Open Collections

UBC Graduate Research

Analysis of Campus Water Base Flows Using Metering Data Crozier, Aina 2018-09-06

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


66428-Crozier_Aina_CIVL_592F_Analysis_campus_FinalReport.pdf [ 3.02MB ]
JSON: 66428-1.0386033.json
JSON-LD: 66428-1.0386033-ld.json
RDF/XML (Pretty): 66428-1.0386033-rdf.xml
RDF/JSON: 66428-1.0386033-rdf.json
Turtle: 66428-1.0386033-turtle.txt
N-Triples: 66428-1.0386033-rdf-ntriples.txt
Original Record: 66428-1.0386033-source.json
Full Text

Full Text

UBC Social Ecological Economic Development Studies (SEEDS) Sustainability Program Student Research Report        Analysis of Campus Water Base Flows Using Metering Data Aina Crozier University of British Columbia CIVL 592F Themes: Water, Buildings September 6, 2018         Disclaimer: “UBC SEEDS Sustainability Program provides students with the opportunity to share the findings of their studies, as well as their opinions, conclusions and recommendations with the UBC community. The reader should bear in mind that this is a student research project/report and is not an official document of UBC. Furthermore, readers should bear in mind that these reports may not reflect the current status of activities at UBC. We urge you to contact the research persons mentioned in a report or the SEEDS Sustainability Program representative about the current status of the subject matter of a project/report”. ANALYSIS OF CAMPUS WATER BASE FLOWS USING METERING DATAUBC SEEDS Directed Studies Research ReportBy Aina CrozierSept. 6th, 2018 AINA CROZIERUniversity of British ColumbiaPhD Student in Civil EngineeringResearch focuses on tools to help inform and assist decision making in urban water system planning and management.Norwegian University of Science and TechnologyMSc Energy and Environmental EngineeringCo-supervised by:Dr. Martino Tran Dr. Steven WeijsDirector, Urban Predictive Analytics LabAssistant Professor, SCARPAssociate Faculty, Civil EngineeringDirector, Hydro-Info-Theory LabAssistant Professor, Civil EngineeringSeptember 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 2EXECUTIVE SUMMARYWater consumption in all buildings equipped with high-resolution meters has been analyzed in detail in this study, with focus on their contribution to the high campus base flow (minimum) consumption, estimated at 50 % of total annual consumption. A systematic approach has been applied to estimate the base flow by assuming that whatever is consuming water when occupancy is minimal during the non-irrigation season is running continuously. Buildings have been identified and ranked accordingly. Since data was not easily acquired and there has been limited previous analysis, a lot of the time was spent on organizing, cleaning and exploring the data, as well as gaining confidence by discussing findings with stakeholders. The 55 primary buildings for which good quality data was available for the study period from July 2017 to July 2018 (the majority Core buildings), consume 24 % of the Campus total supply and 35 % of the Campus base flows. This translated to current high-resolution metered base flow consumption of ~660,000 cubic meters annually, or potential annual cost savings of almost $600,000 (17 % of total water costs) if eliminated completely. It was also found that, as expected, Core buildings with Lab as primary usage have the highest water usage intensities and contribute more to campus base flows. Further, it is stated that more than 70% of the Campus water consumption is currently not accounted for in high resolution (Skyspark data).Recommendations from this study are to investigate whether the already high-resolution metered base flow water consumption in buildings can be reduced, by, first, clearing up any confusions regarding where meters are located and the area and purpose they cover, and, second, evaluating the cost and feasibility of conserving the water. Further, the recommendation is to make sure that already installed high-resolution meters are working properly and connected, and then prioritize additional meters in buildings with significant lab usage, as well as separate irrigation systems to help investigate potential leaks. Additionally, a number of recommendations related to improving data availability and quality are listed in this report, as the overall impression is that there should be substantial opportunity for water conservation in buildings and other systems if reliable data is available. As a contribution to this goal,the report is accompanied by an Excel spreadsheet containing all numerical and descriptive data that have been gathered in this study, including various statistics, metrics and charts to help provide a deeper understanding at the building level. September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 3TABLE OF CONTENTS1. Introduction1. Background & Previous Work 2. Status & Objectives2. Approach1. Challenges & Time Spent2. Data Availability: Campus Supply & Building Consumption 3. Data Quality: Uncertainties & Sources of Error4. Technical & Analytical Methods3. Results1. Spreadsheet with Metadata, Statistics and Metrics2. Building Consumption, Rankings & Benchmarks3. Nighttime Consumption & Estimated Base Flows4. Example: Forestry Sciences4. Conclusions1. Findings Summary2. Recommendations3. Potential Projects5. References & Resources6. Appendices: Spreadsheet & Technical WorkSeptember 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 4BACKGROUNDSeptember 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 5Bud Fraser Senior Planning & Sustainability Engineer“Campus supply water meter data indicates base (minimum) water flows that approach 50% of average flows. This indicates that there are buildings or systems that are consuming large amounts of water during the night when academic building occupancy is at a minimum.”IntroductionPREVIOUS WORKUBC Water Base Flow Analysis (2011) Stantec Consulting engaged by Campus Sustainability Office (CSO) Defined baseflow as the lowest and most consistent water flow for a building at any one point during the study period (Sept 2010 to Sept 2011). Extensive study covering data from campus main supplies and sewer, buildings with digital meters (78) and analog meters (298), analog irrigation meters, and various manual readings.September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 6Total and Base Water Consumption of the Entire Campus (Stantec Consulting) Breakdown of the Campus Annual Water Consumption by Meter Type (Stantec Consulting) IntroductionSTATUSEstimated base flow percentage of annual consumption is still around 50%. Water is relatively cheap compared to electricity, which explains why efforts to reduce consumption may have been down-prioritized.  UBC Water Action Plan (2018) – Renewed sustainability effort to conserve water.“Maintain UBC total water consumption at or below 2017 levels despite anticipated growth of 2%”Makes sense to investigate sources of base flow (low-hanging fruit). Price of water per cubic meter: $0.891(Increasing 5% annually)20112018Water Consumption (m3)Estimated Baseflow (m3)Baseflow/Annual Consumption (%)16 Avenue Meter 404,000*970,895 315,000*403,103 78%*42%U Blvd Meter 3,500,000*2,799,096 1,730,000*1,483,91450%*53%Total 3,900,000*3,769,991 2,050,000*1,887,01753%*50%$1,298,095 (2011) $1,681,332 (2018)September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 7Cost of base flow:*Stantec Consulting: Total and Base Water Consumption of the Entire Campus IntroductionOBJECTIVES1. Gain a deeper understanding of water consumption across campus buildings by analyzing Skyspark high-resolution data along with data from campus main supply meters.o Identifying and ranking buildings with the largest contribution to campus base flow rates.o Identifying potential sources of the base flows.2. Estimate how much of the total water flow within core academic buildings on campus is metered and if possible, provide recommendations for additional metering locations.September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 8IntroductionThese evolved from the various actions listed in the new UBC Water Action Plan.CHALLENGES & TIME SPENTSystematic approach instead of reading off base flow graphically: Easier to reproduce and update numbers. Apply analytical technique to get deeper understanding of characteristics and trends for building consumption.Acquiring, exploring and cleaning data (gaining confidence) took longer than anticipated and triggered many questions.Data has been organized in single spreadsheet, providing a good baseline for further analysis.Summary of data availability and quality, methods, results and recommendations are presented in the following slides.1.Literature review2.Acquiring data3.Exploring and cleaning data4.Stakeholder discussions5.Extracting informationTIME SPENTSeptember 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 9ApproachDATA AVAILABILITYTime period: July 1st 2017 – July 1st 2018 Steam boiler in Power House shut down on June 7th 2017 Campus Water Supply:  No direct access to data (TrendReader) 15 min resolution data  U Blvd and 16th Ave supplies (high- and low-pressure zones)  Available as average flow rates (liters per second) Consumption readings Sept 27th, Dec 20th and March 28th(cubic feet for 16th Ave and cubic meters for U Blvd)Building Water Consumption: Access to Skyspark platform 15 min resolution data 80 buildings listed with water consumption data Available as cumulative consumption (cubic meters)September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 10ApproachCampus Water Schematics indicating Main Supplies, Distribution Pipes, Pressure Zones, PRVs, etc. (EWS)DATA QUALITY: CAMPUS SUPPLYSeptember 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 11Interpretation challenging due to automated interaction between high and low-pressure zones through seven pressure-reducing values (PRVs).Uncertainties & Sources of Error: When and how much of U Blvd supply feeds low-pressure zone. Offset between consumption computed from flow rate measurements and manual readings. Conversion between metric and imperial units.See Appendix 2: Technical Work for more details on how these were handled.ApproachDATA QUALITY: BUILDING CONSUMPTIONMain challenge is understanding where data is coming from and how it has been acquired and treated before being presented in Skyspark.Uncertainties & Sources of Error : Understanding which meter the data is coming from.  Ensuring correct area and space usage of buildings due to sub-fed buildings.  Missing data and confidence factors (communication issues water meter/ION meter/server). Interpreting and handling consumption spikes (cumulative consumption or calculation error). Interpreting zeros (malfunctioning meter or accounted for in following datapoints).See Appendix 2: Technical Work for more details on how these were handled.September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 12ApproachSample of Map indicating Campus Water Distribution Pipes, Buildings and Water Meters. (EWS)TECHNICAL & ANALYTICAL METHODS1. Organized data in single Excel spreadsheet1.2. Included metadata by cross-referencing different sources.3. Cleaned and updated data 2:• Building usage and total areas adjusted for meters covering sub-fed buildings • Consumption spikes after missing data removed from the dataset• Confidence factors and zero percentages used to filter out poor quality data 4. Computed statistics and metrics to get building rankingBase flow consumption estimate found by extracting nighttime consumption in non-irrigation months and scaling up to the whole year.Assuming that whatever is consuming water when occupancy is minimal in non-irrigation season, is running continuously.Map of campus distribution pipes, building names, numbers and water metersMap of campus supply and pressure zonesCCP building list with names and numbers Skyspark building namesStatistics, metrics and rankingsMetadataBuilding consumption dataCampus supply dataSeptember 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 13Approach1 See Appendix 1: Spreadsheet for more details.2 See Appendix 2: Technical Work for more details.SUMMARY SHEET WITH METADATA, STATS AND METRICSSeptember 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 14ResultsBUILDING CONSUMPTION ANALYSIS SHEETSeptember 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 15Results80 Buildings 55 Good quality47 High pressure zone (U Blvd)43 Core (18 sub-fed)27 Lab16 Other Primary Usage4 Ancillary1 Residential3 Other Primary Usage8 Low pressure zone (Both supplies)3 Core1 Lab3 Other Primary Usage5 Ancillary(2 sub-fed)5 Residential25 Poor quality10 Mech fault6 Comm fault1 Elec fault8 Syst faultSKYSPARK BUILDING DATA▪ 55 primary buildings with high-resolution data of good quality in the study period from July 2017 to July 2018 (25 deemed poor quality).▪ The majority are Core buildings located in the High pressure zone, consuming water from the U Blvd campus supply only.▪ 64 of the 218 core buildings on campus are metered with high resolution when including sub-fed buildings into the count.September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 16ResultsTOP 10 WATER CONSUMERSHigh-resolution metered buildings account for 24 % of total campus supply and Top 10 buildings account for 14%.If only considering buildings located in the high-pressure zone (47 buildings), these account for 29% of the U Blvd supply.September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 17m3/yrResultsMetered Core buildings account for 18 % of total campus supply and Top 10 Core buildings account for 11%.If only considering core buildings with lab as primary usage (28 buildings), these account for 15 % of total campus supply.m3/yrAdded to the list when only considering core buildings.More than 70% of campus consumption is not accounted for in high resolution (Skyspark).WUI BENCHMARK▪Water Usage Intensity (WUI) is defined as annual consumption divided by total area (m3/yr/m2).▪Biological Sciences Building South stands out even more as Orchard Commons drops to the bottom.▪Chemistry South, CEME, Biomedical Research Centre and Frank Forward (all Lab buildings) replace other buildings.▪Buildings with Lab as primary usage have highest WUI.Building Type WUI Average(m3/yr/m2)WUI Range(m3/yr/m2)Core (excl. Bio Sci South) 1.8 (1.4) 0.2-9.8Core Lab (excl. Bio Sci South) 2.5 (1.7) 0.2-9.8Ancillary 1.3 0.4-3.6Ancillary Residential 1.0 0.4-2.0September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 18WUIResultsMetered consumption most likely includes water usage in Biological Sciences Building West, thus the area should be more than doubled.NIGHTTIME CONSUMPTION 23% of campus total nighttime supply metered with high resolution (14% in Top 10)  Metered buildings contribute similarly as for whole year. 35% of campus total nighttime supply metered in non-irrigation season (23% in Top 10) Metered buildings contribute more to nighttime consumption outside of irrigation season. Separate irrigation systems thus account for 12% of nighttime consumption during irrigation season? *  CEME and H.R. Macmillan replace AMS Nest and Hennings in Top 10 consumers.Potential annual savings of 616,000 cubic meters or $550,000 (16% of annual water costs) if currently metered base flow consumption in high-pressure zone is eliminated completely.September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 19 41% of U Blvd nighttime supply metered in high pressure zone in non-irrigation season (29% in Top 10) Almost 60% not captured (some in low-pressure zone) . Metered consumption accounts for almost 90,000 cubic meters.m3/yrNighttime is here defined as time period between midnight and six in the morning and irrigation season as beginning of May to end of September.Results* In range of Stantec Consulting estimate in 2011.ESTIMATED BASE FLOW CONSUMPTIONMain assumption: whatever is consuming water when occupancy is minimal in non-irrigation season, runs continuously.35% of estimated campus base flow consumption metered (664,105/1,887,017 m3).Almost $600,000 annually already metered (Campus total $1,681,332).Uncertainty related to absolute values from scaling, but relative size indicates base flow contribution. Except for Orchard Commons, Ponderosa North and the Aquatic Centre, Core Lab buildings are the biggest contributors.Need to investigate Aquatics Lab in Biological Sciences Building South and the distribution of dechlorinated water to the Building West.September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 20Colors indicate high and low pressure zones. ResultsEXAMPLE: FORESTRY SCIENCES▪The Aquatic Ecology Lab (fish tank) has a flow-through water system.▪Lab had an incident that put it out of operation at the end of last year.▪In operation the minimum consumption of 2 cubic meters every 15 minutes converts to 70,000 cubic meters annually. ▪Indicates annual cost of $62,000 to EWS for running the lab – compare to cost of installing recirculating system. September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 21ResultsFINDINGS SUMMARY▪Total Campus base flow is estimated at 1,887,017 cubic meters (~50% of annual consumption).  ▪ U Blvd main supply accounts for 1,483,914 cubic meters of base flow alone (78 %).▪55 of the 80 buildings listed with water consumption data in Skyspark are deemed good quality and considered in this study.▪ The majority are Core Buildings located in the High pressure zone and consume water from the U Blvd campus supply only.▪ 18% of campus total supply is metered in core buildings (64 of the 218 core buildings when considering sub-fed buildings)▪High-resolution metered buildings account for approx. 24% of campus total supply (14% in Top 10 consumers).▪Outside of irrigation season, the contribution to nighttime consumption increases to 35 % (23% in Top 10 consumers).▪ 41% of U Blvd supply when only considering buildings located in high-pressure zone (29% in Top 10 consumers)▪If main assumption holds, this converts to 664,105 cubic meters of annual base flow consumption.▪ High-pressure zone (Core building area) annual base flow estimate alone is 616,000 cubic meters.▪Potential annual savings of almost $600,000 (17% of total water costs) if currently metered base flow consumption is eliminated completely.▪Core Lab buildings have the highest water usage intensities and contribute more to campus base flows.▪More than 70% of the Campus water consumption is currently not accounted for in high resolution (Skyspark).September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 22ConclusionsRECOMMENDATIONS▪ Clear up any confusions regarding where meter is located and the area/purpose it covers.▪ Investigate whether already-metered base flow consumption can be reduced (cost and feasibility).▪ Make sure already installed high-resolution meters are working properly and connected.▪ Prioritize installing high-resolution meters in buildings with lab space and separate irrigation systems.▪ Improve data access: ▪ High-resolution data download of all buildings less time-consuming.▪ Direct access to campus water supply and sewer data. (PRV data available?)▪ Include descriptions (metadata) on meters, logging, conversions, areas and buildings covered etc. ▪ Gather data from already installed high-resolution meters on campus irrigation systems, if any.▪ Improve data quality:▪ Investigate why campus supply flow rates are off by 10% when compared to consumption readings.▪ Minimize uncertainties regarding missing data, zeros and consumption spikes for building meter data. ▪ Include some of the metrics used in this study in further development of Skyspark.September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 23POTENTIAL PROJECTS Should be substantial opportunity for water conservation and system optimization if reliable data at hand:1. Explore correlations and establish building benchmarks when utilizing water consumption and occupancy data along with building specifics and seasonality. 2. Create map visualizations of building consumption for easier analysis and interpretations.3. Investigate potential leaks by analyzing high-resolution campus supply, sanitation, building and irrigation data. 4. Apply clustering technique and machine learning to predict performance and highlight abnormalities. 5. Multi-objective optimization of complete system (PhD research goal).September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 24ConclusionsREFERENCES & RESOURCES1. Stantec Consulting (2011) “UBC Water Base Flow Analysis” Final Report.2. UBC Water Action Plan (TBC 2018): Bud Fraser (Senior Planning & Sustainability Engineer, S&E)3. Water Consumption in Buildings:▪ Water Meters: Colin Mingus (Climate & Energy Engineer, EWS)▪ Data Logging: Blair Antcliffe (Energy Engineer, EWS) ▪ Skyspark Platform: Arthur de Robert (Energy Data Analyst, EWS) ▪ Maps and Schematics: Erin Kastner (Geospatial Information Manager, EWS)4. Water Supply on Campus:▪ Water Mains Data: Jenny Liu (Mechanical Utilities Engineer, EWS)▪ Operational Questions: Aleks Paderewski (Mechanical Utilities Manager, EWS)5. Spreadsheet “Campus Building Data v9.0 2018 working_20180418_NM”: Nicole Montgomery (Coordinator, Reporting, Monitoring & Systems Review, S&E)September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 25APPENDIX 1: SPREADSHEETThe accompanying “Building Consumption Data Analysis” spreadsheet (~70 MB) contains the following:Summary & Rankings: Summary of metadata, statistics and metrics for each listed building in Skyspark (each water meter).Building Analysis: The option to choose one Skyspark listed building (water meter) to see summary and additional charts for detailed analysis.Campus Analysis: Data and cost analysis for the campus water supplies.Data: Gathered and treated data from the Skyspark Buildings App (under Building Utilities) for the time period July 1st 2017 to June 30th 2018.16th Ave Supply: Gathered data on flow rate measurements and computed consumption for the campus supply for same time period.U Blvd Supply: Gathered data on flow rate measurements and computed consumption for the campus supply for the same time period.CCP Building List: Data extracted from the spreadsheet “Campus Building Data v9.0 2018working_20180418_NM” by filtering on existing buildings only.One can find a number of comments explaining more details in the spreadsheet. September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 26AppendicesAPPENDIX 2: TECHNICAL WORKCampus Supply Data:▪ When comparing consumption computed from average flow rate measurements and the manual readings, a constant error of ~10% was found. Assuming that the offset origins from the flow rate measurements or conversion factors, the error was corrected for by adding 10% to every datapoint for the computed consumption.Building Consumption Data:▪ Initially, building area and usage percentages were gathered from Skyspark, but encountering inconsistencies between these numbers and the CCP Spreadsheet that was used as input, it was decided to use this list directly. A lot of manual work was required by cross-referencing buildings with campus map of meters and distribution pipes, to make sure that the complete area covered by each meter (due to subfed buildings) was included, as well as usage percentages updated accordingly.▪ All water meters are configured to send a pulse to the ION meter (data logger) when 0.1 cubic meters of water has been consumed. A server communicating with the ION meters, converts the pulse count into accumulated consumption with a resolution of 15 minutes, which is the data made available in Skyspark. If communication issues are experienced in the data logging process, this result in missing datapoints during the affected time period, and depending on the cause behind, the first datapoint recorded may represent cumulative consumption during the missing time period or it could be a pulse count conversion error. Since there is no way of easily determining the case, and the timing of the consumption is relevant for this study, it was decided to remove all consumption spikes after a period of missing data. ▪ It was discovered that some meters had a very high percentage of zero measurements, which can be due to malfunction meter or low consumption (less than 0.1 cubic meters in 15 minutes). Further, in some cases large spikes in consumption were identified after a longer period of zeros and these were removed using the same rationale as for first record after missing datapoints above. Overall, distinguishing these events is not straight forward, and again, since timing of consumption is relevant for this study, the percentage of zero measurements was used along with confidence factors (missing datapoints) to determine the quality of the data. Manually adjusting and reviewing the results, the requirement for Good quality data was set to confidence above 85% and percentage of zeros during the day less than 75%. The latter translates to minimum hourly data resolution.     September 6th, 2018 ANALYSIS OF CAMPUS WATER BASE FLOWS 27Appendices


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