riverbank typologies. Missing typologies and associated condition scores indicate typologies that were not encountered within the study area

Figure 9. Decision tree used to define and assign bank condition scores to manmade slag riverbank typologies. Missing typologies and associated condition scores indicate typologies that were not encountered within the study area

Figure 10. Decision tree used to define and assign bank condition scores to manmade slag riverbank typologies. Missing typologies and associated condition scores indicate typologies that were not encountered within the study area

Figure 11. Decision tree used to define and assign bank condition scores to manmade riprap typologies. Missing typologies and associated condition scores indicate typologies that were not encountered within the study area

Figure 12. Decision tree used to define and assign bank condition scores to manmade indistinguishable and manmade other typologies. Missing typologies and associated condition scores indicate typologies that were not encountered within the study area

Figure 13. Map showing the location and extent of natural and manmade riverbanks throughout the study area. Locations and extents of specific natural (soil and fill) and manmade (indistinguishable, other, riprap, slag, concrete walls, and steel walls) can be found in the online web application

Figure 14. Length of riverbank within the study area that classifies as natural and manmade materials

Figure 15. Map showing the location and extent of natural and manmade riverbank conditions throughout the study area

Figure 16. Length of natural and manmade riverbanks within the study area that categorized by condition

Figure 17. Map showing the location and extent of riverbanks with shallow, steep, and vertical slopes throughout the study area

Figure 18. Length of natural and manmade riverbanks within the study characterized by shallow (i.e., less than 2H:1V), steep (i.e., greater than 2H:1V), and vertical slopes

Figure 19. Location of distinct areas of erosion and undercutting throughout the study area Figure 20. Location of manmade structures identified during the field assessment and other culvert and outlet locations provided by ALCOSAN Figure 21. Map showing the location and extent of riverbanks with shallow, steep, and vertical slopes throughout the study area Figure 22. Map of final riverbank condition assessment scores for the study area viii

Figure 23. Three riverbank zones described in riverbank stabilization techniques Figure 24. Diagram and photo of installed brush mattress Figure 25. Diagram and photo of bush layering Figure 26. Diagram and photo of a turf reinforcement mat Figure 27. Diagram and photo of live staking in combination with tree revetments and erosion control blanket Figure 28. Photos of root wad stabilization and natural boulder stabilization Figure 29. Diagram and photo of vegetated geogrid Figure 30. Diagram and photo of live fascines, bio-logs, and seeded erosion control blanket Figure 31. Diagram and photo of installed bio-log Figure 32. Diagram and photo of installed vegetated erosion control blanket Figure 33. Five main sources of water-generated erosion acting on riverbanks Figure 34. Cross-sectional and plan views of filter sock application Figure 35. Cross-sectional and plan views of filter berm application Figure 36. Diagram and photo of vegetative filter strip application Figure 37. Diagram and photos of silt fence and super silt fence application Figure 38. Diagrams of stair step grading and grooved slope surface roughening techniques Figure 39. Diagram and photo of erosion control blanket application Figure 40. Diagram and photo of cellular confinement system application Figure 41. Diagram and photo of green roof application Figure 42. Diagram of an open canal and sedimentation basin Figure 43. Diagram of a bioretention swale and vegetated swale Figure 44. Diagram of an underground substrate filter and water storage cistern Figure 45. Typical cross section view and photograph showing the flattening of an existing unstable slope with erosion control fabric placed on the flattened slope Figure 46. Profile view of flattened slop with bioengineering stabilization techniques Figure 47. Typical cross section view and photograph of a terraced slope installed to increase stability and provide access along walking path Figure 48. Example of a terraced slope design with integration of bioengineering techniques for section of riverbank along the Ohio River in Pittsburgh, PA Figure 49. Typical cross section of excavation and replacement of an existing unstable slope ...48 Figure 50. Riprap rock armoring with integration of vegetation and geotextile fabric Figure 51. Profile and plan view diagrams of rock barb installation as a rock armoring/excavation and replacement technique Figure 52. Typical cross section of a buttress constructed to stabilize a slope Figure 53. Typical cross section view showing a constructed wall to stabilize a slope and photos of constructed retaining walls along the riverfronts of Pittsburgh Figure 54. Retaining wall system including riprap armoring at base, geotextile installation behind the wall and vegetated erosion control blanket installation on the top of the wall Figure 55. Diagram of crib wall retaining structure with vegetation integrated into design Figure 56. Gabion wall installation that utilize bioengineering techniques ix

List of Appendices

Appendix A: NRCS Soil Survey of Riparian Corridor within Project Area

1 Introduction

1.1 Background & Objectives

Bank erosion and failure represent natural processes that are integral to the functioning of river ecosystems (Florsheim et al. 2008). However, land use change within the watershed (e.g., increased urbanization and floodplain development) and modifications to the natural channel (e.g., loss of natural vegetation and construction of engineered riverbanks) and flow regime (e.g., upstream reservoirs and locks and dams) can alter the natural rate of riverbank erosion/failure and result in socioeconomic and ecological consequences (Geyer et al. 2000, Allan 2004, Maynord et

al. 2008, Duro et al. 2020). This is particularly true for urban population centers, which are often characterized by extensive development around and significant modification to river channels in order to take advantage of the vital services they provide, including water provision, wastewater assimilation, recreation, navigation, and transport (Grimm et al. 2008).

There is growing consensus over the importance of considering social, political, and environmental factors together when making decisions regarding the rehabilitation and stabilization of riverbanks within urban settings (Findlay and Taylor 2006; Gurnell et al. 2007). Sustainability of the coupled socioeconomic and ecological system requires placing the river at the center of urban planning such that decisions are made to maximize effective use of the river while recognizing that a healthy, functioning river is vital to maintaining its utility. There have been recent calls for the development of tools capable of facilitating the identification, prioritization, and implementation of sustainable development activities and rehabilitation of riverbanks and their associated riparian corridors in urban areas (Johnson et al. 2007; Everard and Moggridge 2012).

Toward this end, the City of Pittsburgh requested assistance from the U.S. Army Corps of Engineers (USACE), Pittsburgh District (District) with developing a series of tools to facilitate effective management of riverbanks and associated riparian areas along the Allegheny, Monongahela, and Ohio Rivers. The specific objectives of this study were to:

1. Assess and map the current condition of riverbanks within City limits, and

2. Develop an applied best practices document that guides future remediation of unstable riverbanks within the context of the planning goals established by the City and its partners.

1.2 Coordination

The current study was conducted under Section 22 of the Water Resources Development Act of 1974, as amended, which provides authority for USACE to provide planning assistance and technical expertise to support states, local governments, and other non-Federal entities in broad, comprehensive water resource planning efforts. All work described in subsequent sections was cost shared on a 50% federal and 50% non-federal basis per Section 22 requirements.

1.3 Study Area

This study incorporates 42.8 miles of riverbank along the Allegheny, Monongahela, and Ohio Rivers, including: all riverbanks within the City of Pittsburgh (35.2 miles); all riverbanks within the Boroughs of Aspinwall (0.4 miles), Baldwin (1.5 miles), Etna (0.8 miles), Millvale (0.8 miles), Sharpsburg (1.6 miles), and Swissvale (0.6 miles); all riverbanks in Shaler Township (1.2 miles);

and a portion of riverbanks within O'Hara Township (0.7 miles) (Figure 1).

Figure 1. Extent of assessed riverbanks along the Allegheny, Monongahela, and Ohio rivers within the City of Pittsburgh and adjacent municipalities. Location of the study area within Allegheny County is also shown. The location of Herrs Island on Allegheny River is also shown with black and red radial.

1.4 Brief History of Pittsburgh Riverbanks

As can be seen in the Figure 1, the metropolitan area of Pittsburgh is centered around the confluence of the Allegheny and Monongahela rivers where they form the Ohio river. A 24-mile long pool, referred to as the Emsworth Pool or Pittsburgh Pool, spans the sections of the three rivers within the city limits with a targeted pool elevation of 710-feet maintained by the USACE Emsworth Locks and Dams (incudes Emsworth Gated Dam at River Mile 6.2 and a backchannel dam at Neville Island at River Mile 6.8) (USACE 2021). The Pittsburgh District operates 23 locks and dams on the Allegheny, Monongahela, and Ohio Rivers. The main purpose of the locks and dams is to improve and maintain the navigability of the rivers. Navigation and engineering of the three rivers surrounding Pittsburgh has served a prominent role in the industrial and economic history of the city and region. Engineering modifications including removing large rocks and snags, constructing low dams to deepen channels, and widening river extents to improve navigation on the three rivers surrounding Pittsburgh can be traced as far back as the late 1700’s.

By 1824, USACE assumed authority over the development of the rivers which resulted in the eventual large scale engineering improvements and canalization efforts that transformed navigation for the region and set the city of Pittsburgh up to serve as a commercial and industrial hub throughout the industrial revolution (Moxley, 2001). The Pittsburgh steel industry as well as coal and gravel mining, natural gas production, timbering, manufacturing, and the expansive railway system relied heavily on the rivers and riverbanks. Industrialization during the 1800’s and 1900’s brought substantial modification to the riverbanks of Pittsburgh including, but not limited to, deforestation, vast removal of natural soils, pollution to the soil and water, and widespread modification and installation of manmade materials along the riverbanks (Andropogon, 2006).

Figure 2. Photographs showing a variety of riverbank land use (left) and a barge carrying coal (right) along the Monongahela River in Pittsburgh, PA

2 Riverbank Condition Assessment & Mapping

From a geomorphic perspective, a riverbank is defined as the landform distinguished by the topographic gradient extending from the bed of the river channel along the lateral land-water margin toward the edge of the floodplain (Florsheim et al. 2008). The riverbank assessment in this study only considers the portion of the riverbank above the water surface and within the 75-foot riparian buffer at the time the survey was conducted. River elevations at Point State Park ranged from 710.5 feet to 710.7 feet during the time of sampling.

Riverbank erosion and failure is controlled by multiple interacting factors, including bank material and condition, river morphology (i.e., shape of the river channel), and vegetation and land use within the riparian zone. The riparian zone is defined as the land area that serves as the interface between the terrestrial and river ecosystems bordering the river channel, including the riverbank and floodplain (Geyer et al. 2000; Wynn and Mostaghimi 2006). The study team developed a series of criteria to assess how each of these critical factors influences the stability of riverbanks throughout the study area and synthesized these criteria into a continuous measure of riverbank condition. This section describes the methods used to develop and assess these criteria and results of the riverbank condition assessment.

2.1 Methods

2.1.1 Criteria Development

The project team developed a preliminary list of criteria to characterize and quantify critical aspects controlling riverbank erosion and failure, including bank material (e.g., native and naturalized soils vs. various manmade materials) (Hook 1980; Florsheim et al. 2008), the presence and type of vegetation (Geyer et al. 2000, Polvi et al. 2004), and extent of impervious surfaces within the riparian area (Allan 2004), as well as the presence and/or extent of existing features that can further contribute to bank instability, including existing erosion and/or undercutting and manmade structures (e.g., culverts/outfalls).

A preliminary site visit was conducted to optimize the data collection process and redefine characteristics to accurately classify physical attributes. Herrs Island was selected for the initial reconnaissance effort due to ease of access and the varying terrain along the riverbank (Figure 1).

This initial effort led to the determination that several attributes were most effectively and efficiently assessed using existing geospatial data within GIS rather than in the field. It was also determined that several attributes needed to be collected continuously, whereas others were best represented as point characteristics. Point characteristics were defined as any attributes that represented discrete features with lengths of approximately 100 yards or less. The final list of attributes, including the methods for assessing and representing them n GIS, is presented in Table

1. Final criteria are described in detail in the following sections.

Table 2. Finalized list of characteristics included in the riverbank assessment. Attributes are categorized based on assessment methods (i.e., those assessed in the field and those assessed using existing data within GIS), as well as the type of feature each attribute represents (i.e., continuous or point).

Attribute Assessment methods Feature type Bank material Field Continuous Material condition Field Continuous Bank slope Field Continuous Erosion Field Point Undercutting Field Point Structure Field Point Riparian impervious GIS Continuous Culvert/outlet GIS Point

2.1.2 Field Assessment

2.1.2.1 Data Collection

A geodatabase containing all assessment criteria was created using geographic information systems (GIS) [ArcMap, version 10.6.1; Environmental Systems Research Institute (ESRI), Redlands, California] and implemented with the ESRI Collector Application installed on a Trimble TDC-100 Global Navigation Satellite System (GNSS) data collector. The Trimble TDS-100 GNSS data collector leverages global positioning systems and other navigation satellite constellations and enabled the study team to geo-reference all riverbank attributes and associated photos.

Field data were collected via boat traveling along the length of each riverbank for the entirety of the study area. The assessment was completed in three days with an average of 12.7 miles mapped per day. Data collection occurred in early spring (2-3 and 10 March 2020) during the dormant growing season to ensure vegetation foliage would not inhibit line of sight to the riverbank.

Continuous attributes (i.e., bank slope, material, and condition) were used to determine the overall typology of the riverbank at any given location. The study team traveled parallel to the riverbank until a change in one of the continuous attributes resulted in a change in the overall bank typology.

The location of a change in typology was georeferenced using the Trimble TDS-100 GNSS data collector and a new riverbank segment was classified in the data collector geodatabase. Within each riverbank segment, point features were georeferenced and characteristics and photos of those features were collected. Distinct riverbank typologies and segments and associated point feature characteristics of those banks were identified and assessed for all riverbanks within the study area.

Detailed descriptions of each attribute collected during the field assessment are provided below.

2.1.2.2 Continuous Features

2.1.2.2.1 Bank Material & Type

Riverbank material was characterized as either naturalized/native or manmade. As referenced above in Section 1.4, the industrial and engineering modifications to the riverbanks within the city of Pittsburgh and the study area resulted in replacement of most of the native soil with urban fill.

The term naturalized is used in the context of this report to describe the sections of the riverbank containing soil (including modified native soils) and fill materials that have returned to a natural state with minimal or no strategic manmade intervention with varying degrees of sucess. The study assumes all the riverbanks in the study area have experienced some degree of anthropomorphic modification. A soil resource report from the US Department of Agriculture (USDA) Natural Resources Conservation Service (NRCS) Web Soil Survey was created for the study area including the 75-foot riparian buffer (NRCS 2019) and can be found in Appendix A: NRCS Soil Survey of Riparian Corridor within Project Area. Urban land is the dominant soil type that was characterized in the soil report, with lesser amounts of native soil present. Urban land may be fill, rubble, or human transported material. Urban land, as classified by the NRCS soil report, makes up most of the areas characterized in the assessment as naturalized riverbank and may or may not be vegetated.

Naturalized riverbanks were divided into two separate types based on bank material, including those comprised of naturalized or native soils (i.e., ‘soil’) which were generally more stabilized and vegetated and those comprised of visibly altered material and rubble (i.e. ‘fill’) which may or may not be vegetated. Manmade riverbanks were defined as those modified through significant human efforts or alterations and were divided into six different material types: riprap, slag, concrete wall, steel wall, indistinguishable, and other. Each material type is defined below in Table 2, including key observations that differentiate them from one another.

Table 2. Definitions and key characteristics for the three naturalized and six manmade riverbank types.

Material/Type Definition & Key Characteristics Naturalized Riverbanks comprised of soil and fill materials that have returned to a natural state with minimal or no strategic manmade intervention Naturalized and

Native Soil Riverbanks where the underlying materials are native soils or naturalized soil if placed by humans. These soils are likely weathered and loosely compacted and may or may not be vegetated. If no nearby modifications were noted, it was assumed that the soils have not been substantially altered in recent history by human activities.

Fill Riverbanks where the underlying materials were assumed to be placed or compacted with manmade intervention. It was assumed that the presence of manmade infrastructure in the riparian area required site development prior to construction and/or destruction, including the use of fill material to augment the riverbank. Materials directly adjacent or underlying existing or relic development and infrastructure was assumed to be fill.

Manmade Riverbanks modified through significant human efforts or alterations.

Riprap Manmade riverbanks composed of strategically placed natural or man-made stone, including boulders, cobbles, gravels and/or gabion baskets.

Slag Manmade riverbanks with discarded slag ─ a stony byproduct separated from metals during the smelting or refining of ore ─ coating the underlying soil/substrates. This material type was commonly found adjacent to existing or pre-existing industrial plants.

Concrete wall Manmade riverbanks composed of various concrete structures.

Steel wall Manmade riverbanks composed of steel structures (e.g., sheet pile walls).

Other Manmade riverbanks comprised of materials with few occurrences throughout the study area (e.g., timber, brick). Material type was recorded during the field survey.

Indistinguishable Manmade riverbanks for which material composition was not readily discernable.

Throughout the study area, there were occasions when the riverbank consisted of a combination of materials. In these instances, the riverbank was classified relative to the material with the lowest elevation, as this was the material the river is most frequently in contact with. Given the size of the study area and timing of data collection during the dormant growing season (i.e., leaf-off period), it was not feasible to obtain more detailed information on the composition of bank material or the associated vegetation type and structure.

Figure 3. Photographs showing a naturalized riverbank (left) and manmade concrete wall (right) along the Monongahela River in Pittsburgh, PA.

2.1.2.2.2 Material Condition

Riverbank material condition was determined during the boat survey by observing the presence of vegetation, presence of manmade materials, and presence and extent of erosion. the Naturalized riverbank condition was assessed with respect to the presence and general type of vegetation.

Previous research has shown the presence and type of vegetation to be a critical factor in controlling riverbank erosion and failure (Geyer et al. 2000, Polvi et al. 2004). Naturalized riverbanks were assigned to one of three conditions based on presence and type of ground cover vegetation, including herbaceous (non-woody stems, include grasses and ferns), non-vegetated, and woody (trees, shrubs, lianas). A detailed survey of plant species was outside of the scope of this study and was not feasible given the method of data collection (boat survey) and time of year (leaf-off period). The survey did not include an inventory of tree species or mapping of tree canopy.

Geospatial data of the existing tree canopy for the city of Pittsburgh from Tree Pittsburgh (2015) was reviewed (https://gis.pittsburghpa.gov/pghenvironmental/).

Manmade riverbanks had two different material conditions—good/fair and deteriorated/failed— that were assigned based on presence and extent of damage and apparent instability. Each material condition is defined below in Table 3.

Table 3. Definitions and key characteristics of bank condition for naturalized and manmade riverbanks.

Material/Condition Definition & Key Characteristics Naturalized Riverbanks comprised of soil and fill materials that have returned to a natural state with minimal or no strategic manmade intervention.

Non-Vegetated Naturalized riverbanks characterized primarily by exposed soils/substrates and a general lack of vegetation.

Woody Naturalized riverbanks dominated by perennial trees and shrubs having stiff stems and bark.

Herbaceous Naturalized riverbanks dominated by plants that have no persistent woody stems above ground (e.g., grasses).

Manmade Riverbanks modified through significant human efforts or alterations.

Good / Fair Manmade riverbanks that appear to be in-tact and stable and with minimal weathering and damage.

Deteriorated /

Failed Manmade riverbanks that appear to be unstable (e.g., the presence of bare soil/fill that signifies erosion or failure of the bank) and/or extensively damaged (e.g., missing material).

Figure 4. Photographs showing a deteriorating manmade slag (left) and naturalized vegetated slope (right) along Allegheny River in Pittsburgh, PA.

2.1.2.2.3 Riverbank Slope

Slope is an important consideration when assessing the stability of both naturalized and manmade riverbanks. The steepness of a slope can be expressed mathematically as either a ratio [i.e., comparing the horizontal (H) to the vertical (V) component to form a ratio, such as 2H:1V], percent grade (i.e., calculated ratio between the vertical and horizontal lengths), or slope angle (i.e., angle between the horizontal and sloping planes). Table 4 and Figure 5 display the relationship between these three mathematical representations of slope.

Table 4. Slope chart showing the relationship between the horizontal (H) and vertical (V) components of slope as represented by ratio, percent grade, and slope angle

H:V Ratio Percent Grade (%) Slope Angle (arctan V/H) 1:2 200 63.4° 1:1 100 45° 2:1 50 26.6 3:1 33 18.4 4:1 25 14 5:1 20 11.3

Figure 5. Horizontal (H) and vertical (V) relationships represented as ratios, percent grade, and slope angle

Most soils have a natural angle of repose ꟷ the steepest angle a material can be piled without slumping or the surface material sliding ꟷ of approximately 30 degrees (i.e., approximately a 2H:1V slope) (U.S. Department of the Interior Bureau of Reclamation 1963). A slope of 2H:1V is also the maximum recommended slope for many riverbank stabilization techniques, including bioengineering and more traditional engineered solutions (e.g., riprap), although a 3H:1V slope or shallower is more advantageous to maintain stable riverine slopes (U.S. Department of Agriculture, Soil Conservation Service 1989; Mississippi Watershed Management Organization 2010).

For the purposes of this study, slope was recorded by visual approximation as a horizontal to vertical ratio along all riverbanks during the boat survey and recorded as ‘steep” if the slope had a H:V ratio of 2:1 or steeper and ‘shallow’ if the slope had a H:V ratio of 3:1 or shallower (Table 5). A ‘vertical’ category was included for manmade banks, as bank stabilization measures can be engineered to be vertical (e.g., concrete and steel walls); however, naturalized slopes are generally unstable at extreme vertical slopes.

Table 5. Definitions of slope assessment criteria for naturalized and manmade riverbanks.

Material/Slope Definition & Key Characteristics Naturalized Riverbanks comprised of soil and fill materials that have returned to a natural state with minimal or no strategic manmade intervention.

Shallow Naturalized riverbanks visually assessed to have slopes shallower than 2H:1V.

Steep Naturalized riverbanks visually assessed to have slopes steeper than 2H:1V.

Manmade Riverbanks modified through significant human efforts or alterations.

Shallow Manmade riverbanks visually assessed to have slopes shallower than 2H:1V.

Steep Manmade riverbanks visually assessed to have slopes steeper than 2H:1V.

Vertical Manmade riverbanks visually assessed to be fully vertical.

2.1.2.3 Point Features

2.1.2.3.1 Erosion and Undercutting

The presence of erosion and undercutting along naturalized riverbanks provides an indication of bank instability and increases susceptibility to future erosion and failure. Erosion and undercutting were considered only for naturalized riverbanks. Any damage to and loss of material to manmade riverbanks was captured in the material condition criteria. Approximate dimensions were assigned to all instances of erosion and undercutting. Instances of erosion were categorized as being either less than or greater than 50 cubic yards. Both sheet flow erosion and channelized erosion of the riverbanks were assessed, as indicted by bare, displaced soil along the riverbank and preferential flow paths in the soil, respectively. Undercutting, as indicated by the presence of unsupported soils near the water line, was categorized as either less than or greater than 300 linear feet. A linear measurement was utilized due to the unknown extent/volume of undercutting below the waterline.

These approximate dimensions enabled accurate visual estimates.

Figure 6. Photographs showing of examples of undercutting (left) and erosion (right) point features along Monongahela River in Pittsburgh, PA

2.1.2.3.2 Manmade Structures

Manmade structures can affect the potential for erosion and failure of riverbanks by altering channel hydraulics. Manmade structures can also influence the appropriateness of various stabilization techniques and/or help define stabilization goals and objectives. The presence and type of manmade structures intersecting the riverbank were noted during the field survey.

Manmade structures commonly encountered during the survey included concrete or masonry walls, ferry terminals, boat ramps/launches, stairs, and bridge piers. The condition of each structure—good/fair or deteriorated/failed—was also noted.

2.1.3 GIS Assessment

2.1.3.1 Continuous Features

2.1.3.1.1 Riparian Imperviousness

The presence of impervious surfaces within the riparian zone can increase the susceptibility of adjacent riverbanks to erosion and failure by increasing stormwater runoff and reducing riparian vegetation. The extent of impervious surfaces—artificial surfaces covered by water-resistant materials (e.g., pavements, rooftops)—within the riparian zone was visually estimated for each riverbank segment using high definition aerial photography (available through ArcGIS Online).

Riparian zones for each riverbank segment were classified as having greater than 66%, between 33% and 66%, or less than 33% imperviousness. For the purposes of this study, the riparian zone was defined as extending 75 feet up-bank from the edge of the riverbank.

2.1.3.2 Point Features

2.1.3.2.1 Culverts and Outlets

Many sewer discharges and other stormwater outlets and culverts exist along the riverbanks, potentially impacting bank condition and appropriate restoration techniques. For example, point discharges can increase soil saturation, altering slope stability. The study team utilized point locations of culverts and other discharge outlets provided by the Allegheny County Sanitary Authority (ALCOSAN) to characterize and quantify all such discharges within the study area.

Figure 7. Photographs showing examples of culvert point location (left) and impervious surface (right) along riverbanks in Pittsburgh, PA

2.1.4 Condition Assignment

2.1.4.1 Riverbank Typologies

The three continuously measured field attributes (i.e., bank material, bank condition, and slope) were used to establish riverbank typologies that provide an initial assessment of the underlying condition and stability of each riverbank segment. Table 6 describes the notation used to define typologies throughout the remainder of the report. Each riverbank segment was given an overall condition score of either poor, fair, or favorable, depending on its underlying typology. We provide a discussion regarding the condition rating for each naturalized and manmade typology observed within the study area in the subsequent sections.

Table 6. Notation used to define riverbank typologies based on continuously assessed field attributes.

Notations are broken out for naturalized and manmade riverbanks.

Characteristic Attribute Notation Naturalized N Soil Material type So Fill Material type F Non-vegetated Material condition Nv Herbaceous Material condition H Woody Material condition W Shallow Slope Sh Steep Slope St Manmade M Slag Material Type Sl Concrete wall Material Type CW Steel wall Material Type SW Riprap Material Type R Indistinguishable Material Type I Other Material Type O Good/Fair Material Condition GF Deteriorated/Failed Material Condition DF Shallow Slope Sh Steep Slope St Vertical Slope V

2.1.4.1.1 Naturalized Riverbanks

The decision tree for defining and rating naturalized riverbank typologies as either favorable, fair, or poor is provided below in Figure 3. Shallow slopes and the presence of rooted vegetation increase stability and reduce the potential for erosion. Riverbanks composed of naturalized soil or fill materials with either herbaceous or woody vegetation cover and shallow slopes received a ‘favorable’ condition score (i.e., N-F-H-Sh, N-So-H-Sh, and N-F-W-Sh typologies; Figure 3).

Naturalized riverbanks with either herbaceous or woody vegetation cover and steep slopes generally received a ‘fair’ condition score (i.e., N-F-H-St, N-So-W-St, and N-F-W-St typologies;

Figure 3). Most soils have a natural angle of repose less than 2H:1V, which means that soils with a steeper slope are more unstable and likely to fail with added saturation such as precipitation or runoff; however, the presence of rooted vegetation acts to stabilize these steeper slopes.

Naturalized, non-vegetated riverbanks with steep slopes were given a ‘poor’ condition score because they are at increased risk of failure with increased saturation and are not stabilized by rooted vegetation (i.e., N-F-Nv-St typology; Figure. 8). There were no naturalized, non-vegetated riverbanks with shallow slopes identified during the survey, so these typologies (N-F-Nv-Sh and N-So-Nv-Sh) were not assigned a condition or included in mapping.

Figure 8. Decision tree used to define and assign bank condition scores to naturalized riverbank typologies.

Missing typologies and associated condition scores indicate typologies that were not encountered within the study area.

2.1.4.1.2 Manmade Riverbanks

Slag Typologies: The decision tree for defining and rating manmade slag riverbank typologies as either favorable, fair, or poor is provided below in Figure 9. Shallow riverbanks composed of deteriorated/failed slag received a ‘fair’ condition score. Slag is generally more resistant to erosion than naturalized soils and provides protection against wave-action undercutting; however, riverbanks composed of slag also have stability concerns when slopes are greater than 2H:1V.

Thus, steep and vertical riverbanks composed of slag received a ‘poor’ condition score, regardless of condition (M-Sl-GF-V, M-Sl-DF-St, M-Sl-DF-V typologies; Figure 9).

Figure 9. Decision tree used to define and assign bank condition scores to manmade slag riverbank typologies. Missing typologies and associated condition scores indicate typologies that were not encountered within the study area.

Concrete and Steel Wall Typologies: The decision tree for defining and rating manmade concrete and steel wall riverbank typologies as either favorable, fair, or poor is provided below in Figure

10. Steel and concrete walls in good/fair condition were assigned ‘favorable’ condition scores, regardless of the slope. Concrete and steel materials in good condition are exceptionally stable and provide protection against erosion/failure. Deteriorated/failed concrete and steel materials installed at shallow angles were assigned ‘fair’ condition scores because the bank materials provide some protection against soil erosion while the shallow slope limits the potential for large-scale failure.

Steep or vertical concrete and steel walls in deteriorated/failed condition received a ‘poor’ condition score due to concerns regarding continued stability and potential for riverbank failure.

Figure 10. Decision tree used to define and assign bank condition scores to manmade slag riverbank typologies. Missing typologies and associated condition scores indicate typologies that were not encountered within the study area.

Riprap Typologies: The decision tree for defining and rating manmade riprap riverbank typologies as either favorable, fair, or poor is provided below in Figure 11.