Open Collections

UBC Undergraduate Research

A Feasibility Study of Re-introducing a Salmon Population to Jericho Park & Lands Brewer, Julia; Irvine, Keegan; Young, Mason; Wang, Brian 2019-04

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


52966-Brewer_J_et_al_ENVR_400_Feasibility_study.pdf [ 1.71MB ]
JSON: 52966-1.0378564.json
JSON-LD: 52966-1.0378564-ld.json
RDF/XML (Pretty): 52966-1.0378564-rdf.xml
RDF/JSON: 52966-1.0378564-rdf.json
Turtle: 52966-1.0378564-turtle.txt
N-Triples: 52966-1.0378564-rdf-ntriples.txt
Original Record: 52966-1.0378564-source.json
Full Text

Full Text

                                                                                          A Feasibility Study of Re-introducing a Salmon Population to Jericho Park & Lands      Primary Research Team: Julia Brewer Keegan Irvine Mason Young Brian Wang  Research Advisors: Manuela Hyan (MASc Candidate) Dr. Tara Ivanochko Dr. Michael Lipsen Dr. Mark Johnson  Jericho Stewardship Group Partners: Frank Heinzelmann Teaghan Smith     Environmental Sciences 400 - UBC April 2019    Executive Summary In collaboration with the Jericho Stewardship Group, a team of students from the University of British                Columbia has completed a study investigating the feasibility of restoring an ocean-connected stream             channel capable of supporting a returning salmon population on the to be developed Jericho Lands, and in                 Jericho Park in Vancouver, Canada. The Jericho Lands are a 90 acre property on Vancouver’s west side                 and is one of the last remaining consolidated large sites within the City boundaries (De Silva, 2018). This                  feasibility assessment was achieved through addressing three major research objectives.  The first objective was to assess the water quality of hydrologic inputs into Jericho Park. To assess this                  objective, water temperature and electrical conductivity (EC) data were collected over a sampling period              from November to March and a total metals analysis was completed on water samples. The results of the                  analyses indicate that over the sampling period, water temperature and EC were within the optimal range                for supporting chum salmon. C​oncentrations of Aluminum, Arsenic, Iron, Manganese, and Zinc were in              exceedance of British Columbia’s short-term water quality guideline (WQG) for supporting freshwater            aquatic life at one sampling location within the central region of Jericho Park. ​At the concentrations                measured, the metals exceeding the WQG could have lethal or sub-lethal effects on salmon, negatively               impacting their survival. In order to determine the suitability of this water for supporting a healthy                 salmon population, more monitoring of water quality in Jericho Park should take place.    The second objective was to investigate whether enough water could be supplied to the park in order to                  support a suitable stream. This was assessed through completion of a water balance analysis and empirical                hydrologic modelling of the catchment. The results of this analysis indicate that the current hydrologic               inputs into the catchment are sufficient for maintaining a small stream capable of supporting chum salmon                under optimal conditions. However, this would require several modifications to the current urban system              including completion of the City of Vancouver’s proposed Integrated Stormwater Management Plan            (ISMP) to separate and reroute storm sewer pipes, and potential construction of a holding tank to mitigate                 flashy stream discharge and maintain stable water flow throughout the year.   The third objective was to investigate the potential for groundwater input into the future stream system                through an analysis of the region’s surficial and subsurface geology. Through this analysis, a moderately               productive aquifer that could serve as a source of water for the future stream system was found. However,                  this aquifer is located below a confining till layer which is preventing natural recharge and upwelling of                 the water source. Therefore in order to access this water source a well and pump system would need to be                    installed.   To further verify the results obtained in this study, the team recommends implementation of a monitoring                program following the development and the implementation of the proposed Integrated Stormwater            Management Plan. It is recommended that a long-term monitoring routine of the water quality in Jericho                Park is upheld. Long-term water monitoring will allow for better understanding of the change in water                quality and long-term suitability of the water source for supporting a healthy aquatic ecosystem. Total               metals analyses are the best method for analyzing metal concentrations as they allow for detection of all                 2   potentially harmful forms. However for metals where the only dissolved concentrations are reported in the               BC WQG (Aluminum and Cadmium), the total metals concentrations likely overestimates the toxicity             (BC Ministry of Environment & Climate Change Strategy, 2018), and a ​dissolved metals analysis may be                a better indicator of the water quality. It is also recommended that a stream gauge be implemented in the                   existing stream to better understand the current water supply. To determine the viability of withdrawing               groundwater in this region, well installation and pump testing will be required. Future researchers and city                planners may want to consider several additional unknown factors including an analysis of optimal              channel morphology to reduce flow velocity and increase water storage capacity, as well as an assessment                of the most appropriate surface substrate and surrounding riparian vegetation for supporting optimal             salmon spawning conditions.  Authors Julia Brewer Julia studies land, air, and water systems in Environmental Science. Has           completed relevant coursework in ecohydrology, hydrology, groundwater       hydrology, aquatic pollution and ecology. Key skills related to the project include            experience with field sampling, data collection and analysis, and technical          scientific writing. Keegan Irvine Keegan studies land, air, and water systems with a specific interest in            biogeochemistry. Has completed relevant coursework in geology, hydrology,        ecohydrology, and ecology. Has skills in data science and GIS to apply towards             the computer modelling, data collection, and analysis required for this project. Mason Young Mason studies land, air, and water systems in Environmental Science. Has           completed relevant coursework in hydrology, groundwater hydrology, and        ecology. Has experience in field sampling, instrumentation, surveying, mapping         techniques, and data analysis. Brian Wang Brian studies land, air and water systems with an interest in atmospheric            processes and surface hydrology. His project related skills and experience include           fluency in statistical software programming languages for data analysis,         knowledge of python and ArcGIS and experience in the use of equipment for             monitoring stream metrics.   3   Table of Contents  Executive Summary 2 Table of Contents ​4 1. Introduction ​5 1.1 Impacts of Urbanization on Hydrology 6 1.4 Similar Restoration Projects in Vancouver 7 1.2 Selecting Native Salmonids 8 1.3 Benefits of the Project 9 2. Methods 9 2.1 Water Quality 9 2.2 Hydrologic Modelling 11 2.2.1 First Principle Analysis 13 2.2.2 Computer Modelling 15 2.3 Groundwater 18 3. Results and Discussion 19 3.1 Water Quality 19 3.1.1 Visual Water Quality 19 3.1.2 Temperature and Electrical Conductivity (EC) 19 3.1.3 Metals Analysis 20 3.2 Streamflow Modelling 24 3.2.1 First Principle Analysis 24 3.2.2 Computer Modelling - GR4J 26 3.3 Groundwater 30 3.3.1 Quadra Sands Aquifer 30 4. Conclusions and Recommendations 35 4.1 Water Quality 35 4.2 Water Availability 36 4.3 Groundwater 37 4.4 Conclusions and Future Considerations 38 References 39 Appendix 43 4   1. Introduction Prior to urbanization, Vancouver accommodated a network of natural stream channels that various pacific              salmon species would visit to spawn each year. Urbanization of the area caused the hydrological elements                supporting the stream channels to be eliminated, along with the annual salmon runs. Salmon have               traditionally been very important species for both local indigenous communities and colonial settlers as a               staple food and a culturally significant icon representing life in coastal British Columbia (CBC News,               2013). The Jericho Stewardship Group (​​), a non-profit, volunteer-based         organization in coordination with Vancouver Park Board that helps to re-naturalize Jericho Park, has              proposed the daylighting of a streams capable of supporting an annual salmon run as part of the future                  urban development plans for the Jericho Lands. The Jericho Lands (Figure 1) are a 90 acre property on                  Vancouver’s west side and is one of the last remaining consolidated large sites within the City boundaries                 (De Silva, 2018). By re-creating a salmon spawning habitat within Vancouver’s city limits, some of the                historical ecological functioning of the land can be restored within an urban context, providing both               re-connection between citizens and British Columbia’s coastal ecology and re-establishing the various            ecosystem services gained from a healthy freshwater stream environment. Additionally, the creation of             stream channel and native vegetation corridors downhill from 8th Avenue to Jericho Park and the ocean                would create a pleasant trail system, thereby promoting recreational activity and community interaction.  The primary research question for this project is as follows: Given the current urban hydrologic               characteristics of the Jericho catchment, will it be possible to restore a historical salmon spawning stream                network to Jericho Lands that is connected to Jericho Park and the ocean?   This report provides a feasibility assessment of reintroducing a salmon population to the Jericho lands               through the pursuit and study of three objectives:  1. Investigate the water quality of major hydrologic inputs to the Jericho Lands system and assess its                suitability for supporting life and the spawning, survival, and success of a select salmon species. 2. Apply a hydrological model to the Jericho Lands system and determine an estimate for the annual                discharge of this catchment overall which can be used to estimate stream channel dimensions and               flow velocity. 3. Investigate the potential for groundwater input into the Jericho Lands catchment area that could              add additional flow in the proposed stream system.  This project is in the second year of its development, beginning with a previous group of students from                  UBC who conducted a similar study in the 2017-2018 academic year (Li et al., 2018). Their project                 focused on the basic stream characteristics required for the restoration of salmon spawning habitat and               suggested two potential locations for the stream headwaters to be located in accordance with the available                water supply of the area. Our project builds upon this previous work to ensure that a high level of detail                    and quality information is available to use when deciding how best to develop an appropriate stream                channel network.  5    Figure 1. ​Map showing location of Jericho Park and Lands in Vancouver, BC, Canada. A) Map of Greater Vancouver, showing Jericho Park and Lands outlined in red; B) Map of Jericho Park and Lands. Retrieved from Google Maps (2019). 1.1 Impacts of Urbanization on Hydrology Urban catchments have several features which significantly reduce the life-sustaining properties that are             usually associated with healthy stream ecosystems. This effect has been termed “urban stream syndrome’              (Meyer et al., 2005). Changes in surface and sub-surface permeability causes more intense stormflow              events, higher pollutant concentrations in runoff, abnormal stream morphologies, and declining species            richness (Meyer et al., 2005). These changes have caused declines in sensitive fish and macroinvertebrate               populations (Walsh et al., 2005). That being said, these impacts have not necessarily resulted in a                universal decline in species abundance, and many species can still be supported if they are aptly tolerant                 to the conditions (Walters et al., 2003). For this reason, we have selected a model salmon species which is                   the most tolerant to the negative aspects of urban stream syndrome to determine permissible aspects of                6   water quality and stream characteristics. We also provide recommendations to improve the ecological             integrity of an urban stream. 1.4 Similar Restoration Projects in Vancouver Several other stream restoration projects in the Vancouver area have been completed or are currently in                development. A stream restoration project in the nearby Spanish Banks catchment was successfully             completed in 1999 and held 60 returning salmon five years later in 2004 (Rainforest Applied Ecology,                2007). The stream is now stocked each year with salmon fry in hopes that they will return to spawn in the                     stream. Some of the salmon fry released into the stream are from local elementary school students, a                 program which also provides a valuable educational experience related to BC’s natural ecosystems.             Another stream restoration project is taking place just east of the Jericho catchment at Tatlow Park, in                 which they do not expect any fish to return to spawn (Vancouver Park Board, 2018). These two                 restoration projects are located in very different environmental and hydrological settings (Table 1). The nearest successful salmon spawning stream rehabilitation project is Still Creek in Burnaby, located              roughly 22 km east from the Jericho urban watershed (Table 1). Still Creek is 17 km long (average 0.2 %                    grade) with a contributing watershed area of 1050 ha (CVPS, 2002). In 2002, the stream only had 3 km of                    exposed channel, much of which was channelized with concrete, straightened, and had very little riparian               area. The urbanized catchment (68% impervious) fosters conditions of rapid fluctuations in streamflow             during precipitation events (CVPS, 2002). Summer low-flows were recorded at 0.04 m​3​/s, with mean              annual flows of 0.4 m​3​/s. This recorded mean annual streamflow is much larger than the proposed                required streamflow for the Jericho system by Li et al. (2018). However, Still Creek is larger and has a                   larger contributing catchment area. Restoration efforts have returned some ecological functioning to the             stream, ameliorating the water quality and have had chum salmon naturally returning to spawn since               2012. The Still Creek restoration project is a hopeful testament to the possibility of supporting a chum                 salmon spawning habitat in a similar urban environment.               7   Table 1. ​Site comparison for nearby stream restoration projects. Average slope and stream length              calculations were done using ArcGIS. Land use estimates are based on visual estimates of satellite               imagery from Google Maps (2018).    Tatlow Park Spanish Banks Still Creek Jericho Location relative to Jericho East West East N/A Land use ~ 30% pervious ~ 70 % impervious ~ 80 % pervious ~ 20% impervious ~ 30% pervious ~ 70% impervious ~40% pervious ~60% impervious   Presence of fish? Not expected due overall low volume of water contained in the stream network Coho and Chum salmon fry released each year from hatchery  Chum salmon naturally returning  since 2012  N/A Average slope 3.4​o 12​o 0.11​ o  4.2​o Maximum slope  12.7​o  41.5  N/A  41.4​o Stream Length 606 m 1516 m   17000 m  ~4000 m (maximum flow path from ArcGIS) 1.2 Selecting Native Salmonids Prior to urbanization, coho and chum salmon historically used streams in the Vancouver watershed as               spawning grounds (Page et al., 2013). These species are still found spawning in a handful of locations in                  the Greater Vancouver region (Page et al., 2013). Although coho are still released into the nearby Spanish                 Banks stream along with chum (Rainforest Applied Ecology, 2007) and still spawn elsewhere near              Vancouver (Page et al., 2013), we chose to focus on the ecological requirements of chum salmon for a                  variety of reasons. Out of all native pacific salmon species, chum salmon are known to spawn at a greater                   range of habitat conditions (Geist et al. 2002), can survive at the highest maximum temperature limit                (Richter & Kolmes, 2005) and choose to spawn at warmer locations (Geist et al., 2002). This is important                  for survival in our urbanized catchment because the potential water outputs from heating, ventilation and               air conditioning (HVAC) systems and the urban heat island effect which could result in an increase in the                  temperature of water being expelled into the stream environment. Chum salmon fry can also survive               8   through higher rates of fine sediment input into the spawning stream relative to other species (Jensen et                 al., 2009). While this is beneficial to their survivability of chum, fine sediment should not pose a                 significant risk as they would only be actively introduced into the stream channels once all construction                activities in the area have come to an end. The life history characteristics and habitat requirements for                 chum salmon are presented in Table 2.   Table 2.​ Life history data and habitat requirements for Chum Salmon. Adapted from Li et al. (2018).  Full range Mean or preferred range Timing of run and spawning November to January (Groot and Margolis, 1991) N/A  Timing of fry run February to June (Groot and Margolis, 1991) March to end-of April (Groot and Margolis, 1991) Temperatures - 0.1 - 23.8 ​o​C (Brett and Alderdice, 1951) *Survivable range for fry  Incubation: 4.0-13.0 ​o​C, Rearing: 12.0-14.0 ​o​C,  Migration: 8.3-15.6 ​o​C,  Spawning: 7.2-12.8 ​o​C (BC MECC, 2018) Spawning substrate Silt to gravel (Groot and Margolis, 1991) Gravel < 15 cm diameter (Groot and Margolis, 1991) Stream depth (spawning) Exceeds 10cm  (Isaak et al., 2007) 30cm (Isaak et al., 2007) *mean stream depth of investigated salmon bearing stream Stream velocity (spawning) N/A 10 - 30 cm/s (Geist et al., 2002) 1.3 Benefits of the Project Restoring a salmon spawning habitat within a degraded urban ecosystem can have positive effects both               culturally and environmentally. The nutrient released from salmon spawning promotes the growth of             riparian areas around streams which provide natural water filtration and increase species richness of              ecosystems (Bilby et al., 2003). Additionally, salmon spawning areas can increase beneficial dissolved             nutrient content and insect biomass (Minakawa & Gara, 1999). These effects have been found to be even                 greater with chum salmon spawning due to the high-density spawning habits associated with the species               (Bilby et al., 2003). Citizen feedback from the previously mentioned stream restoration projects reveals              that not only positive environmental improvements, but cultural benefits and citizen well-being will also              be afforded through the educational and recreational opportunities created from this project, if it were to                move forward (Vancouver Park Board, 2018). Stream daylighting is part of the City of Vancouver’s               (2016) Integrated Stormwater Management Plan. This project could assist with stormwater management            for the upcoming urban developments on Jericho Lands by diverting stormwater to the stream network,               9   creating a financial incentive by reducing the amount of storm-sewer pipes that will need to be built while                  consequently supplying the stream network with additional water. 2. Methods 2.1 Water Quality Several water characteristics were used as indicators to assess the water quality of the stormwater that is                 feeding into Jericho Park from the Jericho watershed, which would be the major hydrologic input to the                 proposed system. Visual water quality was assessed through direct observation during sampling. Specific             characteristics observed during the visual analysis included the overall appearance of the water (clarity,              turbulence, obvious signs of pollution) and identification of whether the water was flowing, and how               quickly. In addition to the visual analysis, several water quality parameters were measured over time to                establish a baseline of the stormwater input pre-development of Jericho Lands. Measured parameters             include: temperature, electrical conductivity (EC), and total metals concentrations. Stream temperature is            a significant factor in determining the suitability of water for supporting salmon. While chum salmon are                the most tolerant pacific salmon species to high water temperatures (Geist et al., 2002), obtaining               adequate data on the variations in water temperatures over the spawning period was necessary. Chum               salmon do not have a specific preference for EC, however, this parameter was measured as a proxy for                  overall water quality, as an abnormally high EC is generally an indicator for water pollution ​(Das et al.                  2006). As many freshwater fish, including salmon are particularly vulnerable to metals (Price, 2013), and               elevated concentrations of metals are common in urban stormwater (​Göbel et al. 2007)​, metal              concentrations were measured to ensure their levels were within recommended guidelines for freshwater             ecosystems.  Collection of EC and temperature measurements took place at three locations within Jericho Park (Figure               2): two sites along an existing stormwater-fed stream in Jericho Park - the upper stream site (SU) and                  lower stream site (SL), and one stormwater outflow site within the centre of the park - the central forest                   site (FC). In all sampling locations, the water was non-stagnant (i.e., flowing), however, water in the                stream was flowing at a higher rate than in the central forest outflow area. Measurements were taken                 using a Decagon GS3 sensor connected to a ProCheck-12 reading device to measure temperature and EC.                Since we anticipated significant variability in these parameters throughout the year, temperature and EC              measurements took place under a bi-monthly sampling regime from November to March - the general               period in which chum salmon would inhabit the stream.  For the metals analysis, stormwater samples were collected in sterile polypropylene bottles at four              sampling sites within the park (Figure 2): three sites along the stormwater-fed stream - SU, SL and a                  middle stream sampling site (SM), and one stormwater outflow site within the centre of the park - FC.                  Water at the SM site was flowing at the same rate as the SL and SU sites. Sampling took place over three                      days in the beginning, middle and end of February. Thus, in total twelve metals water samples were                 collected throughout the park for analysis. Once collected, the water samples were treated with a strong                acid preservative and stored at 5 degrees Celsius before being submitted for analysis. A total metals                analysis was performed by ALS Environmental Laboratory, in Burnaby, BC. To recover metal             10   concentrations, water samples were digested with nitric and hydrochloric acids, and analyzed. Since the              toxicities of many of metals vary with hardness, total hardness was additionally calculated by ALS as the                 concentrations of calcium (Ca) and magnesium (Mg) expressed as calcium carbonate (CaCO3)            equivalent.  Figure 2.​ Stormwater sampling locations at Jericho Beach Park. Temperature and electrical conductivity measurements were made at the central forest (FC), upper-stream (SU), and lower stream (SL) sampling sites. Water samples for the metals analysis were made at FC, SU, SL, and middle stream (SM) sampling sites. The legend describes the colours associated with sewage types, diamonds represent sewage outflow areas and arrows represent the corresponding flow paths leading to each outflow. Retrieved  from VanMaps (2019).  To determine the suitability of the water for supporting chum salmon, the measured water quality               parameters were compared with optimal water quality guidelines (WQG) for freshwater aquatic            ecosystems documented in reports by the Government of British Columbia Ministry of Environment and              Climate Change Strategy (BC MECC). The WQG for metal concentrations present water quality             recommendations in two forms: a short-term WQG, and a long-term WQG. The short-term WQG              represents the maximum concentration of total and dissolved metals which must not be exceeded at any                time, whereas the long-term WQG represents an average concentration which should not be exceeded              over a 30-day period. In this study, measurements were not made with the necessary frequency to obtain a                  30-day average and thus conclusions regarding water quality are based primarily on the short-term WQG,               where available. 2.2 Hydrologic Modelling For the purposes of our study, we chose to re-examine the available water supply at the park using                  historical climate data to determine how much water could potentially be supplied to a stream through                11   natural hydrologic processes. From this, an estimate of stream size, flow rate, and water supply deficit                was calculated for the spawning season, informed by preferential ranges of stream depth, flow speed, and                spawning season of chum salmon (Groots & Margolis, 1991). Climate data was sourced from a               meteorological station at UBC, using 10 years of data from January 1st, 2009 to December 31st, 2018 and                  all GIS data was sourced from Geogratis Canada and Vancouver Open Data Catalogue. Features of the                urban catchments contributing to the Jericho lands are summarized in Table 3. There are two urban                catchments which could potentially contribute water to the stream without the need for uphill pumping               (Figure 3). This area was delineated by interpreting the storm pipe flow direction and surface slope. The                 combined area of these catchments was used when calculating precipitation and discharge quantities. We              assume that the combined area of these two catchments could be possibly used to supply water to the                  stream because it runs through the development area, meaning that there is potential for both of the                 catchments to contribute water to a stream given proper planning.  Figure 3. ​A map of the two contributing urban catchments that could possibly supply a stream on the Jericho lands with rainwater.        12       Table 3.​ ​Watershed characteristics for the two contributing urban catchments that drain into Jericho Lands. Characteristic Value Source Urban watershed area 495 ha ArcGIS Impervious area 62 % Li et al., 2018 Pervious area 38 % Li et al., 2018 Average slope 4.2​o ArcGIS Maximum slope 41.4​o ArcGIS Dominant SCS soil type C (silty/sandy clay loam) Luttmerding, 1984 Longest flow path ~4000 m  ArcGIS Weighted curve number 85.64 McCuen, 1989 Weighted runoff coefficient 0.25 McCuen, 1989 Annual precipitation -  10 year average 1137.7 mm UBC climate data  Two approaches are taken to investigate this issue of water availability. The first approach is based on                 first principles employing physical hydrological theory to predict annual volume of water leaving the              catchment as streamflow. This approach will help give a sense of stream dimensions assuming the most                optimal conditions tailored to Chum salmon. The second approach is inferring discharge volume based on               established hydrological computer models calibrated to the site. 2.2.1 First Principle Analysis  The goal of the first principle analysis is to create a physical picture of the stream based on the available                    water leaving the urbanized catchment as discharge. First principle draws extensively on a physical              approach which, for hydrological purposes is the conservation of mass. The law of conservation of mass                implies that the storage within the catchment (a hydrological area outlining zones where each drop of                precipitation will end up in the same outlet going to the sea) will equal the amount of water entering the                    catchment minus from the amount of water leaving the catchment. This creates the relationship outlined               in Equation 1:  13   Storage Input OutputΔ =  −  (Eqn. 1)  The only input of water into the catchment is precipitation ( ) assuming no additional input from the          P        sewage systems and from groundwater. This simplification to ignore groundwater interactions are due to              the dominant Langley-Cloverdale surface soil type which are moderately poor to poorly drained             (​Vancouver Soil Map: Langley-Cloverdale Soil Management Group, n.d.). The area’s sub-surface           geology also supports this assumption and will be elaborated in the groundwater section below.              Precipitation data from 2009 to 2018 was obtained from the UBC Climate Station. Outputs of water from                 the catchment are evapotranspiration ( ) and overland discharge ( ) assuming no water piped out of    TE     Q        the catchment through the stormwater system or as groundwater. Eliminating anthropogenic piping of             stormwater is a reasonable assumption in light of the Integrated Stormwater Management Plan (ISMP)              proposed by the Vancouver’s Greenest City Action Plan and Climate Change Adaptation Strategy. The              ISMP calls for the total separation of sanitary sewers and stormwater which we assume could then be                 diverted to the stream. Equation 2 shows the completed catchment water balance equation:  Storage P  ET  Q Δ =  −  −  (Eqn. 2)  ET is calculated from multiplying the surface crop coefficient ( ) by potential evapotranspiration         Kc     (​PET​). Potential evapotranspiration is calculated according to Priestley-Taylor equation (1972) using the            EcoHydRology R package. This function calculates PET from mean air temperature, latitude and             longitude of the catchment (used to calculate incident solar radiation), albedo (assumed to be 17%), slope                and aspect of the catchment. The albedo value is estimated by the sum of the albedo for a grass forested                    mixed area (average of grass and deciduous forest albedo is 0.2) multiplied by 38% pervious surface                cover and albedo for a suburban snow free surface (0.15 according to Oke (2017)) multiplied by 62%                 pervious surface cover (Oke, 2017; Li et al., 2017).  The surface crop coefficient is an experimentally derived variable, assuming no limitation on plant              growth, that shows the ratio of the actual crop evapotranspiration to potential evapotranspiration (​Single              crop coefficient (Kc), n.d.​). In the Jericho catchment, the two main crops are assumed to be grass fields                  and coniferous forests. for grass is 0.85 and for forested stands are 0.78 (​Single crop coefficient   Kc      Kc         (Kc), n.d.; Wang et al., 2007​). Visually, the park seems to have equal coverage of grass and forest                  therefore taking the average of the two coefficients yields a weighted value of 0.815.Kc   where T  K ET  E =  c * P K  0.815 c =  (Eqn. 3)  In order to find discharge, change in storage is assumed to be zero over a water year (October to October                    of next year). Water storage of the catchment goes up during the winter season and gets depleted during                  the summer season. This assumes that over a year the change in storage is zero. This assumption allowed                  equation 2 to be rearranged into Equation 4.    P  ETQ =  −  (Eqn. 4)  14   To consider the urbanized component of this catchment, it is likely that stays the same however            P     TEwill be increased due to impervious surface cover. A similar approach as the crop coefficient can be done                  to find the actual over an urban catchment by introducing a term. is calculated from finding    TE         KU  KU     the ratio of annual actual data to the annual calculated using the climate data. Actual data     TE     ETP       TE  are sourced from Vancouver Sunset Urban Climate Research Tower where its flux footprint covers an               urban residential area similar to that of Jericho Catchment (Christen, n.d.). Data was analysed for the                entire year of 2017 and 2018 separately to find the average to be 0.719. To get the total value           KU        TE  each component is multiplied by the proportion of the catchment that is impervious (characteristic of               urban land cover surface) which is 68% and proportion of the catchment that is pervious (characteristic of                 natural land cover surface) which is 32% (Li et al, 2018).   T  [(0.32 A ) ET ] [(0.68 A ) ET ],E urban =  *  catchment * Kc * P +  *  catchment * KU * P  (Eqn. 5)   where K  0.815, K .719 c =   U = 0   Under optimal condition for observing the largest streamflow, it is assumed that soil is at saturation where                 the difference between and are all observed as discharge in the river ( ). Soil saturation processes   P  TE         Q     are also ignored. Doing this calculation at an annual scale will yield an annual value as a volume. This              Q       will indicate the total amount of water available each year to be fed into the stream. This value is summed                    across the entire year which will overestimate summer dry periods and underestimate winter wet periods.               To get a sense of what the stream physically looks like, it is possible to use stream velocity and stream                    depth that is prefered for Chum salmon to calculate stream width assuming a rectangular stream channel.                Although a rectangular stream channel is unrealistic, this shape will provide maximal stream width. The               prefered stream velocity by chum salmon is 10 to 30 cm/s (Geist et al, 2002) and the prefered stream                   depth is 13 to 50 cm (Groot & Margolis, 1991). Taking the maximum stream velocity and maximum                 depth will result in a minimum channel width and vice versa for a maximum channel width. Equation 6                  shows the equation to convert Q to channel width.  hannel width 1 year)/(3.15e7s) 1/depth) 1/stream velocity)  C = Q * ( * ( * ( (Eqn. 6)  Maximum stream velocity can be computed using the same stream channel assumption. Assuming stream              width of 2m which is aesthetically pleasing while spatially reasonable. Stream depth is assumed to be                30cm which is the mean stream depth of Middle Fork Salmon River (Isaak et al., 2007). Equation 7 is                   used to convert Q to stream velocity.  tream velocity 1/(3.15e7)) 1/depth) 1/width)  S = Q * ( * ( * ( (Eqn. 7)  Note that is the volume of water leaving from the outlet point of the entire catchment (where all  Q                  streamflow leaves the catchment and into the ocean). Therefore this stream width calculation is at the                outlet point and how this stream width decreases with distance upstream needs to be investigated in a later                  study in order to get a profile of stream width with distance upstream.  15   2.2.2 Computer Modelling  An empirical modelling technique was used to assess how the physical characteristics of the site would                affect the discharge of a hypothetical stream channel network. Initially, a variety of lumped and               semi-distributed models were considered but the data required to effectively calibrate them and provide              meaningful estimates of parameter values was unattainable so we selected a simpler model with fewer               requirements. The selected model is an empirical model called GR4J (Pagano et al., 2010), which               produces daily flow values for a catchment with four parameters to optimize. These four parameters are:                the root zone storage capacity, the catchment water exchange coefficient, the one-day maximal capacity              of the routing reservoir, and the unit hydrograph base-time for the catchment. Although this is a more                 simplistic model, there is evidence to suggest that models which use few free parameters can still provide                 good approximations of streamflow in many hydrological settings (Perrin et al., 2001). GR4J’s data              requirements include daily precipitation, potential evaporation, air temperature, and streamflow. The           10-year climate data time series was grouped by julian day to find average daily precipitation and air                 temperatures. This was done to represent a climatic average for Vancouver to use in streamflow estimates                which would then be used to calibrate GR4J. The climate datasets contained several small breaks in the                 record which were gap-filled using spline interpolation.   Since we are simulating a non-existent stream, we estimated potential daily streamflow characteristics             using several different methods. First, daily streamflow rates were estimated using a modified rational              method:  Q​R​ = C*A*(i-PE) (Eqn. 8)  Where Q​R is daily flow rate (mm​3​/day), C is the runoff coefficient, i is rainfall intensity (mm/d), PE is                   potential evaporation (mm/day), and A is the area of the catchment (mm​2​). The runoff coefficient was                determined using the average slope of the catchment, the dominant soil type, and a weighted average for                 the pervious and impervious areas of the catchment. Runoff coefficients are empirically derived for              different surfaces and represent the ratio between overland flow and infiltration. For impervious areas, C               equals 0.6 assuming the area is representative of single-house residential areas resting on a slope between                2​o - 5​o​. For pervious areas, C is being treated as a combination of fields and forest with silty clay-loam                    soils and equals 0.11. The weighted average for both areas would be 0.41. Although the rational method                 was developed to calculate peak discharges from hourly precipitation events, we are repurposing it as a                way to calculate the proportion of water that will runoff due to daily rainfall rates in relation to surface                   type and magnitude of potential evaporation for an urban system. This method is very similar to how the                  streamflow was modeled by Li et al. (2018).  We also used the curve number (CN) method to calculate a different estimate of discharge (Q​CN​):  S = 1000/CN - 10 (Eqn. 9) I​a​ = 0.2*S  (Eqn.10) Q​CN​ = A* (P-PE-I​a​)​2​/(P-PE+0.8*S) (Eqn. 11) 16    Where S is the potential maximum retention, CN is the weighted curve number for the catchment                (unitless), I​a is initial abstraction due to to features in the landscape (unitless), P is daily precipitation                 (mm), PE is potential evaporation (mm), and A is the area of the catchment (ha). The CN for the                   catchment (CN​w​) was calculated using the weighted formula for urban land use:  CN​w​ = CN​P​(1-f)+f*98 (Eqn. 12)  Where CN​P is the CN corresponding to the soil type of the pervious area, f is the fraction of impervious                    area, and 98 is the CN associated with the urban environment.  Potential evaporation was calculated using the Oudin formulation and was used for both the rational and                CN method calculations (Oudin et al., 2005):  PE = λρRe * 100T +5 a  (Eqn. 13)  Where R​e is global radiation (W m​-2​), is the latent heat of vaporization for water (kJ K​-1​), is the       λ           ρ   density of water (kg m​-3​), and T​a is mean daily air temperature (K). PE is considered equal to zero if air                     temperatures are below -5 ​o​C. To calculate T​a​, a 10-year daily average was taken between 2009-2018.                Global radiation is estimated using the latitude of Vancouver (49.3​o N) and solar incident angle as a                 function of the date. Values for potential evaporation will generally be lower than potential              evapotranspiration, however the future vegetation cover over the catchment is unclear so transpiration             values will likely change. Also, potential evaporation had been shown to provide accurate modelling              results when using GR4J (Oudin et al., 2005).  Travel time for the watershed was calculated using the Carter lag equation (McCuen, 1989) for partially                sewered watersheds:  t​c​ = t​t​ = 100 * L​m​0.6​ * S​m​-0.3 (Eqn. 14)  Where t​t is travel time (minutes), L​m is the length of longest flow path (miles) and S​m is the average                    gradient of the watershed (%). Since we are simulating a unit hydrograph, we have to assume that time of                   concentration (t​c​) is equal to the travel time of the watershed (t​t​). This equation was calibrated for a                  similar urban watershed that has a significant amount of water channeled to the outlet through pipe-flow.                It gives us a travel time of 0.26 days. Our synthetic unit hydrograph would then have a base-time of 2t​c                    equal to 0.52 days. This means that throughout a precipitation event, all runoff will be routed through the                  watershed within ~12 hours.  The rational and CN method can be used to approximately simulate streamflow for the urbanized               catchment because both methods model runoff rates assuming very little subsurface storage (i.e. flood              events). Due to the catchment having a poorly drained soil type and a large impermeable area, we assume                  that the urban catchment will respond similarly to a rain event in a natural catchment with saturated                 17   subsurface storage. Other than the modelling assumptions stated in this section, there are several others               which are summarized in the appendix (Table A1).   The 4 parameters used by GR4J were calculated in several different ways to provide a range of simulated                  streamflow values with differing assumptions, as summarized in Table 4. GR4J parameter calibrations             using streamflow generated from the rational and CN method (Equations 8-13) was done using the airGR                package in R. The GR4J model was also run twice using parameters that were selected, termed custom                 calibration 1 and 2. Custom calibration #1 (C1) assumes that there is no storage in the catchment, no                  water lost or gained from other catchments, and no water stored in the routing reservoir. This parameter                 set is indicative of a heavily urbanized catchment that will direct all precipitation to direct runoff. Custom                 calibration #2 (C2) assumes that there is some available soil water storage, a significant amount of water                 lost from the catchment via sewers, and a small amount able to be stored in the routing reservoir. C2                   represents a rough approximation to the current watershed characteristics. For modelling an urban             environment where no groundwater interaction is assumed, we consider intercatchment exchange to be             equivalent to the water lost into to the storm sewer system.   Table 4. ​The four parameters used by GR4J optimized for our catchment using various assumptions. Both the CN and Rational method calibrations were done using the Michel calibration found within the airGR R package. Parameter Rational Method Calibration CN Method Calibration C1 C2 Root zone storage capacity (​mm​) 1.0x10​5 1.2x10​5 0 1500 Intercatchment exchange (​mm/day​) -13.1 -6.4 0 -1000 Routing reservoir storage capacity (​mm​) 5.8 4.1 0 5 Base time of the unit hydrograph (​days​) 0.52 0.52 0.52 0.52  Using the modelling results, stream width was calculated from the mean discharge over the period that                salmon would be in the stream, along with a preferred flow velocity and depth for spawning chum salmon                  (Equation 6). Water deficit was then defined as the amount of water that must be supplied to the stream                   each day to maintain a constant streamflow suitable for salmon spawning. A mean daily deficit was                calculated for each month to show how the requirement for additional water will change throughout the                spawning season. Using the same stream dimensions we calculated the potential velocity (V) of the               stream using the Manning Equation (Equation 15; McCuen, 1989):  V = (1.486/​n​)*R​h​2/3​*S​1/2 (Eqn. 15)  18   Where ​n is the manning roughness coefficient (unitless), 1.486 is a conversion factor, R​h is the hydraulic                 radius (m), and S is the slope of the stream (ratio). A manning roughness coefficient of 0.035 was selected                   assuming the stream would have bed conditions of a natural gravel stream channel with some grass and                 weeds. The hydraulic radius is the ratio between the cross sectional area of the stream and the wetted                  perimeter. 2.3 Groundwater Groundwater can be an important source of streamflow generation, especially during periods of             no rainfall in which groundwater discharge can potentially sustain streamflow as part of base flow (Miller                et al., 2016). Deep stores of groundwater that do not naturally recharge streams can be accessed manually                 using well and pump systems. Due to the complexity and cost of drilling wells, however, direct                measurements to determine the viability of using groundwater as a significant input to streamflow were               not conducted in our study. Instead, we reviewed available groundwater and geological information to              estimate the potential of using groundwater as a source of input. Geological maps, cross sections, and                literature combined with lithological data from wells were used to understand the surficial and sub               surficial geological characteristics of the proposed development site and the surrounding area.   Geology largely influences the way that water moves and is stored in the ground and is therefore an                  important aspect to consider when trying to understand groundwater availability. Porosity and            permeability, which depend on the size, shape, and mixture of grains and particles that compose soil and                 rock, for instance, are the two most important properties that dictate the storage and flow of water in                  porous materials (Brickeret et al., 2017). Hydraulic conductivity is a parameter that describes how easily a                fluid flows through a porous medium, and is dependent on permeability. Groundwater flows much faster               through materials of higher permeability or hydraulic conductivity, such as sand, compared to materials of               lower permeability or hydraulic conductivity, such as silt and clay (Van der Spek et al., 2013). In solid                  bedrock, which typically has very low hydraulic conductivities, most of the water is stored in and                transported through fractures (Blessent et al., 2011).  Most of the existing groundwater information was obtained from the Government of British Columbia’s              Provincial Groundwater Observation Well Network Database which contains reports on Vancouver’s           aquifer systems. The information provided in these reports are based on the raw data such as well records,                  water quality data, reports, air photographs, and geologic maps that were used to interpret the aquifer.                Because there are no active groundwater monitoring wells located in or near the development site, the                exact depth to the water table and confining layer, aquifer thickness, and possible well yield for the area                  of concern is unknown, but some assumptions can be made based on available information.  19   3. Results and Discussion 3.1 Water Quality 3.1.1 Visual Water Quality In general the stormwater running in the forest stream and central park outflow site appeared clear with no                  obvious signs of pollution or excess sediment. At the FC site, the water runs through a forested area                  containing a fair amount of leaf litter and dirt and as a result this site had a murkier appearance than the                     forest stream sites. At the SU sampling location, an oily sheen layer was observed on top of the water on                    two of the sampling dates in the middle and end of February suggesting potential oil or gasoline based                  contamination. 3.1.2 Temperature and Electrical Conductivity (EC) Over the sampling period, the average water temperature was 7.6 ​o​C. A maximum temperature of 10.7 ​o​C                 was measured at the SU site in early December, and the minimum temperature of 2.5 ​o​C was measured at                   the SL stream site in early February (Figure 4). Optimal temperatures for chum salmon vary throughout                their life cycle. Optimal ranges for incubation, rearing, migration and spawning are 4.0-13.0 ​o​C, 12.0-14.0               o​C, 8.3-15.6 ​o​C, and 7.2-12.8 ​o​C respectively (BC MECC, 2018). Although the minimum temperature              recorded was outside of the optimal range, the species is likely to have a tolerance for temperatures                 outside of its optimal range in the short term. A study by Brett and Alderdice (1951) found that lethal                   temperatures for chum salmon fry are below -0.1 and above 23.8 ​o​C.   It is important to note that measurements were only taken over the period of November to March and                  therefore did not capture temperatures over the full range of the fry-run which generally occurs from                February to the end of April ​(Table 1). Higher maximum and average temperatures could be expected if                 measured over the entire period in which salmon inhabit the stream. Additionally, the month of February                during the year of sampling (2019) was much colder than typical with an average temperature of 0.4 ​o​C                  compared to the 4.9 ​o​C typically experienced in February. According to an Environment Canada              meteorologist’s report, this was the coldest February on record in Vancouver (Daily Hive, 2019), which               may have additionally skewed the average water temperatures measured to the low end. Thus the results                can be considered a lower estimate of the water temperatures to be expected in Jericho Park and indicate                  that the temperatures will likely not exceed lethal temperature limits.   The average EC measured over the sampling period was 0.267 dS/cm with a minimum recording of 0.167                 dS/cm and a maximum of 0.505 dS/cm (Figure 4). According to a BC MECC report (2015) typical EC                  values for freshwater range between 0.05 dS/cm and 1.5 dS/cm. In general, EC has a linear relationship                 with total dissolved solids (TDS) and therefore high EC values are often an indicator for water pollution                 (Das et al. 2006). The EC values measured in our study were on the low end of what is typical for                     freshwater systems, suggesting that there is no obvious pollution source contaminating the stormwater             feeding Jericho Park.  20    Figure 4.​ Time Series of measured Temperature and electrical conductivity of stormwater outflowing into Jericho Park. Optimal parameter ranges obtained from BC Ministry of Environment & Climate Change Strategy (2015; 2018) are represented by the area between the black dotted lines. The red dotted line represents the average over the sampling period. 3.1.3 Metals Analysis A summary of the BC MECC Guidelines for freshwater aquatic species (Table A2), and a report of the                  total metals analysis (Table A3)  are presented in the Appendix section of this report. The results of the metals analysis shows that distinctive differences in metal concentrations were              measured between the different sampling sites (Figure 5). In particular, with exceptions for boron and               molybdenum, samples collected from the central forest (FC) sampling site contained higher levels of              metals than the three forest stream sites. Although there was not a sufficient amount of data collected to                  perform a formal statistical comparison, these differences were consistently one order of magnitude or              greater. In the cases of aluminum, iron, and lead, the concentration differences between the FC site and                 the lower stream (SL) site were in the order of two magnitudes or more (Table A3). Lower concentrations                  of metals were measured at the SL site compared to the middle stream (SM) and upper stream (SU) sites.                   In general, metal concentrations were observed to decrease moving downstream which could suggest that              a filtering effect took place as water moved through the streambed substrate and pollutant concentrations               dissipated.   There are a few factors which may have contributed to the discrepancy in metal concentrations measured                at the FC site compared to the three forest stream sites. Firstly, as mentioned in the introduction of this                   21   report, the water at the FC site flowed at a much slower rate than in the stream. The water level at the                      central forest site was also much shallower than the stream sites and was murky due to decomposing leaf                  litter. The shallowness and slow moving nature of the water likely contributed to the elevated levels of                 pollutants at the FC site site as the rate of dispersion of contaminants would have been lower, resulting in                   the pollutants being more concentrated. Another important factor distinguishing the FC site from the other               sites is that the stormwater outflow which sustains the water in the FC site is drained from a different area                    of the Jericho catchment than the outlet which supports the forest stream sites. As seen in Figure 1, the FC                    site is supported by stormwater draining from the east side of Jericho Park while the SU, SM, and SL                   sites are supported by a storm sewer pipe draining from the northwest region of the park. Therefore, a                  pollution event may have occurred somewhere on the eastern side of the Jericho catchment contaminating               the water measured at the FC site exclusively. In order to better understand the characteristics of the                 stormwater draining into the FC region, more research should be done on its specific flow path in order to                   identify potential contamination sources along this path.   The results of the analysis also showed that in a few cases, the concentrations of metals measured                 throughout the sampling period were in exceedance of the BC MECC (2018) water quality guidelines               (WQG) for freshwater aquatic life (Figure 5). As reported in Table 5 and observed in Figure 5, the                  measured concentrations of aluminum (Al), arsenic (As), iron (Fe), manganese (Mn) and zinc (Zn) were               in exceedance of the short-term maximum recommended levels. These exceedance events all took place at               the FC site. The concentrations exceeded the WQG on all three sampling days at the FC site for Al and                    Fe, on the first two days at the FC site for Zn, and on the first sampling day only at the FC site for As and                          Mn (Table A3). Throughout the month of February, metal concentrations for those metals exceeding the               standard at the FC site decreased consistently for each subsequent measurement (Table A3). This could               suggest that a pollution event occurred early on in February (or in an earlier month) which lead to                  elevated metal concentrations in the water flowing into the FC site. Over time, metal concentrations at                this site decreased as they became diluted with cleaner water throughout the sampling period. This theory                however relies on various assumptions since there was not enough data collected to obtain a full time                 series of the metal concentrations at this location. Further monitoring of the metals in the stormwater at                 the FC site should be done to determine if the water from this source is viable for the subsistence of                    freshwater aquatic species including chum salmon. 22   Figure 5. ​Total metal concentrations by site for cases where water quality guidelines (WQG) were exceeded, measured in mg/L. Asterisks represent the BC Ministry of Environment & Climate Change Strategy (2018) WQG concentrations for freshwater aquatic life in mg/L. WQG for manganese and zinc varied with hardness, thus multiple asterisks are displayed on the plots for these metals. The ​upper and lower bounds for each box is the interquartile range, and the middle bar is the median​.    In this study, comparisons could only be made with the short term WQG for metals, thus conclusions                 cannot be made regarding the long-term (30 day average) concentrations. This may be a more valuable                measure of the overall water quality of these sources as it would minimize potential unusual spikes in                 contaminant concentrations due to, for example, an unusual pollution event. There are two important              things to note about the WQG for Al specifically. Firstly, the WQG varies depending on pH below a pH                   of 6.5. Water pH was not measured as a parameter in this study, and thus the pH was assumed to be                     greater than 6.5. This assumption was based on an average, freshwater pH value of 7.5 for the Lower                  23   Fraser Valley, reported by McKean and Nagpal (1991), with the British Columbia Ministry of              Environment. However, the pH of the water measured at our sampling sites may have varied from this                 due to the impacts of urbanization. Secondly, the WQG for Al was set using dissolved aluminum and in                  this study, a total Al analysis was performed. Total metals analyses are recommended by the BC MECC                 (2018) because they ensure that all forms of the metals that may potentially be toxic are included in the                   measurement. This allows for complete confidence in the safety of the water in the event that the                 concentrations detected are below the guideline. However, it is possible that a large fraction of the metals                 measured are in forms that are not biologically reactive and thus total metals analyses usually               overestimates toxicity to a considerable extent, especially in a turbid water body (BC MECC, 2018).               Thus, although the detected concentrations of total Al were above the WQG in this case it cannot be                  concluded with confidence that this water is unsuitable for supporting aquatic species. In order to confirm                this, further water quality assessment including a dissolved metals analysis should be performed. Potential Impacts of Measured Metal Concentrations In general the WQG are based on toxicity levels for the most sensitive species, and represent                recommended levels for the survival and health of these species. Although measured metal concentrations              at the FC site exceeded these guidelines, this is not to say that all aquatic species would be unable to                    survive in such conditions, however, exposure may cause other negative impacts which reduce their              success. Most aquatic species, including salmon are particularly vulnerable to metals due to continuous              direct contact with water, and the high solubility of metals in water. In general, metals cause harm to fish                   through disrupting gill function, and the olfactory system, which are both essential for survival (Price,               2013). Sub-lethal effects associated with metal toxicity in salmon include alteration of behaviors such as               predator avoidance, foraging, and migration, as well as physical impairments related to growth and              development, swimming efficiency, and immune system responses (Price, 2013). All of the metals             measured in this study can negatively impact salmon to some degree, even below levels known to cause                 lethal effects. According to research by Price (2013), there have been several instances reported in               literature where metals such as Cu, Ni, and Zn have caused sub-lethal toxic effects in various species of                  salmon at concentrations below the British Columbia WQG for freshwater aquatic species. In contrary,              the BC WQG for Al, Cd, Pb, and Ag have been found to protect salmon against both lethal and sublethal                    toxic effects. This suggests that multiple lethal or sublethal effects could be observed in chum salmon and                 other species of fish, due to exposure to the concentrations of metals measured at specific sites in this                  study. Sources of Metals in Urban Stormwater Since Jericho lands are located in a heavily urbanized area, various factors may have contributed to the                 elevated concentrations of metals in the stormwater draining into this region. In urban environments, one               of the major sources of heavy metals is industry and vehicle traffic combustion. Jericho Park and Lands is                  surrounded by several heavy traffic roadways including 4th Avenue, and Northwest Marine Drive. Heavy              metals produced in combustion processes originate as dust particles which settle on ground and plant               surfaces through dry and wet deposition and may eventually get transported in stormwater runoff (Göbel               et al. 2007). Since the sampling sites were fed by storm sewer pipes draining the roads and lands                  surrounding Jericho, deposition of material from vehicle emissions may be significant source of metal              24   contamination in the stormwater in this region. Another potential source of metals in the stormwater in                this region is leaching from roofs and pipes. Many of the metals sampled in this study (e.g. Al, Cu, Pb,                    Zn) are commonly used in roofing, piping and gutter systems in residential areas (Göbel et al. 2007). This                  could be especially relevant in the context of this study due to old infrastructure on the Jericho Lands                  from the historical garrison. Rain which falls and collects on these surfaces could dissolve metal               materials incorporating them into the stormwater and transporting them as the runoff washes away.              Additionally, common types exterior paints have been found to contain various metals such as lead​, zinc,                cadmium, manganese, nickel, copper, cobalt, and chromium (Mielke et al., 2001). Therefore ​elevated             metals concentrations could have resulted from paint or other common residential and construction             materials washing or being dumped into storm sewer catch basins.  3.2 Streamflow Modelling  3.2.1 First Principle Analysis The objective of the first principle analysis is to analyze how much water in the Jericho catchment is                  available as discharge in a black box model and the maximized channel width and streamflow associated                with that discharge volume.   Annual Discharge  Over the 10 years, mean annual precipitation is found to be 1138.89 per meter square of the           mm       catchment. Mean daily evapotranspiration rate is found to be 1.60 per day. The mean annual         mm       discharge volume, as the residual of the P and ET, is found to be 2748149 which is equivalent to 0.09              m3      . Figure 6 presentes the above result in an annual basis over the 10 years as a time series./sm3      Figure 6.​ Annual precipitation, discharge and actual evaporation for the Jericho catchment on an annual basis from 2009 to 2018. Precipitation and temperature data used to calculate potential 25   evapotranspiration are sourced from UBC Climate Station. Discharge is calculated as the residual of the two terms.   Maximized Channel Width  The reason for finding the maximized channel width is to see if the water available can produce a                  reasonably wide stream. Channel width is calculated from discharge assuming a rectangular stream             channel. Average maximum channel width is 6.7 meters calculated from streamflow of 10 cm/s and               stream depth of 13cm. Minimum channel width is 0.58 meters calculated from streamflow of 30 cm/s and                 stream depth of 50cm. Logically speaking a 6.7 meter wide stream locating in Jericho Lands is                unreasonable but it does indicate the catchment has the capacity of supporting a wide shallow stream.                Table 5 presents the above result in an annual basis over the 10 years.   Table 5. ​Channel width calculated from the preferred streamflow (10 to 30 cm/s) and the prefered stream depth (13 to 50 cm). Minimum channel width is calculated from taking the maximum streamflow and maximum depth and vice versa for calculation for maximum stream width. Year Maximum Channel Width (m) Minimum Channel Width (m) 2009 0.474 5.468 2010 0.557 6.432 2011 0.408 4.709 2012 0.589 6.791 2013 0.294 3.388 2014 0.643 7.417 2015 0.525 6.057 2016 0.792 9.135 2017 0.660 7.614 2018 0.868 10.015  Maximized Streamflow  Streamflow is calculated assuming a fixed channel width of 2 meters and a constant depth of 30cm which                  is the mean stream depth of a salmon spawning stream in Idaho that was studied by Isaak et al. (2007). 10                     year average streamflow are found to be 14.5 cm/s. Table 6 presents the above result in an annual basis                   over the 10 years as a time series. The preferential range of stream flow rate for chum salmon was found                    by Geist et al. (2002) to be 10 cm/s to 30 cm/s. The calculated streamflow is in the lower range (below 15                      cm/s), 6 of the past 10 years, indicating the possibility of water shortage. An extremely dry year was                  observed in Vancouver in 2013 with only 848.50 mm of total precipitation. Due to this, a maximal                 streamflow rate of 7.34 cm/s was achieved over this year, which is not within the preferential range                 26   outlined by Geist et al. (2002). In light of climate change’s effect on precipitation, it introduces                uncertainties regarding water availability in future years.  Table 6.​ Stream flow rate calculated using channel width of 2m and a depth of 30cm.  Year Annual Mean Streamflow (cm/s) 2009 11.8463 2010 13.93576 2011 10.20377 2012 14.71309 2013 7.341511 2014 16.06966 2015 13.1228 2016 19.79149 2017 16.49631 2018 21.69963 3.2.2 Computer Modelling - GR4J Mean flow rate for the spawning season varied widely with model runs, indicating that flow speeds will                 differ depending on the degree to which the catchment is urbanized (Table 7). The C1 modelling scenario                 shows abnormally high flow rates, influenced by a totally impermeable urban catchment that is diverting               all rainwater to a single outlet. This is a highly unlikely discharge value whose purpose is to merely                  demonstrate the magnitude of influence that catchment storage has on controlling streamflow. The other              scenarios show mean discharge speeds that are comparable to what was found in the first principal                analysis and are better estimates for the current characteristics of the catchment. Out of these scenarios,                the CN calibration showed the largest discharge values, owing to having the largest root zone storage                capacity which enables greater base flow and less flashy hydrographs. It is important to note that the                 rational and CN method calibration runs are assuming very large amounts of soil moisture storage and                little inter-catchment exchange to produce a conservative discharge estimate. Table 7. ​Results from all runs of the GR4J model. Parameters used in each model run are presented in                   Table 4 above. Nov-June mean flow values represent expected discharge during the time period the               stream would be hosting salmon.  Rational Method Calibration CN Method Calibration C1  C2  Annual mean flow (m​3​/s)  0.059 0.125 15.00 0.051 Nov-June mean flow (m​3​/s)  0.084 0.173 9.09 0.092 Annual flow volume (L) 1.8x10​9 3.9x10​9 2.9x10​11 2.9x10​9 27      Figure 7.​ Annual simulated hydrographs from the four different GR4J model runs showing discharge at the outlet. The shaded area represents the period when salmon will be present in the stream.  The mean flow rates from the C2 scenario are comparable to the CN and rational calibrations (Figure 8),                  showing that root zone storage can be greatly decreased if water loss due to inter-catchment exchange is                 increased. Water loss from the catchment due to natural inter-catchment exchange is unlikely in an urban                context, so this parameter is better regarded as the amount of water lost to the storm drainage system.                  Detailed annual hydrographs for each model scenario are shown in Figure 7 which reveal how even                though the mean flow rates can be similar (Figure 8), the C2 hydrograph is much flashier than those of the                    rational or CN method.   28    Figure 8. ​Average discharge for 3 model scenarios. The leftmost and rightmost shaded segments of the graph indicate the periods which salmon would be present in the stream (November - June).  Possible stream widths were calculated using the simulated discharge estimates, and preferential ranges of              spawning velocity and depth, found in Table 1. Using mean stream discharges from the CN, rational, and                 C2 model scenarios, mean stream width was found to be 3.2m over the flow season (Nov-June), with a                  height of 0.3m and flow velocity of 0.2 m/s. Although this stream width may be able to be supported                   during times of steady flow, the relative flashiness of the simulated urban hydrographs (Figure 7) means                that streamflow will be highly variable. To support salmon in a stream of this size, a sustained discharge                  of 0.19 m​3​/s would be needed. Since this value is larger than the average discharge for each model                  scenario (excluding C1), stream dimensions would need to be reduced to lower the risk of periods with no                  flow. A more conservative target streamflow of 0.12 m​3​/s was calculated to support stream with               dimensions of 2 m width, 0.3m height, and 0.2 m/s flow. For this scenario, monthly streamflow deficits                 were calculated and are shown in Figure 9.  29    Figure 9.​ Expected daily water volume deficit (megalitres) by month of the flow season. Averages were taken from CN, rational, and C2 discharge scenarios. Flow deficit is based on a target flow rate of 0.12 m​3​/s. The largest water supply deficiencies are expected in April and May, with an average daily requirement of                 3.56 ML/day and 7.47 ML/day respectively. These values were taken from a 10-year climatic average, so                there will likely be years with more and less rain over these months. When considering optimal stream                 dimensions, an active floodplain would likely be required to manage flood risks in November and               December regardless of the chosen width and height, which will increase the area that the stream                development will need. The total quantities of rainfall on the catchment throughout the spawning season               would be able to support the stream, as shown in the first principal analysis, however the flashiness of the                   hydrographs show that the stream would go from high flow (possibly flooding) to low flow very rapidly.                 Mitigation strategies would need to be in place to retain water over time and maintain a steady stream                  flow to facilitate a functional spawning habitat. Flashiness is also exacerbated by a high average slope of                 the catchment, causing flow velocities to potentially reach 4.3 m/s, as determined by the Manning               equation (Eqn. 15). This speed is much greater than the preferred stream velocity of chum salmon, and                 would transport water out of the catchment too quickly to sustain a constant discharge in the preferred                 range. Several possible approaches to mitigate for the combined effects of urban runoff velocity,              catchment slope, low storage volume, and water loss to storm drains are discussed in the               recommendations section below.  Stream Location The most cost-effective and easy-to-engineer stream locations on the Jericho lands would be positioned in               areas of high flow accumulation based on the topography of the area. Figure 10 shows possible stream                 locations derived from a digital elevation model (DEM) of the catchment. The current topography of the                area limits flow from the Jericho uplands into the ponds (light blue) so an additional channel would need                  to be constructed from the ponds to the ocean to facilitate fish entering and exiting the stream. When                  30   planning the urban development of the area, the two flow paths shown in Figures 10A and 10C should be                   considered because they will most easily direct water towards them. However, given the expected flow               speeds of the water, routes will need to be planned to mitigate the incline of these flow paths using                   strategies like step pools, side channels, and planned meandering. Location 2 is only shown as 600m in                 length although it could be extended up to the southern boundary of the park. The locations should be                  chosen considering that location 1 will direct more flow given the topography, yet location 2 is nearer to                  the storm sewer with highest flow rates in the area. A culvert would need to be built beneath W 4th                    Avenue to , regardless of the chosen location. It should be noted that the high flow storm sewers show                   which sewers would direct water based on a DEM of the surface topography, meaning that they are only                  accurate if the underground sewer system is always on the same angle as the surface.   Figure 10.​ Map of the Jericho lands showing routes of highest overland flow accumulation (blue), highest sewer flow accumulation (pink/red), and an existing small stream (green). All flow is northerly following the white contour lines (10 m spacing). Flow accumulation pixel sizes reflect the 20 m resolution DEM that was analysed. 3.3 Groundwater 3.3.1 Quadra Sands Aquifer According to government data from the BC MECC (2011), there is a large aquifer present below most of                  Vancouver and Burnaby (Figure 11). The aquifer is approximately in size and formed out of        3 km7 2       unconsolidated glaciomarine quadra sands (BC MECC, 2011a). Quadra sand is a late Pleistocene             lithostratigraphic unit that is highly distributed throughout the Georgia Depression, British Columbia            31   (Clague, 1976). It mainly consists of horizontally and cross stratified, well sorted sand that is overlain by                 till deposited during the Fraser Glaciation and underlain by fluvial and marine sediments deposited during               the preceding non-glacial interval. A report on the aquifer based on data collected for the Government of                 British Columbia describes the aquifer as being mainly confined, and made of sand and gravel of glacial                 or preglacial origin (BC MECC, 2011a). Depth to water values for the 14 wells range from 17 to 285 ft                    below ground level with a geometric mean of 120 ft. The median and average depth to water is 46 m and                     47 m respectively.    Figure 11.​ Quadra sands aquifer #0049 shown to extend over the development area in Jericho Lands.   Figure 11 shows the aquifer to extent over the development area. Groundwater stored in confined aquifers                cannot recharge stream naturally because they are often too deep and the confining layer largely retains                the water. To supply water to the stream network, the aquifer would have to be accessed by a well drilled                    through the confining layer and the stored water would need to be pumped out using an installed pump                  system. The aquifer, in its entirety, is classified as being moderately productive, moderately vulnerable,              and low in demand (BC MECC, 2011a). Productivity refers to the ability of an aquifer to supply                 groundwater for use (Abesser & Lewis, 2015). It is assessed using observations of well yield, the nature                 of the aquifer material, specific capacity, and transmissivity (Berardinucci & Ronneseth, 2002). Table 8              summarizes the value ranges for the productivity indicators of a moderately productive aquifer             (Berardinucci & Ronneseth, 2002).    32   Table 8.​ ​Value ranges associated with the different indicators of productivity for a moderately productive aquifer. (L/s/m = litres per second per metre) (Berardinucci and Ronneseth, 2002). Indicators of Productivity Value  Nature of Aquifer Material Sand (medium) Well Yield 0.3 – 3.0 L/s  Specific Capacity 0.4 – 4.0 L/s/m  Transmissivity 0.0005 – 0.005 /sm2    The values vary depending on the conditions of the area; therefore, site specific aquifer testing is                necessary to determine the specific availability of groundwater and the rate at which it could be                withdrawn from the aquifer.  The vulnerability of an aquifer refers to how vulnerable it is to contamination from surface sources.                Vulnerability is assessed and interpreted based on type, thickness and extent of geologic sediments              overlying the aquifer, depth to water or depth to top of confined aquifers, and type of aquifer material                  (Berardinucci & Ronneseth, 2002). A potential benefit of low permeability sediments overlying aquifers is that they serve as natural barriers,                making it more difficult for contaminants to get through (Parker et al., 2004). The degree of natural                 protection, however, varies across an aquifer so the thickness of the overlying till layer at Jericho Lands                 should be investigated further to get a better understanding of the vulnerability near the development site.                Contamination from naturally occurring substances in the rock and soil can also threaten the quality of                groundwater, therefore it will be important to test the quality of groundwater in the area before using it as                   source of water for a stream. Finally, being low in demand means that there are currently low levels of groundwater being withdrawn                from the aquifer. A moderate level of development is assigned to aquifers in British Columbia that have a                  low demand but moderate productivity. A moderate level of development indicates that demand is              moderate relative to water availability and additional development of this aquifer should be given careful               consideration (Berardinucci & Ronneseth, 2002). 3.3.2 Surficial Geology The surficial geology of Vancouver is largely composed of unconsolidated sediments deposited by             glaciers and rivers during the Quaternary period (Armstrong, Roots, & Staargaard, 1990). Most of the               glacial deposits in Upland Vancouver are drift associated with the Vashon Glaciation approximately             13,000 to 18,000 years ago. The Vashon drift consists largely of sandy loam lodgement tills that have                 been modified by marine processes, and are thinly overlain in many areas by Capilano sediments               composed of beach sands, marine and non-marine silts and clays, and river sands, with gravel (Rood &                 33   Hamilton, 1994). A map of Metro Vancouver’s surficial geology by Armstrong (1979) illustrates the              surficial geology of Jericho Lands and the surrounding area (Figure 12). Jericho Beach and the Park are                 composed of sand and gravel while the surficial geology of Jericho Lands is largely till, consisting of                 clay, silt, sand, and stone varying in size, up to 25 m thick (​Armstrong & Hicock, ​1979). The eastern side                    of the park, extending upland in the south-east direction, has a surface geology of marine and                glaciomarine silt loam to clay loam with minor sand and silt, normally less than 3 m thick but up to 10 m                      thick in upland areas (​Armstrong & Hicock, ​1979).   Figure 12.​ Map of the surficial geology of Jericho Lands and the surrounding area (​Armstrong & Hicock, ​1979).  Glacial till typically has low hydraulic conductivities, ranging from approximately m/s to m/s         01 −12   01 −6  depending on its composition, while clean sand, a more permeable medium, has hydraulic conductivities              ranging from m/s to m/s (Freeze & Cherry, 1979). This feature supports the assumption of no  01 −6   01 −2              groundwater input in the first principle analysis methodology. The surficial till in the development area is                characterized as being a lodgment till that was deposited from the base of a moving glacier (Rood &                  Hamilton, 1994). Lodgement tills are particularly dense due to compaction, and therefore typically have a               lower porosities, permeabilities, and hydraulic conductivities than other depositional types of till.            Therefore, the surficial till in the area is likely a zone where the flow of water is restricted (i.e., an                    aquitard) and would be incapable of yielding a viable water supply to a production well. The area of                  Capilano sediments towards the eastern end of the Jericho Lands is comprised of fine grained silt and clay                  which have lower hydraulic conductivities and do not make good aquifers either (Freeze & Cherry, 1979).                Table 9 contains lithological data obtained from a well located in the Capilano Sediments surficial               geology area shown in Figure 12, east of Jericho Lands and provides some evidence of the characteristic                 near surface clay (BC MECC, 2011b).  34   Table 9. ​Lithological data for well #71204 east of Jericho Lands at the north west corner of Dunbar and                   8th (BC MECC, 2011b). From (meters) To (meters) Lithology Raw Data 0 0.6 Gravelly Till 0.6 1.2 Loose Gravelly Clay 1.2 3.6 Clay Hardpan 3.6 6.4 Sandstone 6.4 6.7 Salt & Pepper Sandstone Soft 7.6 8.2 Salt & Pepper Sandstone  3.3.3 Subsurface Geology  Underlying the surficial till is the Quadra Sand unit comprised of sand and gravel, followed beneath by                 fluvial and marine sediments and bedrock. The bedrock that underlies Jericho Lands and the majority of                Metro Vancouver is sedimentary conglomerate, sandstone, shale with thin lignite; lesser basalt flows, sills              and minor pyroclastics (Ministry of Energy, 2014). A report on risk management and settlement control               for urban tunneling under Vancouver investigated the surficial and sub surficial geology in the area and                found similar compositions (Ciamei & Moccichino, 2009). The bored tunnels extend from Waterfront             skytrain station to Olympic Village station, east of Jericho Lands. The soils in the geological report are                 described as being glacial tills with minor non-till like soils and some interspersed granitic boulders. The                underlying rocks are reported as being sedimentary, composed mostly of sandstone (~75%), siltstone and              claystone. A report on the geology of Vancouver by Armstrong et al. (1990) indicates that most of the                  area between Kitsilano and Point Grey beaches is underlain by brown sandstone and mudstone (shale).               Sandstone is a porous and more permeable bedrock formation compared to others, such as granite or                volcanic bedrock, and is therefore the most common aquifer type, but compaction can potentially reduce               much of the pore space (Hale & McDonnell, 2016). Shale is much less permeable, and therefore usually                 must be fractured or cracked to be an aquifer, which is less common. There is, however, no information                  indicating the presence of a bedrock aquifer near Jericho Lands, therefore a well will need to be drilled to                   find the depth to the bedrock, and to determine its capacity to bear and transmit water. Figure 13 shows                   the relative positions of the geological layers.      35    Figure 13​. Cross-section of surficial and sub surficial geology through the center of Jericho Lands and Jericho Park and Beach. This cross-section is purely interpretive. 4. Conclusions and Recommendations 4.1 Water Quality In general, the measured temperature and EC values were within the optimal range for supporting chum                salmon as recommended by the Government of British Columbia (BC MECC, 2018; 2015). The low EC                values measured at all sites indicate that no significant pollution source is contaminating the stormwater               in this region. Since water temperatures were not recorded over the total range of chum salmon season, it                  is likely that maximum and average stream temperatures over this period would be higher than measured                in this study. In order to minimize temperature impacts on salmon, we propose a few recommendations                for retaining lower stream temperatures. In general incoming solar radiation and air temperature are              recognized as the major sources of thermal energy for streams (Johnson, 2004). Research has found that                shading decreases maximum temperatures both by inhibiting solar radiation from reaching the stream and              increasing convective heat exchanges which decrease air temperature (Johnson, 2004). Therefore, it is             recommended that a future stream be surrounded with vegetation which promotes maximum shading of              the streambed. Another factor which has been shown to significantly influence stream temperatures is the               substrate type. Research has found alluvial type substrates to be correlated with lower fluctuations in               diurnal and seasonal temperature patterns and lower stream temperatures than bedrock substrates. Thus,             land developers may want to consider using alluvial type substrates during stream construction in order to                keep stream temperatures lower and more stable on a diurnal and annual basis.  36   Total metal concentrations were observed to vary between sampling sites. In general, water sampled from               the FC site contained the highest concentrations of metals. Concentrations of Al, As, Fe, Mn, and Zn were                  in exceedance of British Columbia’s short-term WQG for supporting freshwater aquatic life at this site. In                order to make confident conclusions regarding the suitability of the stormwater sampled for supporting              chum salmon spawning and survival, it is recommended that more water sampling take place. Obtaining               sufficient water quality data to compare with the long-term WQG is likely a good approach as this may                  provide a more valuable measure of the ​overall water quality of these sources and would minimize                potential unusual spikes in contaminant concentrations. Additionally, only dissolved metals are accounted            for in the WQG for Al and Cd, thus for these metals, it may be valuable to perform a dissolved metals                     analysis in addition to a total metals analysis to ensure that only potentially toxic forms are being                 considered in comparison with the guidelines.  4.2 Water Availability In summary, first principle analysis assumed no groundwater exchange (input and output) in the Jericho               catchment and in an annual scale the change in catchment water storage is zero. The analysis also                 assumed that soil is at saturation therefore the difference between precipitation and evapotranspiration             leaves the catchment as discharge. First principle analysis found that at the catchment outlet there is                enough water to support a small salmon spawning stream residing in the lower region of the preferential                 range for Chum salmon. However during dry years such as 2013, the stream will be too shallow to                  support salmon species due to water shortage. Assuming climate change does not further affect              precipitation regime in Vancouver, an average channel width ranging from 0.6m to 6.7m can be expected                while assuming a rectangular stream channel, constant streamflow of 0.2 m/s and constant stream depth of                18.5cm are both preferential for chum salmon ​(Geist et al, 2002)​. Additionally, considering the historical               average level of discharge, a streamflow of 0.15 m/s can be expected while assuming a rectangular stream                 channel, constant stream depth of 30cm and channel width of 2m. These metrics are calculated at the                 catchment outlet point and additional analysis is needed to acquire a streamflow profile with regard to                distance upstream.   Computer modelling of the catchment shows that an optimal discharge at the outlet of the stream could                 only be supported solely by rainfall during December and possibly November given the current              hydrological and climatic characteristics. Our results suggest that average discharge could range from             0.084-0.173 m​3​/s over the flow season (Nov-June) however, flow rates fluctuate rapidly in response to               precipitation events, meaning that sustained discharges within this range are unlikely to occur without              careful planning. We can conclude that several major issues must be addressed when planning for the                stream:  1. Urbanization planned for the Jericho lands will increase evaporation, runoff speeds, and increase             proportion of precipitation discharged as quickflow. 2. Stormwater must be diverted from the upper areas of the catchment to the headwaters in order to                 generate the required water supply. 3. Holding tanks with a controlled discharge will likely be needed to account for the lack of soil                 water storage and maintain a steady stream flow. 37   4. The stream will need to meander across the landscape to decrease flow velocity and increase the                routing reservoir storage capacity.   In terms of maximizing available water available for the stream, it is crucial that the tasks set out by the                    integrated stormwater management plan (ISMP), proposed by the City of Vancouver, are completed. The              main focus of ISMP is to separate stormwater from the combined sewage pipe; if this water is made                  available to the stream via stormwater pipes it would enable the use of much more rainwater which was                  previously non-viable (​City of Vancouver, 2016​). The total water supply can also be increased if               evaporation is limited by allowing water to infiltrate into a medium as opposed to be trapped on the                  surface and exposed to the atmosphere. Evaporation will consequently increase with further urbanization             of the catchment, so development strategies such as absorbent landscapes, infiltration swales, rain             gardens, pervious paving, and infiltration trenches (​City of Vancouver, 2016​) will need to be              implemented. Additional measures include highly reflective roofing material will increase the albedo of             the catchment, thereby decreasing the surface temperature and energy available for evaporation to take              place. To increase base-flow of the stream (and reduce flashiness), strategies such as infiltration bulges,               green roofs, tree well structures, rainwater harvesting, and detention tanks will be needed (​City of               Vancouver, 2016​). Flooding events should be expected given Vancouver’s climatic norms so ample space              for a floodplain with a functional riparian area would be recommended. 4.3 Groundwater Geological information from literature and well data indicate that the surficial geology of Jericho Lands               and the upland area is largely composed of glacial till with areas of silt and clay towards the east. These                    materials have low hydraulic conductivities and are indicative of a near surface confining layer below               Jericho Lands and the watershed area. Underlying the surficial geology is sand and gravel which               comprises the extensive glaciomarine Quadra Sands aquifer system. The confining layer prevents direct             surface to aquifer connectivity, therefore water from this source will only be accessible using a well and                 pump system. Below the aquifer, the bedrock is composed of sandstone and lesser shale.  The large size of the aquifer and its classification as being moderately productive and in low demand                 suggests that it could support withdrawals. Site specific testing, however, should be done to determine the                actual viability of accessing and withdrawing groundwater in the area as the groundwater conditions will               not be constant throughout the entirety of the aquifer. Wells, for instance, should be installed at the                 development site to enable direct aquifer testing and the collection of more precise lithological              information throughout the entire depth. Aquifer testing should be done to determine the transmissivity,              hydraulic conductivity, water quality and storativity characters of aquifers that ultimately determine its             ability to supply usable water. It is also important to test the groundwater quality to confirm that it is                   within a safe range for supporting chum salmon as human contaminants and natural metals can degrade                water quality. Finally, while there is no indication that the bedrock in the area is an aquifer, ​t​he higher                   porosity and permeability of sandstone bedrock make it a common aquifer type. Similar to the confined                aquifer, pump testing would be required to quantify its yield potential. However, because the bedrock has                not already been mapped as an aquifer in the area, it is unlikely to be one. Therefore, given the high cost                     of drilling and lack of information, it would likely not be worth it. 38   4.4 Conclusions and Future Considerations In conclusion, the introduction of an ecologically functional stream that can support chum salmon is still                uncertain and greatly depends on how construction and implementation of the ISMP will affect water               availability and water quality. Since future developments will change the hydrological characteristics of             the area, we recommend to perform a similar water balance approach and water quality analysis               post-construction. It is possible to reintroduce a chum salmon population to the lands with careful               planning throughout the development phase and a long term management plan to upkeep the stability of                the ecosystem. Several other unknowns that should be considered during the design process are listed               below.    1. Designing a self-sufficient aquatic ecosystem that will require minimal human maintenance. 2. A holding tank at the headwaters could be designed as a water feature and should be integrated                 into the urban planning as early as possible. 3. Possible stream channel morphologies must be carefully considered to create a stream that             optimizes both flow speeds and area required so that it can be well integrated into the dense urban                  environment. 4. Streambed surface is important for spawning site selection so an appropriate substrate must be              selected to facilitate chum salmon. 5. Efforts will be needed to select beneficial native plant species to integrate into the stream/riparian               zone to reduce flow speeds, contribute to water filtration, and support biodiversity. 6. Sunlight exposure of the stream could be optimized to support plant life while simultaneously              limiting evaporation. 7. Sediment load in the stream from construction will be a concern for the survivability of fry. A                 management plan to mitigate for this will be needed. 8. Predatory behavior of house cats and raccoons on the spawning salmon is expected and could               negatively influence reproduction rates of salmon. 9. Other freshwater fish species could be considered to inhabit the stream, especially if it will not be                 possible to connect the stream channel to the ocean given planning restrictions and sea-level rise. 10. Floodplain characteristics will need to be considered given the large fluctuations of discharge             which are likely to occur. Features like high embankments, and highly permeable areas             immediately surrounding the stream are recommended.   39   References Abesser, C., & Lewis, M. (2015). A semi-quantitative technique for mapping potential aquifer productivity on the national scale: Example of england and wales (UK). Hydrogeology Journal, 23(8), 1677-1694. doi:10.1007/s10040-015-1295-5 Armstrong, J.E. & Hicock, S.R. (​1979), Geological Survey of Canada, "A" Series Map 1486A, 1 sheet,​  (Open Access) Armstrong, J.E., Roots, C. F., & Staargaard, C. F. (1990). Vancouver geology (3rd, expand ed.). Vancouver, B.C: Geological Association of Canada, Cordilleran Section. Berardinucci, J. and Ronneseth, K. (2002). Guide to Using the Aquifer Classification Maps for the Protection and Management of Groundwater. BC Ministry of Water, Land and Air Protection Water, Air and Climate Change Branch. Retrieved from:  Blessent, D., Therrien, R., & Gable, C. W. (2011). Large-scale numerical simulation of groundwater flow and solute transport in discretely-fractured crystalline bedrock. Advances in Water Resources, 34(12), 1539-1552. doi:10.1016/j.advwatres.2011.09.008 Brett J.R. (1951). Temperature Tolerance in Young Pacific Salmon Genus Oncorhynchus. Journal of the Fisheries Research Board of Canada, 9: 265-309. British Columbia Ministry of Environment and Climate Change Strategy (BC MECC). (2015). Water Quality Assessment and Objectives for Coldstream Creek - Technical Report. Retrieved from: wqos/thomp-ok-wqos/technical_coldstream_wqo_jan_2016.pdf  British Columbia Ministry of Environment and Climate Change Strategy (BC MECC) - Water Protection and Sustainability Branch. (2018). British Columbia Approved Water Quality Guidelines: Aquatic Life, Wildlife & Agriculture. Retrieved from: wqos/approved-wqgs/wqg_summary_aquaticlife_wildlife_agri.pdf  British Columbia Ministry of Environment and Climate Change Strategy (BC MECC) - Water Protection and Sustainability Branch. (2011a). Groundwater Aquifers. Retrieved from:  British Columbia Ministry of Environment and Climate Change Strategy (BC MECC) - Water Protection and Sustainability Branch. (2011b). Groundwater Wells. Retrieved from:  40   CBC News. (2013, March 16). ‘Pacific salmon named B.C.’s official fish. CBC News. Retrieved from  Christen, A. (n.d.). Vancouver-Sunset. Retrieved from City of Vancouver. (2016). Citywide Integrated Rainwater Management Plan. Vol 2. City of Vancouver Planning Services (CVPS). (2002). Still Creek Rehabilitation and Enhancement Study. Clague, J. J. (1976). Quadra sand and its relation to the late wisconsin glaciation of southwest british columbia.​ Canadian Journal of Earth Sciences, 13​(6), 803-815. doi:10.1139/e76-083 Das, R.,  Samal, N.R.,  Roy, P.K.,  Mitra, D. (2005). Role of electrical conductivity as indicator of pollution in shallow lakes. Asian Journal of Water Environment and Pollution, 3(1), 143-146. De Silva, Margaret. “A Closer Look at the Jericho Lands Development.” Vancouver Magazine, 18 Oct. 2018, ​​. Freeze, R. A., & Cherry, J. A. (1979). ​groundwater​. Englewood Cliffs, N.J: Prentice-Hall. Geist, D. R., Hanrahan, T. P., Arntzen, E. V., McMichael, G. A., Murray, C. J., & Chien, Y. J. (2002). Physicochemical characteristics of the hyporheic zone affect redd site selection by chum salmon and fall Chinook salmon in the Columbia River. North American Journal of Fisheries Management, 22(4), 1077-1085. DOI:;2 Göbel, P., Dierkes, C., Coldewey, W. G. (2007). Storm water runoff concentration matrix for urban areas. Journal of Contaminant Hydrology, 91, 26-42. DOI:​ ​10.1016/j.jconhyd.2006.08.008 Google Maps (2019). Map of Vancouver and Jericho Park and Lands. Retrieved from,-123.2018971,13z Groot C and Margolis L. 1991. Pacific Salmon Life Histories. UBC Press: Vancouver. Hale, V. C., & McDonnell, J. J. (2016). Effect of bedrock permeability on stream base flow mean transit time scaling relations: 1. A multiscale catchment intercomparison. Water Resources Research, 52(2), 1358-1374. doi:10.1002/2014WR016124. Isaak, D. J., Thurow, R. F., Rieman, B. E., & Dunham, J. B. (2007). Chinook salmon use of spawning patches: Relative roles of habitat quality, size, and connectivity. Ecological Applications, 17(2), 352-364. doi:10.1890/05-1949 Kondolf GM and Wolman MG. (1993). "The sizes of salmonid spawning gravels." Water Resources Research 29(7): 2275-2285. 41   Jensen, D. W., Steel, E. A., Fullerton, A. H., & Pess, G. R. (2009). Impact of fine sediment on egg-to-fry survival of Pacific salmon: a meta-analysis of published studies. Reviews in Fisheries Science, 17(3), 348-359. DOI: ​ Johnson, S.L. (2004). Factors influencing stream temperatures in small streams: substrate effects and shading experiment. Canadian Journal of Fisheries and Aquatic Sciences, 61, 913-923. DOI: 10.1139/F04-040.  Li, L., Stewart, L., Yuan, D., & Wen, C. (2018). Restoring a Salmon Spawning Stream to the Jericho Watershed. Luttmerding, H.A. 1984. Soils of the Langley-Vancouver Map Area: Report 15, British Columbia Soil Survey volume 6. Province of British Columbia, Ministry of Environment, Assessment and Planning Division, Victoria, BC. 345 pp. McCuen, R. H. (1989). Hydrologic analysis and design (pp. 143-147). Englewood Cliffs, NJ: Prentice-Hall. McKean, C.J., Nagpal, N.K. (1991). Ambient Water Quality Criteria for pH - Technical Appendix. Ministry of Environment, Province of British Columbia.. Retrieved from: Meyer, J. L., Paul, M. J., & Taulbee, W. K. (2005). Stream ecosystem function in urbanizing landscapes. Journal of the North American Benthological Society, 24(3), 602-612. DOI: Mielke, H.W., Powell, E.T., Shah, A., Gonzales, C.R., Mielke, P.W. (2001). Multiple metal contamination from house paints: consequences of power sanding and paint scraping in New Orleans. Environmental Health Perspectives, 109(9): 973-978. DOI: ​10.1289/ehp.01109973 Oke, T. (2017). ​Urban climates​. Cambridge, United Kingdom: Cambridge University Press. Oudin, L., Hervieu, F., Michel, C., Perrin, C., Andréassian, V., Anctil, F., & Loumagne, C. (2005). Which potential evapotranspiration input for a lumped rainfall–runoff model?: Part 2—Towards a simple and efficient potential evapotranspiration model for rainfall–runoff modelling. Journal of hydrology, 303(1-4), 290-306. Pagano, T., Hapuarachchi, P., & Wang, Q. J. (2010). Continuous rainfall-runoff model comparison and short-term daily streamflow forecast skill evaluation. CSIRO Tech. Rep. EP103545, 70 pp., CSIRO, Australia. Page, N., P. Lilley, D. Matsubara and N. Tokgoz (2013). Hastings Creek Watershed: Ecology and Hydrotechnical Assessment. Unpublished report for District of North Vancouver prepared by Raincoast Applied Ecology and Kerr Wood Leidal Associates. 111 pp + appendices. 42   Parker, B. L., Cherry, J. A., & Chapman, S. W. (2004). Field study of TCE diffusion profiles below DNAPL to assess aquitard integrity. Journal of Contaminant Hydrology, 74(1-4), 197-230. doi:10.1016/j.jconhyd.2004.02.011  Perrin, C., Michel, C., & Andréassian, V. (2001). Does a large number of parameters enhance model performance? Comparative assessment of common catchment model structures on 429 catchments. Journal of Hydrology, 242(3-4), 275-301. Price, M.H. (2013). Sub-lethal metal toxicity effects on salmonids: a review. Report prepared for SkeenaWild Conservation Trust. Smithers, BC. 64 pages. Available at: Rainforest Applied Ecology. (2007.) Spanish Bank Creek Restoration Project Project Summary [Brochure]. Richter, A., & Kolmes, S. A. (2005). Maximum temperature limits for Chinook, coho, and chum salmon, and steelhead trout in the Pacific Northwest. Reviews in Fisheries Science, 13(1), 23-49. DOI: Rood, K.M. and R.E. Hamilton. (1994). Hydrology and water use for salmon streams in the Fraser Delta Habitat Management Area, British Columbia. Can.Manuscr. Rep. Fish. Aquat. Sci. 2038: 187 p. Single crop coefficient (Kc). (n.d.). Retrieved March 11, 2019, from  Vancouver Park Board. (January 2018.) Tatlow and Volunteer Park stream restoration information display open house [Brochure]. Vancouver Soil Map: Langley-Cloverdale Soil Management Group. (n.d.). Retrieved March 11, 2019, from ​ Walsh, C. J., Roy, A. H., Feminella, J. W., Cottingham, P. D., Groffman, P. M., & Morgan, R. P. (2005). The urban stream syndrome: current knowledge and the search for a cure. Journal of the North American Benthological Society, 24(3), 706-723. DOI:  Walters, D. M., Leigh, D. S., & Bearden, A. B. (2003). Urbanization, sedimentation, and the homogenization of fish assemblages in the Etowah River Basin, USA. In The Interactions between Sediments and Water (pp. 5-10). Springer, Dordrecht. DOI: Wang, K., Wang, P., Li, Z., Cribb, M., & Sparrow, M. (2007). A simple method to estimate actual evapotranspiration from a combination of net radiation, vegetation index, and temperature. Journal of Geophysical Research: Atmospheres​, ​112​(D15).  43   Appendix  Table A1.​ ​A list of assumptions for our modelling approach that are not stated in text. Model Assumptions 1. Rainfall is distributed evenly over the entire catchment area. 2. All precipitation that falls on the combined catchment areas can be directed to a single outlet. 3. The watershed system is linear, meaning that discharge will increase and decrease linearly in response to precipitation events. 4. The maximum flow rate occurs when water is contributed to the outlet from the entire catchment. 5. Stream dimensions are uniform from headwaters to outlet.  Table A2.​ ​British Columbia Ministry of Environment & Climate Change Strategy (2018) water quality guideline metal concentrations  for freshwater aquatic ecosystems.  Metal BC MECC WQG (mg/L) Form Guideline Type Aluminum (Al) *** 0.1 Dissolved Al Short Term Maximum Arsenic (As) *** 0.005 Total As Short Term Maximum Boron (B) 1.2 Total B Not specified Cadmium (Cd) (varies with hardness) FC1 – 0.00070 FC2 – 0.00059 FC3 – 0.00059 SL1 –0.00044 SL2 –0.00047 SL3 –0.00043 SM1 –0.00043 SM2 –0.00050 SM3 – 0.00046 SU1 – 0.00046 SU 2 –0.00051 SU 3 – 0.00050 Dissolved Cd Short Term Maximum Cobalt (Co) 0.11 Total Co Not Specified 44   Copper (Cu) (varies with hardness) FC1 – 0.11 FC2 – 0.097 FC3 – 0.096 SL1 –0.072 SL2 –0.077 SL3 –0.071 SM1 –0.071 SM2 –0.083 SM3 – 0.077 SU1 – 0.076 SU 2 –0.084 SU 3 –0.083 Total Cu Short Term Maximum Iron *** 1 Total Fe Short Term Maximum Lead (Pb) (varies with hardness) FC1 – 0.102 FC2 – 0.083 FC3 – 0.082 SL1 –0.056 SL2 –0.061 SL3 –0.056 SM1 –0.055 SM2 –0.067 SM3 – 0.061 SU1 – 0.060 SU 2 –0.069 SU 3 –0.067 Total Pb Short Term Maximum Manganese (Mn) (Varies with hardness) FC1 – 1.85 FC2 – 1.65 FC3 – 1.64 SL1 –1.36 SL2 –1.42 SL3 –1.35 SM1 –1.35 SM2 –1.49 SM3 – 1.41 SU1 – 1.41 SU 2 –1.50 SU 3 –1.49 Total Mn Short Term Maximum Molybdenum (Mo) 2 Total Mo Short Term Maximum 45   Selenium (Se) 0.002 Total Se Long Term Average (ST not available) Silver (Ag) 0.0001 Total Ag Short Term Maximum Zinc (Zn) *** (varies with Hardness)   FC1 – 0.055 FC2 – 0.041 FC3 – 0.041 SL1 – 0.033 SL2 – 0.033 SL3 – 0.033 SM1 – 0.033 SM2 – 0.033 SM3 – 0.033 SU1 – 0.033 SU 2 – 0.033 SU 3 – 0.033 Total Zn Short Term *​WQG for Cd, Cu, Pb, Mn, and Zn based on site hardness measurements *WQG for Al assumes water pH ≥ 6.5. For pH < 6.5, the Al WQG varies with pH by the following equation: dissolved Al = e​(1.209 - 2.426 (pH) + 0.286 K)​, where K = (pH)​2​.  Table A3.​ ​Stormwater total metals concentrations and hardness reported by site.  Metal Sample No. Central Forest (mg/L) Lower Stream (mg/L) Mid-stream (mg/L) Upper Stream (mg/L) Exceeds Guideline? Hardness [CaCO​3​] 1 119 74.8 73.1 79 NA 2 101 80.2 86.1 87.4   3 100 73.9 79.5 86.1   Aluminum 1 2 3 9.15 4.52 2.99 0.0357 0.0561 0.0437 0.355 0.0525 0.05 0.167 0.0862 0.0495 Y Arsenic   1 0.00659 0.00017 0.00032 0.00023 Y 2 0.000367 0.00018 0.00018 0.0002   3 0.00258 0.00016 0.00019 0.00018   Boron 1 0.016 0.015 0.015 0.014 N 46     2 0.014 0.015 0.014 0.013     3 0.015 0.015 0.015 0.015   Cadmium 1 0.000489 0.000016 0.000089 0.0000435 N   2 0.000182 0.000109 0.000084 0.000146     3 0.000127 0.000026 0.000036 0.0000439   Cobalt 1 0.00821 <0.00010 0.00038 0.00037 N   2 0.00353 <0.00010 0.00016 0.00024     3 0.00244 <0.00010 0.00013 0.00021   Copper 1 0.0232 0.00277 0.00663 0.00439 N   2 0.0109 0.00444 0.00485 0.00543     3 0.00744 0.0035 0.00398 0.0037   Iron 1 10.7 0.144 0.836 0.581 Y   2 5.32 0.201 0.277 0.451     3 3.46 0.171 0.258 0.392   Lead 1 0.0387 0.000259 0.00191 0.00122 N   2 0.0166 0.000366 0.000381 0.00176     3 0.0118 0.000326 0.000453 0.000654   Manganese 1 2.84 0.00654 0.0582 0.051     2 0.951 0.0148 0.0327 0.0477 Y   3 0.675 0.00805 0.0255 0.0405   Molybdenum 1 0.00063 0.00181 0.00192 0.00266 N   2 0.000453 0.00165 0.00193 0.00213     3 0.000404 0.00173 0.00207 0.00248   Selenium 1 0.00021 <0.00005 0.000081 0.000055 N 47     2 0.000177 0.000064 <0.00005 0.000052     3 0.000111 0.000055 0.000058 0.000072   Silver 1 0.000078 <0.00001 <0.00001 <0.00001 N   2 0.000042 <0.00001 <0.00001 <0.00001     3 0.000025 <0.00001 <0.00001 <0.00001   Zinc 1 0.126 0.004 0.0128 0.0084 Y 2 0.0538 0.0068 0.0083 0.0106   3 0.0372 0.0052 0.0076 0.0077    48 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items