10.15.2018 WRAB Packet - Agenda 4 Water Supply and Climate ChangeCITY OF BOULDER
WATER RESOURCES ADVISORY BOARD
INFORMATION ITEM
MEETING DATE: October 15, 2018
AGENDA TITLE: Information Item – Water Supply Planning Model Update
PRESENTERS:
Jeff Arthur, Director of Public Works for Utilities
Joe Taddeucci, Water Resources Manager
Kim Hutton, Water Resources Engineer
EXECUTIVE SUMMARY:
In 2003, the City of Boulder (city) began considering climate change in its water supply
planning. In 2008, the city engaged in an innovative approach to simulate the effects of climate
change on municipal water supply and concluded that the city can reliably meet water demands
in all but the hotter and drier modeled scenarios. The city has since refined its planning model
with updated municipal demand data and improvements in hydrologic and climate science and
modeling techniques. The refined model allows staff to evaluate whether projected build-out
water demands can be reliably met under varied hydrology and climate scenarios. The city
simulated historical baseline conditions and seven climate change scenarios to determine water
supply and demand changes in the year 2050. In the baseline and four of the wetter climate
change scenarios, build-out water demands can be met reliably. In three of the drier scenarios,
minor violations of the reliability criteria occur, and modified system operations will be
evaluated to determine if the water supply system can adequately meet demands. The purpose of
this item is to provide information on potential climate change effects on municipal water supply
and to provide WRAB members an opportunity to ask questions.
BACKGROUND:
Since its incorporation in 1871, the City of Boulder’s long-term planning efforts have prioritized
high quality and reliable municipal water supply. Water supply planning traditionally considered
the following elements:
•future water demand derived from projections of population growth and land use changes
and water use trends;
•water supply availability due to normal fluctuations in precipitation and streamflow and
the associated effects on the city’s water rights yield;
•building infrastructure capacity capable of meeting demands; and
•source water quality.
As a result of several significant planning analyses in the early 1900s through the late 1980s, the
city expanded and diversified its municipal water supply system. From the 1900s to the 1960s,
the city grew its water supply from the mouth of Boulder Creek to North Boulder Creek and the
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 1
Silver Lake Watershed. From the 1950s – 2001, the city expanded its supplies to Middle Boulder
Creek and acquired the Barker system. From the 1950s – 2001 the city was annexed to the
Northern Colorado Water Conservancy District and developed supplemental, west slope
Colorado-Big Thompson and Windy Gap supplies including construction of Boulder Reservoir
and the Boulder Reservoir Water Treatment Plant.
In 1989, City Council adopted the Raw Water Master Plan, which established water supply
“reliability criteria”. These reliability criteria provide policy guidance for the city’s source water
supply by defining the acceptable frequency of water restrictions for droughts of varying
severity. The reliability criteria are central to the city’s water supply philosophy1 and are used to
test the adequacy of the city’s water supply through computer modeling.
Due to the complexity of the city’s water rights and supply system, the city uses a computer
model2 (Boulder Creek Model) to simulate water supply operations and evaluate long-term water
supply adequacy. Water supply is considered adequate if the system can operate without
exceeding the frequency of water restrictions specified by the reliability criteria. With
advancement in climate sciences and recognition of the effects greenhouse gas emissions may
have on water supplies and demands, the city began considering climate change in water supply
planning in 2003. Climate modeling has become central to water supply management because of
the effects temperature and precipitation have on streamflow and water demand. In 2008, the city
engaged with NOAA, University of Colorado, Stratus Consulting and Hydrosphere Consultants
to develop a method of assessing the effects of climate change on water supply. This assessment
incorporated climate change scenarios developed by the international science community in the
Boulder Creek Model. The results of this analysis indicated that Boulder would have to
implement more water supply reductions than described in the reliability criteria under some of
the hotter and drier scenarios. It was also recommended that the cit y monitor climate science and
adapt water supply planning appropriately. Starting in 2015, the city began to refine the climate
change assessment capabilities of the Boulder Creek Model using updated and higher resolution
climate information and modeling techniques from the science community.
ANALYSIS:
The Boulder Creek Model uses streamflow, weather (temperature and precipitation) and
municipal demands as model inputs and simulates a number of components, including Boulder
Creek basin water rights administration, Northern Water CBT and Windy Gap availability and
the city’s raw water supply. The model includes a 100-year record of observed streamflows and
weather and an annual water demand derived from the 2015 Boulder Valley Comprehensive Plan
demographic projections3 as a “baseline” scenario. To assess water supply reliability for various
climate change scenarios, the baseline scenario is modified with climate-adjusted streamflow and
climate-adjusted water demand. The climate-adjusted data is developed from Global Climate
Model projections of future changes in temperature and precipitation.
1 The reliability criteria are included in the city’s current Drought Plan (Volume I-2010 and Volume II-2004) and
most recent Source Water Master Plan (Volume 1 and Volume 2, 2009).
2 The model was first developed by Hydrosphere Consultants, which became AMEC Earth & Environmental, in the
late 1980s. The city currently contracts with Rozaklis and Associates and Lynker Technologies for modeling
support.
3 Demographic projections obtained from 2015-2040 Boulder Valley Comprehensive Plan Update Projections and
2015-2040 Boulder Valley Comprehensive Plan Update Projections Methodology.
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 2
The following updates and changes were made between the 2008 and 2015 models:
•The city’s annual build-out demand was changed from 28,600 acre-feet to 22,000 acre-
feet4;
•Outdoor variation of municipal demand as a function of weather has been incorporated
into the model whereas previously it was fixed;
•Water rights administration modules were refined;
•Use of higher spatial resolution climate and hydrology models allows for simulation of
varied weather across the Boulder Creek Model’s individual water source and use areas;
and,
•Climate projections from the latest two international modeling efforts (CMIP3 and
CMIP5) were used in the analysis5 whereas the previous model only used CMIP3 data.
Using the refined climate information and Boulder Creek Model, the city analyzed the water
supply system performance in the year 2050 under seven climate change scenarios. The seven
scenarios were carefully selected from 209 climate projections to be representative of a range of
potential future conditions. While all 209 projections indicate a warmer future, they vary
between wetter and drier as well as seasonal precipitation patterns. The seven selected scenarios
represent future temperature projections of 1.6 – 3°C warmer and precipitation projections of 9
percent drier to 23 percent wetter than the reference period6. The seven scenarios also represent a
wide range of projected changes to streamflow and outdoor water demand, which are critical
components of water supply management.
Conclusions reached based on the most current modeling efforts are similar to results of the 2008
study and are summarized as follows:
•Annual municipal demands increase in all climate change scenarios due to the effects of
increased evapotranspiration on outdoor water use. While build-out demand is modeled at
22,000 acre feet per year, simulated climate-adjusted demands range from 23,100 to
25,400 acre feet per year as illustrated by the graph in Attachment 1.
•The annual volumetric change in climate-adjusted Boulder Creek streamflows varies
from 109 percent more to 39 percent less, but all scenarios show higher flows in May and
lower flows in July – September which reflects a shift to earlier runoff as illustrated by
the graph in Attachment 2.
•More supply comes from storage (rather than direct flow) during late summer due to
lower late-summer streamflow.
•Boulder can reliably meet build-out water demands in the baseline scenario and four of
the wetter climate change scenarios. In three of the drier climate change scenarios, the
decreased available streamflow and increased municipal demands result in minor
violations of the reliability criteria.
NEXT STEPS:
4 Build-out water demand was calculated in the 2016 Water Efficiency Plan to be 19,980 acre-feet per year assuming
current conservation efforts and climate conditions and natural replacement of indoor water fixtures. A 10% factor
of safety is added to the build-out demand in the Boulder Creek Model.
5 An ensemble of 209 climate projections from CMIP3 and CMIP5 were analyzed as part of this study. The
ensemble includes a variety of Global Climate Models and greenhouse gas emission scenarios that generate a range
of projected temperature and precipitation scenarios.
6 Future changes to temperature and precipitation are measured against the reference period of 1970 – 1999.
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 3
Modeling efforts to date assume that current system operating practices would occur unchanged
in 2050. Future analyses will consider adaptation strategies such as modified operations as
described below.
1.Modify operating criteria in the model to adapt to changes in runoff and temperature.
Improvements in system performance are expected under the hotter and drier scenarios.
2.Extend the model forecasting to the year 2070 to provide a longer planning horizon.
Variations in temperature and precipitation projections increase over this period.
3.The current model is run using a 100-year record of observed hydrology and weather
(which drive water demand). Develop hydrology and water demand data for the period
1566 – 2002 from reconstructions of tree ring data in order to model a wider range of
climate and streamflow variability.
4.As needed, evaluate the effects of policy changes to the reliability criteria, drought
response, water conservation goals, water rights acquisition and storage development on
system performance.
As the modeling work continues, staff will return to WRAB in 2019 with any significant
updates. Once the climate change and water supply analyses are complete, results will be used to
update the drought plan and drought response triggers.
Attachments
1 – Baseline and climate-adjusted average annual and monthly municipal demands
2 – Baseline and climate-adjusted Boulder Creek streamflows
3 – Updated Boulder Climate Change Water Supply Assessment
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 4
Attachment 1. Baseline and climate-adjusted average annual and monthly municipal demands
0
200
400
600
800
1000
1200
Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug SepQuartermonthly demand in acre-feet2050 DataObject 23 -Average COB Demand 1915-2014
Baseline CMIP3+5_ll_2050 CMIP3+5_9010_2050 CMIP3+5_7525_2050
CMIP3+5_c_2050 CMIP3+5_2575_2050 CMIP3+5_1090_2050 CMIP3+5_ur_2050
Baseline
CMIP3+5_ll_2050
CMIP3+5_9010_2050
CMIP3+5_7525_2050
CMIP3+5_c_2050
CMIP3+5_2575_2050
CMIP3+5_1090_2050
CMIP3+5_ur_2050
Scenario Avg Annual Total (af)
22,075
25,365
24,889
24,283
23,840
23,625
23,140
23,346
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 5
Attachment 2. Baseline and climate-adjusted Boulder Creek streamflows
0
20000
40000
60000
80000
100000
120000
140000
160000
180000
200000
Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug SepAcre-Feet2050 Average Total Inflows 1915-2014
Baseline CMIP3+5_ll_2050 CMIP3+5_9010_2050 CMIP3+5_7525_2050
CMIP3+5_c_2050 CMIP3+5_2575_2050 CMIP3+5_1090_2050 CMIP3+5_ur_2050
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 6
Beginning in 2003, as part of its water supply planning efforts, the City of Boulder (Boulder) has
periodically assessed the vulnerability of its water supply system to climate change. Boulder’s
previous assessment in this area was completed in 2008 and utilized Boulder’s previously
developed Boulder Creek Watershed Model (BCWM) and the climate change science and
related data that were available at that time.
Starting in 2015, Boulder began this effort to update its climate change water supply
assessment to incorporate the latest climate change modeling, recent downscaling applications
of climate change science to regional statewide water supply planning, and related refinements
in the BCWM. This interim report on the Updated Climate Change Water Supply Assessment
describes the potential effects of a selected series of seven climate change scenarios on
Boulder’s water supply system that may be expected to occur by the year 2050.
This interim report summarizes the methods and outputs related to the climate change analysis
and the effects of the selected range of climate changes scenarios on Boulder’s existing water
supply system. The range of potential options for mitigating or adapting to climate change
effects on the reliability of Boulder’s water supply will be addressed in the final version of this
report.
2.0 Climate Change Analysis
This study utilizes the methods and data products developed as a part of the Colorado River
Water Availability Study (CRWAS) Phase II project (CRWAS-II), completed by the Colorado
Water Conservation Board (CWCB). CRWAS-II used data from the Coupled Model
Intercomparison Project Phase 3 (CMIP3) and Phase 5 (CMIP5) archives to process climate
projections for the entire state of Colorado. The CMIP3 archive of twenty-first century
projections aligns with the IPCC Assessment Report 4 (AR4) and contains 112 projections from
16 GCMs forced with three Special Report on Emission Scenarios (SRES) emissions pathways
(A1B, A2, and B1) (CWCB, 2012). The CRWAS-II project used a subset of 97 projections from
the 234 runs that comprised the CRWAS-II archive. The CMIP5 model runs are forced by
representative concentration pathways (RCPs), which are used to represent different
assumptions about the effect of past and future greenhouse gas emissions. There are four
different RCPs used for the CMIP5 experiment: 2.6, 4.5, 6.0, and 8.5. The CRWAS-II project
developed an ensemble of 112 CMIP3 and 97 CMIP5 projections for a total of 209 projections
for future time horizons of 2040, 2050 and 2070 (CWCB, 2015a).
2.1 Methods
2.1.1 General Approach
The climate change analysis completed for Boulder used projected future changes in weather to
calculate hydrology change factors, which were used to perturb historical natural flows. Climate
change projections from global climate models (GCMs) provide historical (baseline) and future
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 7
Attachment 3
Draft Interim Report
Updated Boulder Climate Change Water Supply Assessment
October 2, 2018
1.0 Introduction
(projected) values for temperature and precipitation using different radiative forcings scenarios
(RCPs) or emissions scenarios (SRES). Lynker developed weather change factors (for
precipitation) and offsets (for temperature) using the future and historical weather from the
GCMs as a part of the CRWAS-II project. These change factors were used to perturb the
historical gridded weather data and then used to model hydrology. The hydrology model was
run with the historical gridded weather and the perturbed gridded weather to produce simulated
baseline modeled flow and simulated future modeled flow, respectively. The difference between
these two modeled datasets is used to develop a time series of hydrology change factors that
are applied to natural flow to arrive at climate-adjusted natural flow. This multi-stage process
was used for CRWAS-II and has three advantages: it helps to reduce model bias, it improves
the representation of the effect of climate on interannual variability, and it bases all projections
on a familiar and accepted flow baseline. A schematic depicting the climate change adjustment
workflow is shown in Figure 2-1.
Figure 2-1: Climate Change Analysis Workflow
2.1.2 Historical Weather Forcings
Daily gridded historical weather forcings were needed to run the hydrology model and get flow
output. This data needed to represent observed weather over the duration of Boulder’s flow
period of record, 1907 to 2014. However, the CRWAS-II project relied on the Maurer el al.
(2002) 1/8-degree daily gridded meteorological dataset from 1950 to 2000, which was later
extended to 2014 by Andy Wood (Lynker, 2014). Since Boulder’s flow data extended well
before 1950, the Livneh et al. (2013, as extended) 1/16-degree daily gridded meteorological
dataset was used, which spans from 1915 to 2015. There was no daily gridded meteorological
data corresponding with Boulder’s flow data from 1907 to 1914. Therefore, the natural flows
were adjusted directly using average change factors as described below and in Section 2.1.5.
We anticipated slight, but not unreasonable differences or biases between the Maurer et al. and
Livneh et al. meteorological datasets. The Livneh et al. dataset was used in its entirety from
1915 to 2014, supplemented by a static adjustment of Boulder’s flow data from 1907 to 1914.
Since the Livneh et al. dataset was in 1/16-degree grids and Boulder’s hydrology model is
currently set-up for a 1/8-degree grid, the forcings data were aggregated to a 1/8-degree grid. A
snapshot of results from the comparison between the Maurer and Livneh datasets over the
common period of record (1950 to 2014) is shown Appendix A.
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 8
Attachment 3
2.1.3 GCM-based Weather Change Factors
The CRWAS-II project developed weather change factors for temperature and precipitation on a
grid cell by grid cell basis for each of the 209 available projections over the three future periods
(2040, 2050 and 2070). The average monthly baseline data (1 set of 12 monthly values) from a
climate projection and the average monthly future data (1 set of 12 monthly values) from the
same climate projection are used to obtain one set of average monthly change factors. These
change factors were developed for the three future time periods, 2040, 2050 and 2070. A 30-
year averaging period was used for both the baseline (historical) period and the future period
from the climate projection. Thus, the 2040 analysis used the average change from 2025 to
2054, the 2050 analysis used the average change for years 2035 to 2064, and the 2070
analysis used the average change from 2055 to 2084. The corresponding baseline (historical)
averaging period for each projection was 1970 to 1999. These change factors were then applied
back to the historical gridded meteorological dataset (discussed in the Section 2.1.2), to provide
climate-adjusted temperature and precipitation (CWCB, 2015b). The CRWAS process for
developing climate-adjusted weather data is shown in Figure 2-2.
Source: Colorado River Water Availability Study (CWCB, 2012).
Figure 2-2: CRWAS Climate-adjusted Weather Schematic
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 9
Attachment 3
The CRWAS-II project developed seven pooled model projections, each consisting of the mean
conditions of 10 individual projections, to provide more information than a single projection, and
to represent the range of model-to-model disagreement without relying on the full ensemble of
209 CMIP3 and CMIP5 projections. The aggregated climate scenarios were developed using
characteristic points of runoff and consumptive irrigation requirement (CIR) to consider
influences in water supply and demand, respectively, from each projection (Table 2-1). The
designation “9010” represents the 90th percentile CIR and 10th percentile runoff, which will have
a more severe impact on water supply systems (more demand and less supply), while the
designation “2575” represents the 25th percentile CIR and 75th percentile runoff, which will have
a less severe impact on water supply systems (less demand and more supply) (CWCB, 2015b).
Table 2-1: Characteristics of CRWAS-II Future Climate Scenarios
Designation CIR Percentile Runoff Percentile
Lower Left 100% 0%
9010 90% 10%
7525 75% 25%
Center 50% 50%
2575 25% 75%
1090 10% 90%
Upper Right 0% 100%
2.1.4 Climate-adjusted Weather
The weather change factors were already developed as a part of the CRWAS-II project. They
were used to perturb the historical daily gridded meteorological dataset for use in the hydrology
modeling for Boulder. We used seven aggregated projections (Lower Left [LL], 9010, 7525,
center, 2575, 1090, and Upper Right [UR]) for the 2050 future period, for a total of 7 climate
projections. A snapshot of the weather change factors for grid cells within the Boulder Creek
watershed is provided in Section 2.2.1.
2.1.5 Climate-adjusted Hydrology
The Variable Infiltration Capacity (VIC) hydrology model was used calculate runoff and
evaporation within the Boulder Creek watershed using the meteorological data. There were two
types of VIC model runs used for this analysis, baseline hydrology and future hydrology. The
historical gridded meteorological dataset was used to force the model once to obtain the
modeled baseline hydrology. Then the model was run using the perturbed (climate-adjusted)
weather to obtain modeled future hydrology. The change between the baseline and future
datasets represents the anticipated change from the future (climate-adjusted) weather. These
hydrology change factors were averaged to produce a time series of monthly change factors
representing the period of record (1915 to 2014). The flow change factors were then used to
perturb Boulder’s historical natural flow data to produce climate-adjusted flow, which was used
in Boulder’s water supply model (CRAM). A similar set of change factors were calculated for
CIR to perturb historical values of irrigation use within Boulder’s water supply model. The CIR
change factors were calculated for the months of April through October, due to abnormalities in
the VIC model’s calculation of PET, where actual ET can increase above PET. As previously
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 10
Attachment 3
stated, since the Livneh et al. daily meteorological dataset began in 1915, so the model runs
within the Boulder model were limited to the 100-year period of 1915-2014.
2.2 Results
2.2.1 Analysis of Weather Results
The meteorological change factors from the CMIP3 and CMIP5 (CMIP3+5) projections were
plotted according to change in temperature and precipitation for Boulder basins 1 through 14.
The temperature and precipitation change factors for the seven aggregated projections in future
period 2050 are shown in Figure 2-3.
Figure 2-3: 2050 Change Factors
2.2.2 Analysis of Hydrology Results
The hydrology change factors were averaged together according to 16 inflow basins used by
Boulder’s water supply model. A map of Boulder’s basins is provided in Figure 2-4 and a list of
the basin names is provided in Table 2-2. The average monthly flow change factors were
plotted for two of Boulder’s basins, Basin 6 (Boulder Creek above Barker Reservoir) and Basin
11 (South Boulder Creek above Gross Reservoir), both of which represent larger headwater
inflows. The flow change factors for the seven projections within future period 2050 are shown
for Basin 6 in Figure 2-5 and Basin 11 in Figure 2-6.
The change factor plots can be interpreted as an increase in flow for values greater than one,
and a decrease in flow for values less than one. The change factors for Basin 6 and Basin 11
are similar, both showing the largest flow increase in May and flow decreases for most other
months, especially July through December. In both Basin 6 and Basin 11, the Center scenario
(yellow line) shows flow decreases for most months, June through February. These figures
UR1090
2575
Center
7525
9010
LL
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5Precipitation Change FactorTemperature Change (deg C)
2050
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 11
Attachment 3
highlight the expected changes in the timing of runoff, where the hydrograph’s peak will arrive
earlier due to increases in temperature.
Figure 2-4: Map of Boulder’s Basins
Table 2-2: Boulder CRAM Basins
Basin ID CRAM Inflow Description
01 Inflow 1 NBC above Silver Lake
02 Inflow 2 NBC above SL Pipeline
03 Inflow 4 NBC above LW Pipeline
04 Inflow 5 Sherwood Creek
05 Inflow 7 NBC above MBC
06 Inflow 10 MBC above Barker Res
07 Inflow 11 MBC above NBC
08 Inflow 15 BC above Orodell
09 Inflow 16 Four Mile Creek
10 Inflow 25 Bear Creek
11 Inflow 12 Eldorado Virgin Flows
12 Inflow 13 SBC gains at DW
13 Inflow 14 SBC gains at Eldo Gag
14 N/A
15 N/A
16 Colorado River at Hot Sulphur Springs
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 12
Attachment 3
Figure 2-5: Basin 6 2050 Runoff Change Factors
Figure 2-6: Basin 11 2050 Runoff Change Factors
0.0
0.5
1.0
1.5
2.0
2.5
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecChange FactorBasin 6 2050 Runoff Change Factors
UR 1090 2575 C 7525 9010 LL
0.0
0.5
1.0
1.5
2.0
2.5
3.0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecChange FactorBasin 11 2050 Runoff Change Factors
UR 1090 2575 C 7525 9010 LL
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 13
Attachment 3
The average monthly CIR change factors for future period 2050 are shown below for Basin 6
(Figure 2-7) and Basin 11 (Figure 2-8).
Figure 2-7: Basin 6 CIR Change Factors
Figure 2-8: Basin 11 CIR Change Factors
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
18.0
20.0
1 2 3 4 5 6 7 8 9 10 11 12Change FactorBasin 6 2050 CIR
UR 1090 2575 C 7525 9010 LL
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
1 2 3 4 5 6 7 8 9 10 11 12Change FactorBasin 11 2050 CIR
UR 1090 2575 C 7525 9010 LL
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 14
Attachment 3
3.0 Water Supply System Analysis
The results from the climate change analysis (climate adjusted flow, irrigation use and
precipitation) were used to model future climate scenarios using the Boulder Creek Watershed
Model
3.1 Boulder Creek Watershed Model
The Boulder Creek Watershed Model is an application of the CRAM water allocation tool, which
is a generalized network variant of linear programming. The Model was originally developed for
Boulder as part of Boulder’s 1988 Raw Water Master Plan and has been continuously refined
and upgraded since then. Boulder uses the Model to assess the reliability of its water supply
system, to do supply and demand-side planning, to evaluate the effects of facility and
operational changes, and to do short-term operational forecasting.
When used to assess the reliability of Boulder’s water supply system, the Model imposes a
relatively fixed set of Boulder’s municipal demands, typically future projected “buildout”
demands, upon variable time series of natural inflows, transbasin imports, and competing
agricultural and municipal water rights. The Model explicitly emulates Boulder’s drought
recognition and response triggers as described in Boulder’s Drought Plan. The degree to which
Boulder’s municipal demands are met and the frequency of drought restrictions are evaluated in
the context of Boulder’s water supply reliability criteria.
The Model simulates all necessary components of the Boulder Creek basin and Boulder’s raw
water supply including natural stream flows, reservoirs and direct diversions, water uses and
return flows, exchanges, transbasin imports into the basin, and demands by downstream South
Platte water rights. It is operated on a quarter-monthly time step (i.e., weekly) using historical
inflow and weather data for the historical period of 1915-2014. The Model also includes times
series of paleo natural flows derived from tree ring chronologies covering the period of 1566-
2002, which can be used to test the effects of more severe and sustained drought than are
represented in the 1915-2014 period of historical stream flow records.
3.2 2050 Water Supply Analysis
Boulder’s municipal demand was set at a fixed annual volume of 22,000 acre-feet per year in
the historical baseline scenario, which is 10% greater than Boulder’s projected buildout water
demand as described in Boulder’s 2016 Water Efficiency Plan. Adding a 10% safety factor to
Boulder’s modeled demand is consistent with Boulder’s previous reliability assessments.
Boulder’s municipal demand is comprised of both indoor and outdoor demand components.
The indoor demand was held constant while the outdoor demand was modeled as varying from
year to year in response to monthly and annual variations in irrigation requirement, both in the
historical baseline scenario and in the seven pooled climate change scenarios.
The seven pooled 2050 climate change scenarios were simulated in the Boulder Model to
determine future changes to water supply and demand. In each figure, the heavy black solid
line indicates baseline or historical conditions, the brown solid line represents the Upper Right
scenario, which is typically the most favorable future conditions, and the dotted orange line
represents the Lower Left scenario, which is typically the least favorable future conditions.
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 15
Attachment 3
The average monthly total inflows for 2050 show a change in the timing of the flows, where the
peak flow still occurs in June, but there is increased flow in May. This results in decreased
runoff for July, August and September compared to the historical baseline. This result is
common across all 2050 climate projections, including Upper Right. The average monthly
inflows are shown in Figure 3-1. Annual flow volumes increase significantly in two of the
scenarios and decrease significantly in three of the scenarios. Average annual natural flow
volumes at Orodell and Eldorado are shown in Table 3-1.
Figure 3-1: Average Monthly Total Inflows
0
20000
40000
60000
80000
100000
120000
140000
160000
180000
200000
Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug SepAcre-Feet2050 Average Total Inflows 1915-2014
Baseline CMIP3+5_ll_2050 CMIP3+5_9010_2050 CMIP3+5_7525_2050
CMIP3+5_c_2050 CMIP3+5_2575_2050 CMIP3+5_1090_2050 CMIP3+5_ur_2050
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 16
Attachment 3
Table 3-1: Changes in Average Annual Flow Volumes by Scenario
Boulder’s average annual demands increase in all of the seven pooled climate change
scenarios, because of increased outdoor demands. Boulder’s average monthly demands under
each of the scenarios are shown in Figure 3-2, and Boulder’s average annual demands are
shown in Table 3-2.
Figure 3-2: City of Boulder Average Monthly Demands
Scenario
Mean Annual
Natural Flow,
Boulder Creek at
Orodell, AF
Percent
change
Mean Annual
Natural Flow, S.
Boulder Creek
near Eldorado, AF
Percent
change
baseline 71504 56040
upper right 125026 75% 116847 109%
10-90 89655 25% 74848 34%
25-75 77586 9% 62753 12%
center 68903 -4%54466 -3%
75-25 58362 -18% 45225 -19%
90-10 52427 -27% 40611 -28%
lower left 44684 -38% 34075 -39%
0
200
400
600
800
1000
1200
Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug SepQuartermonthly demand in acre-feet2050 DataObject 23 -Average COB Demand 1915-2014
Baseline CMIP3+5_ll_2050 CMIP3+5_9010_2050 CMIP3+5_7525_2050
CMIP3+5_c_2050 CMIP3+5_2575_2050 CMIP3+5_1090_2050 CMIP3+5_ur_2050
Baseline
CMIP3+5_ll_2050
CMIP3+5_9010_2050
CMIP3+5_7525_2050
CMIP3+5_c_2050
CMIP3+5_2575_2050
CMIP3+5_1090_2050
CMIP3+5_ur_2050
Scenario Avg Annual Total (af)
22,075
25,365
24,889
24,283
23,840
23,625
23,140
23,346
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 17
Attachment 3
Table 3-2: City of Boulder Average Annual Demands
Agricultural demands served by competing irrigation rights in the Boulder Creek basin decrease
slightly in the Upper Right and 10-90 scenarios and increase by varying degrees in the other
scenarios, as shown in Table 3-3. However, irrigation diversions do not increase significantly
compared to the historical baseline, primarily because natural flows significantly decrease in the
mid and late summer period in the climate change scenarios. Most of the irrigation demands in
the Boulder Creek basin are served only by direct flow rights, and just a small fraction of those
rights are sufficiently senior to command the reduced flow in the creek during the summer
season.
Table 3-3: District 6 Mean Annual Irrigation Demand and Diversions
The effects of the seven pooled climate change scenarios are illustrated in Figures 3-3 through
3-10 and summarized in Table 3-4. The Upper Right, 10/90, 25/75 and Center scenarios have
slight to moderate negative effects on Boulder’s water supply range, while still meeting
Boulder’s reliability criteria. The 75/25, 90/10 and Lower Left scenarios have more serious
effects, including demand reductions more frequently than once in 20 years on average.
It should be noted that the results shown in Figures 3-3 through 3-10 reflect the operation of
Boulder’s existing water supply system under current operating rules and drought recognition
and response guidelines, without any mitigation or adaptation measures. There are several
nonstructural and structural mitigation and adaptation measures that should be explored.
Municipal Demand (acre-feet)
climate
change
lower left
climate
change
90-10
climate
change
75-25
climate
change
center
climate
change
25-75
climate
change
10-90
climate
change
upper
right
historical
baseline
Indoor Demand 12439 12439 12439 12439 12439 12439 12439 12439
Outdoor Demand 13196 12450 11844 11401 11186 10701 10907 9636
Total Demand 25635 24889 24283 23840 23625 23140 23346 22075
% increase in outdoor demand
relative to baseline 37% 29% 23% 18% 16% 11% 13%
Scenario
District 6
Mean Annual
Irrigation
Demand, AF
Percent
change
District 6
Mean Annual
Irrigation
Diversions, AF
Percent of
Irrigation
Demand
Met
baseline 128610 117943 92%
upper right 112450 -13% 110894 99%
10-90 121926 -5%113259 93%
25-75 139462 8% 120095 86%
center 150136 17% 120705 80%
75-25 173186 35% 121640 70%
90-10 198793 55% 119390 60%
lower left 218924 70% 115751 53%
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 18
Attachment 3
For example, the rate of treated water deliveries from Boulder’s 63rd Street water treatment
facility (WTF) has a significant effect on the drought resistance of the water supply system.
Increased fall and winter season deliveries from this facility result in greater storage volumes
maintained through the winter in Boulder’s mountain reservoirs. The trade-off is the increased
expense of pumping water from the 63rd Street WTF and the hydropower revenue foregone by
reducing fall and winter season releases from Barker Reservoir and the North Boulder Creek
watershed reservoirs. The Lower Left scenario was re-run with a single change to Boulder’s
operating rules: an increase in fall/winter season treated water deliveries from the 63rd Street
WTF from the recent historical rate of 3.25 MGD to 8 MGD. The results are illustrated in Figure
3-11 and are included in Table 3-4. This change alone reduced the number of annual demand
reductions by nearly 50%.
Additional nonstructural and structural climate change adaptation and mitigation measures
should be explored including: (1) defining optimal methods for increasing production from the
63rd Street WTF to reduce drought effects while not unduly reducing hydropower generation, (2)
refining Boulder’s drought recognition and response guidelines to allow for slightly greater
drawdowns of its mountain reservoirs during extended droughts while still maintaining adequate
reserves, (3) making successive use of Boulder’s fully consumable water rights via recapture,
storage and exchange of Boulder’s fully consumable municipal return flows, and (4) exploring
options to developed or acquire additional storage via joint use agreements and/or purchase.
Figure 3-3: Boulder’s Water Demands and Deliveries and Mountain Storage, Baseline Scenario
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
1915 1921 1926 1931 1936 1941 1946 1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 2011 Treated Water Demands and Deliveries, Acre-feetCombined Barker and Wateshed Storage Contents, Acre-FeetTreated Water Demands, Deliveries and Mountain Storage Contents,
Baseline Scenario
Boulder Annual TW Demand before DRI Combined Barker and Watershed Storage Contents
Boulder Annual TW Deliveries after DRI
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 19
Attachment 3
Figure 3-4: Boulder’s Water Demands and Deliveries and Mountain Storage, Upper Right Scenario
Figure 3-5: Boulder’s Water Demands and Deliveries and Mountain Storage, 10-90 Scenario
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
1915 1921 1926 1931 1936 1941 1946 1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 2011 Treated Water Demands and Deliveries, Acre-feetCombined Barker and Wateshed Storage Contents, Acre-FeetTreated Water Demands, Deliveries and Mountain Storage Contents,
Upper Right Scenario
Boulder Annual TW Demand before DRI Combined Barker and Watershed Storage Contents
Boulder Annual TW Deliveries after DRI
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
1915 1921 1926 1931 1936 1941 1946 1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 2011 Treated Water Demands and Deliveries, Acre-feetCombined Barker and Wateshed Storage Contents, Acre-FeetTreated Water Demands, Deliveries and Mountain Storage Contents,
10/90 Scenario
Boulder Annual TW Demand before DRI Combined Barker and Watershed Storage Contents
Boulder Annual TW Deliveries after DRI
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 20
Attachment 3
Figure 3-6: Boulder’s Water Demands and Deliveries and Mountain Storage, 25-75 Scenario
Figure 3-7: Boulder’s Water Demands and Deliveries and Mountain Storage, Center Scenario
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
1915 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Treated Water Demands and Deliveries, Acre-feetCombined Barker and Wateshed Storage Contents, Acre-FeetTreated Water Demands, Deliveries and Mountain Storage Contents,
25/75 Scenario
Boulder Annual TW Demand before DRI Combined Barker and Watershed Storage Contents
Boulder Annual TW Deliveries after DRI
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
1915 1921 1926 1931 1936 1941 1946 1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 2011 Treated Water Demands and Deliveries, Acre-feetCombined Barker and Wateshed Storage Contents, Acre-FeetTreated Water Demands, Deliveries and Mountain Storage Contents,
Center Scenario
Boulder Annual TW Demand before DRI Combined Barker and Watershed Storage Contents
Boulder Annual TW Deliveries after DRI
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 21
Attachment 3
Figure 3-8: Boulder’s Water Demands and Deliveries and Mountain Storage, 75-25 Scenario
Figure 3-9: Boulder’s Water Demands and Deliveries and Mountain Storage, 90-10 Scenario
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
1915 1921 1926 1931 1936 1941 1946 1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 2011 Treated Water Demands and Deliveries, Acre-feetCombined Barker and Wateshed Storage Contents, Acre-FeetTreated Water Demands, Deliveries and Mountain Storage Contents,
75/25 Scenario
Boulder Annual TW Demand before DRI Combined Barker and Watershed Storage Contents
Boulder Annual TW Deliveries after DRI
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
1915 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Treated Water Demands and Deliveries, Acre-feetCombined Barker and Wateshed Storage Contents, Acre-FeetTreated Water Demands, Deliveries and Mountain Storage Contents,
90/10 Scenario
Boulder Annual TW Demand before DRI Combined Barker and Watershed Storage Contents
Boulder Annual TW Deliveries after DRI
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 22
Attachment 3
Figure 3-10: Boulder’s Water Demands and Deliveries and Mountain Storage, Lower Left Scenario
Figure 3-11: Boulder’s Water Demands and Deliveries and Mountain Storage, Lower Left Scenario,
with 8 MGD Winter Delivery from 63rd Street Water Treatment Facility
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
1915 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Treated Water Demands and Deliveries, Acre-feetCombined Barker and Wateshed Storage Contents, Acre-FeetTreated Water Demands, Deliveries and Mountain Storage Contents,
Lower Left Scenario
Boulder Annual TW Demand before DRI Combined Barker and Watershed Storage Contents
Boulder Annual TW Deliveries after DRI
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
100000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
1915 1920 1925 1930 1935 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Treated Water Demands and Deliveries, Acre-feetCombined Barker and Wateshed Storage Contents, Acre-FeetTreated Water Demands, Deliveries and Mountain Storage Contents,
Lower Left Scenario with 8 MGD Winter Delivery from 63rd Street WTP
Boulder Annual TW Demand before DRI Combined Barker and Watershed Storage Contents
Boulder Annual TW Deliveries after DRI
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 23
Attachment 3
Table 3-4: Drought Response Annual Reductions to Boulder’s Municipal Water Demand
(acre-feet)
climate
change
lower left
with 63rd
Street WTP
operational
change
climate
change
lower left
climate
change
90-10
climate
change
75-25
climate
change
center
climate
change
25-75
climate
change
10-90
climate
change
upper
right
historical
baseline
Average 3008 3178 2978 2357 2341 2309 2215 2254 2025
Maximum 5651 5651 5502 2579 2507 2482 2215 2254 2025
Maximum as a % of annual demand 22% 22% 22% 11% 11% 11% 10% 10% 9%
Number of years with reductions 11 21 10 6 3 3 1 1 1
Scenario
Annual Demand Reduction
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 24
Attachment 3
4.0 References
Colorado Water Conservation Board (CWCB). 2012. Colorado River Water Availability Study,
Phase I final report.
Lynker Technologies (Lynker). 2014. CRWAS Phase II Climate, Task 2, Climate Projections and
Hydrology Forcings, October 10, 2014.
Colorado Water Conservation Board (CWCB). 2015a. CRWAS Phase II Climate, Task 1,
Literature Review, September 8, 2015.
Colorado Water Conservation Board (CWCB). 2015b. CRWAS Phase II Climate, Task 1,
Approach for Constructing Climate Scenarios. September 8, 2015.
Livneh B., E.A. Rosenberg, C. Lin, B. Nijssen, V. Mishra, K.M. Andreadis, E.P. Maurer, and D.P.
Lettenmaier, 2013: A Long-Term Hydrologically Based Dataset of Land Surface Fluxes
and States for the Conterminous United States: Update and Extensions, Journal of
Climate, 26, 9384–9392.
Maurer, E.P., A.W. Wood, J.C. Adam, D.P. Lettenmaier, and B. Nijssen, 2002, A Long-Term
Hydrologically-Based Data Set of Land Surface Fluxes and States for the Conterminous
United States, J. Climate 15, 3237-3251.
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 25
Attachment 3
5.0 Appendix A
A comparison was completed between the Maurer and Livneh meteorological datasets from
1950 to 2014 for precipitation, minimum temperature and maximum temperature. The average
annual precipitation in millimeters (mm) is shown for 1/8-degree grid cells, which are used in the
VIC hydrology model. The two datasets are very similar, with the largest differences occurring
along the continental divide, at the headwaters of the St. Vrain and Colorado River watersheds
(Figure 5-1). The average annual minimum temperatures were very similar between the Maurer
and Livneh datasets, with the largest differences occurring at high elevation locations such as
the continental divide, including northernmost two cells (Figure 5-2). Similarly, the average
annual maximum temperatures for the Maurer and Livneh datasets were very similar, with the
largest differences occurring at the high elevation grid cells (Figure 5-3).
The Maurer and Livneh datasets were also compared in detail for two grid 1/8-degree grids
within the Boulder Creek watershed to ensure averaging four 1/16-degree Livneh grids was
comparable to using one 1/8-degree Maurer grid. The average 1/16-degree Livneh grid is
shown as a dotted blue line, which can be compared with the Maurer 1/8-degree grid shown as
a solid black line. The results for the upper Boulder watershed are provided for precipitation,
minimum temperature, maximum temperature and wind from 1949 to 2014 in Figure 5-4 through
Figure 5-7, respectively. The average annual values for the upper watershed are shown in
Table 5-1. Note that daily average wind values are used prior to 1949, thus the average annual
wind for 1915 to 1948 is static.
Table 5-1: Annual Average Upper Watershed Meteorological Values
Annual Average
Grid Cells Precipitation
(mm)
Maximum
Temperature
(° C)
Minimum
Temperature
(° C)
Wind (m/s)
1/16-degree average 730 10.0 -5.48 3.00
1/8-degree value 688 9.84 -5.61 3.08
The results for the lower Boulder watershed are provided for precipitation, minimum
temperature, maximum temperature and wind from 1949 to 2014 in Figure 5-8 through Figure
5-11, respectively. The average annual values for the lower watershed are shown in Table 5-2.
Table 5-2: Annual Average Lower Watershed Meteorological Values
Annual Average
Grid Cells Precipitation
(mm)
Maximum
Temperature
(°C)
Minimum
Temperature
(°C)
Wind (m/s)
1/16-degree average 398 18.5 2.70 3.17
1/8-degree value 395 18.5 2.62 3.22
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 26
Attachment 3
Figure 5-1: Maurer vs. Livneh Precipitation Comparison
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 27
Attachment 3
Figure 5-2: Maurer vs. Livneh Minimum Temperature Comparison
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 28
Attachment 3
Figure 5-3: Maurer vs. Livneh Maximum Temperature Comparison
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 29
Attachment 3
Figure 5-4: Upper Watershed Precipitation Comparison
Figure 5-5: Upper Watershed Minimum Temperature Comparison
-12.0
-10.0
-8.0
-6.0
-4.0
-2.0
0.0
Temperature (deg C)Annual Average Minimum Temperature
1/16-Grid 1 1/16-Grid 2 1/16-Grid 3
1/16-Grid 4 1/8-Grid 1/16-Grid Average
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 30
Attachment 3
Figure 5-6: Upper Watershed Maximum Temperature Comparison
Figure 5-7: Upper Watershed Wind Comparison
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
Temperature (deg C)Annual Average Maximum Temperature
1/16-Grid 1 1/16-Grid 2 1/16-Grid 3
1/16-Grid 4 1/8-Grid 1/16-Grid Average
2.0
2.2
2.4
2.6
2.8
3.0
3.2
3.4
3.6
3.8
Wind (m/s)Annual Averge Wind
1/16-Grid 1 1/16-Grid 2 1/16-Grid 3
1/16-Grid 4 1/8-Grid 1/16-Grid Average
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 31
Attachment 3
Figure 5-8: Lower Watershed Precipitation Comparison
Figure 5-9: Lower Watershed Minimum Temperature Comparison
-1.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
Temperature (deg C)Annual Average Minimum Temperature
1/16-Grid 4 1/16-Grid 3 1/16-Grid 2
1/16-Grid 1 1/8-Grid 1/16-Grid Average
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 32
Attachment 3
Figure 5-10: Lower Watershed Maximum Temperature Comparison
Figure 5-11: Lower Watershed Wind Comparison
15.0
16.0
17.0
18.0
19.0
20.0
21.0
22.0
Temperature (deg C)Annual Average Maximum Temperature
1/16-Grid 4 1/16-Grid 3 1/16-Grid 2
1/16-Grid 1 1/8-Grid 1/16-Grid Average
2.0
2.2
2.4
2.6
2.8
3.0
3.2
3.4
3.6
3.8
4.0
Wind (m/s)Annual Average Wind
1/16-Grid 4 1/16-Grid 3 1/16-Grid 2
1/16-Grid 1 1/8-Grid 1/16-Grid Average
10/15/2018 WRAB Agenda 4 - Water Supply & Climate Change
Page 33
Attachment 3