NYC Energy Water Use 2013 Report
NYC Energy Water Use 2013 Report
NYC Energy Water Use 2013 Report
ENERGY AND
WATER USE
2013 REPORT
AUGUST 2016
2 |
Copyright
2016 Urban Green Council. All rights reserved.
Disclaimer
None of the parties involved in the funding or creation of this
studyincluding Urban Green Council, its members, and its
contractorsassume any liability or responsibility to the user
or any third parties for the accuracy, completeness, or use of
or reliance on any information contained in the report, or for
any injuries, losses or damages (including, without limitation,
equitable relief) arising from such use or reliance. Although the
information contained in the report is believed to be reliable
and accurate, all materials are provided without warranties of
any kind, either express or implied, including but not limited
to warranties of the accuracy or completeness of information
contained, merchantability, or the fitness of the information for
any particular purpose.
As a condition of use, the user pledges not to sue and agrees
to waive and release Urban Green Council, its members, and
its contractors from any and all claims, demands, and causes
of action for any injuries, losses, or damages (including without
limitation, equitable relief) that the user may now or hereafter
have a right to assert against such parties as a result of the use
of, or reliance on, the report.
Executive Summary.........................................................................................................................................4
Background and Context.............................................................................................................................. 8
Understanding Energy Use and Improving
Energy Efficiency in Large Buildings..................................................................................................... 10
Year Four Benchmarking Data Results.................................................................................................32
Historical Comparisons...............................................................................................................................40
Policy Opportunities.................................................................................................................................... 48
Appendix...........................................................................................................................................................52
This is the City of New Yorks fourth report analyzing data collected
from Local Law 84 of 2009 and its first report analyzing data from
Local Law 87 of 2009. This report focuses on 2013 energy and water
usage reported in 2014. Both laws are part of the Citys Greener,
Greater Buildings Plan, designed to reduce greenhouse gas emissions
from New York Citys largest buildings.
The report was written and designed by Urban Green Council at the
direction of the Mayors Office of Sustainability. Urban Green Council
and New York Universitys Center for Urban Science and Progress
(NYU CUSP) performed the data analysis and developed the graphs
and charts included in this report. The organization responsible for
each graph or chart is listed in its caption.
The individual contributors from each group are included in the
reports Appendix.
4 |
0%
Building
GHG Emissions
% change
B
uilding Energy Use
(Weather Normalized
Source)
% change
-3%
-6%
-9%
2010
2011
2012
2013
Data Year
1
Local Law 84 requires energy and water benchmarking for all large buildings occupying at least 50,000 square feet, or properties with
multiple buildings that, together, encompass 100,000 square feet. City-owned properties larger than 10,000 square feet are also required
to benchmark each year, but are not included in this report, which focuses on the private sector.
The data displayed here represent only properties that have reported greenhouse gas emissions or weather normalized source energy use data
in all four benchmarked years. They represent approximately one-third of all benchmarked properties reporting in 2014 on 2013 use data. These
are some of the largest buildings in New York City and they have never been out of compliance with LL84. This result does not account for the
energy use and emissions reductions caused by Hurricane Sandy. Emissions calculations use EPA coefficients, not NYC-specific coefficients.
| 5
AUGUST 2016
10% Office
4% Other
3
The City of New York. (2016). Inventory of New York City Greenhouse Gas Emissions in 2014. Retrieved from www.nyc.gov/html/planyc/
downloads/pdf/NYC _GHG_Inventory_ 2014.pdf
4
This is based on areas calculated using the Department of City Planning (DCP)s Primary Land Use Tax Lot Output (PLUTO) database.
Guidelines for data quality are defined in Appendix B under the LL84 data cleaning methods.
6
A full-sized power plant is assumed to have a nameplate capacity of 1 gigawatt and operate at a 90 percent capacity factor. This is similar to
Unit 2 (1.032 GW) and Unit 3 (1.051 GW) at Indian Point Energy Center in Buchanan, New York. Together they produce 55 to 57 trillion Btus
annually at a capacity factor of 91 percent. www.safesecurevital.com/about-us/technical-overview.html. https://www.epa.gov/energy/egrid
6 |
Lighting
Other
Space Cooling
Domestic Hot Water
MULTIFAMILY
HOUSING
Ventilation
Elevators &
Escalators
Process Loads
OFFICE
0%
20%
40%
60%
80%
100%
7
This order has been determined by aligning LL84 electrical and fuel energy use with the proportion of each end use reported in energy
audits required under LL87.
| 7
AUGUST 2016
8
Urban Green Council & Steven Winter Associates. (2011). There are Holes in Our Walls. Retrieved from http://urbangreencouncil.org/sites/
default/files/there_are_holes_in_our_walls.pdf
9
Falk, L., Robbins, L., New York State Energy Research and Development Authority. (2010). Results from NYSERDAs Multifamily Performance
Programs:20% Reduction in Multifamily Buildings. Retrieved from http://aceee.org/files/proceedings/2010/data/papers/1958.pdf
8 |
| 9
Where appropriate in this report, data for individual building sectors are
further divided by their size: low-rise buildings with seven or fewer floors;
high-rise buildings with eight or more floors; and, very large buildings that
are larger than 500,000 square feet, regardless of the number of floors. The
division at seven floors is common in building analysis in New York City since
this is the height at which buildings need more complicated building systems,
such as elevators. The very large buildings tend to use more complicated
and centralized building systems. The systems used in each group varied
significantly within the LL87 data, and understanding those differences can
lead to the development of more targeted approaches and recommendations
for building system improvements.
Benchmarked data for individual buildings are public and available online.10
This report adds to the usability of this data by offering a comprehensive look
at benchmarking data as well as building energy data available to date. This
includes private energy audit data and other data not readily available online.
The analysis of this data provides:
10
New York City Mayors Office of Sustainability. Data Disclosure & Reports. Retrieved from www.nyc.gov/html/gbee/html/plan/
ll84_scores.shtml
10 |
11
Con Edison supplies steam to customers from the southern tip of Manhattan to 96th Street on the west side and 89th Street on the east
side: Con Edison. Con Edison Steam Energy Brochure. Retrieved from http://www.coned.com/steam/PDF/SteamEnergyBrochure.pdf
12
Process loads (e.g., kitchen refrigeration) consume energy in support of manufacturing or commercial process other than conditioning
spaces and maintaining occupant comfort. Conveyance loads (e.g., elevators and escalators) consume energy to move people and other
resources in buildings.
13
For lighting, auditors reported the fraction of floor area served by different lighting types. For all other systems, such as domestic
hot water and cooling, large buildings include base building systems that serve common needs, such as heat and hot water, as well as
smaller, tenant systems, such as window air conditioners and lighting, that serve tenant spaces. As expressed in Section 2, audits focused
on inspecting base building energy systems but did not consistently sample tenant space energy systems, thus limiting an accurate
representation of whole building energy use.
| 11
AUGUST 2016
FUEL TYPE
END USE
Space Heating
Electricity
Plug Loads/Misc
Lighting
Natural
Gas
Space Cooling
Domestic
Hot Water
Fuel Oil
District
Steam
Ventilation
Conveyance
Process Loads
Other
Figure 4: Flow of Fuel Types to End Use (LL84 and LL87 data) 1 4
Electricity represents more than half of the audited source energy, while space heating,
fueled mainly by natural gas, represents the largest end use. (Urban Green Council)
14
This chart has been created by aligning LL84 electrical and fuel energy use with the proportion of each end use reported in LL87. Natural
gas allocated to lighting, plug loads, and other electrical equipment generates electricity in cogeneration plants that are discussed later
in this report.
15
District steam is generated by utility plants and delivered to buildings through underground pipes, losing some of its energy in
transmission. The steam becomes liquid water condensate after use, but this hot condensate is not returned to the heating plant. Its energy
is lost unless a use is found at the building, affecting the efficiency of the overall system.
16
The modified Sankey diagram (Figure 6) displays the flow of fuel energy from its sources in the citys utilities, into the different sectors,
then into building boilers and furnaces, and finally, into building distribution systems. This diagram was developed using the combination
of fuel data from LL84 and system information from LL87, and covers only the energy used for heating in audited properties.
12 |
Figure 5: Source Energy Use Intensity by Energy Use Types and Property Type (LL87 data)
(height of bar is proportional to the total area of the property type). Source energy use intensity varies widely
by property type. (NYU CUSP)
Space Heating
Space Cooling
Lighting
Other
MULTIFAMILY HOUSING
OFFICE
WAREHOUSE
RETAIL
HOTEL
SENIOR CARE FACILITY
K-12 SCHOOL
RESIDENCE HALL/DORM
0%
20%
40%
60%
80%
100%
conditioner, moving heat from one place to another. An air conditioner works
on the principle that heat and pressure are linked within a contained volume.
The refrigerant fluid inside an air conditioner is forced to expand, a drop in
fluid pressure causes it to boil at room temperature, in a cycle that removes
heat from inside of the building. A heat pump can also work in reverse and
provide heat indoors. Baseboard heaters contain electric heating elements
enclosed by a metal pipe and fins to transfer heat to the surrounding air.
Because steam heat systems frequently run inefficiently, they offer abundant
opportunities for energy savings. Converting steam systems to hydronic systems
is effective, but expensive. However, significantly reducing fuel use in most steam
systems can be accomplished cost-effectively through operator training, system
maintenance, and better functioning controls. The relative EUIs and water use of
steam and hydronic systems are presented, below, in Figures 8 and 9.
The heat distribution systems introduced in the diagram on the next page can be
broken down further to analyze how different distribution technologies are used
in low-rise, high-rise, and very large audited buildings. Electric heat distribution
| 13
AUGUST 2016
Figure 6: How Large NYC Buildings are Heated (LL84 and LL87 data)
The flow of heating fuel energy through LL87 audited building sectors, heating
equipment, and distribution systems. (Urban Green Council)
FUEL
PROPERTY
TYPE
HEATING
SYSTEM
Natural
Gas
Multifamily Housing
DISTRIBUTION
SYSTEM
Steam Boiler
Steam
Distribution
Hydronic
Distribution
Oil
Office
District
Steam
Other
Electricity
District Steam
Electric Resistance
Rooftop Gas/PTAC/Furnace
Heat Pump
Furnace
Forced Air
Distribution
Electric
Distribution
14 |
Electric
Forced Air
LOW
RISE
25%
Hydronic
Vacuum Steam
HIGH
RISE
2-pipe Steam
1-pipe Steam
20%
OTHER
15%
VERY
VERY
LARGE
LARGE
VERY
VERY
LARGE
LARGE
10%
HIGH
RISE
5%
LOW
RISE
0%
MULTIFAMILY HOUSING
17
OFFICE
OTHER
As referenced in Section 2, the groups of buildings are defined as follows: very largebuildings 500,000 square feet and larger; high rise
buildings under 500,000 square feet with more than seven aboveground stories; and low risebuildings under 500,000 square feet with
seven aboveground stories or fewer.
| 15
AUGUST 2016
pipe steam systems are the least efficient heating systems. They are prevalent in
audited multifamily buildings and heat 37 percent of their floor area. Two-pipe
systems typically use less heating energy than one-pipe systems and
serve 25 percent of audited multifamily area.
Two other systems that require gas, oil, or district steamvacuum steam and
hydronic systemseach serve approximately 20 percent of the multifamily audited
area. A vacuum steam system is a two-pipe system that uses a vacuum pump to
enable even greater efficiency and control, while hydronic systems have separate
supply and return pipes to circulate hot water through radiators.
Combined benchmarking and auditing data show that the median EUI of
multifamily buildings using one-pipe steam heating systems is 13 percent higher
than the median EUI of buildings with other principal heating systems.1 8 Moreover,
the data show a wide distribution in energy use among one-pipe steam systems, as
shown below in the long upper tail in Figure 8. These variations indicate a number
of inefficiently run systems. The higher energy use found in some one-pipe steam
systems may be due to limited operation and maintenance improvements, which
minimize top efficiency, or other causes.1 9 In contrast, vacuum steam systems
160
140
120
100
80
60
40
20
0
1-PIPE STEAM
2-PIPE STEAM
HYDRONIC
VACUUM STEAM
18
To ensure quality data for this comparison, the audited building submissions were cleaned. Please refer to the Appendix for details on the
data cleaning methodology.
19
Energy Efficiency for All, Steven Winter Associates, & Natural Resource Defense Council. (2015). Clanging Pipes and Open Windows:
Upgrading NYC Steam Systems for the 21st Century. Retrieved from http://energyefficiencyforall.org/sites/default/files/EEFAUpgrading%20NYC%20Steam%20Systems.pdf
20
The shapes are a smoothed representation of the distributions of system-specific EUI, excluding electricity for pumps, burners, or controls,
and any fuel associated with domestic hot water. The width of each violin plot represents an estimate of the likelihood that a building with
a specific technology has a certain EUI. That is, based on the data sampled, buildings with two-pipe steam systems are most likely to have
a EUI of around 48 kBtu/sf. Stouter shapes means there is less variation in energy use. Long, narrow shapes mean some buildings are very
efficient and others are very inefficient. Median and quartile values for each distribution are also shown. The shapes are truncated at the
maximum and minimum reported values.
21
One-pipe steam systems are often found in audited buildings with high water usage.
(Urban Green Council)
200
180
16 |
160
140
120
100
80
60
40
20
0
1-PIPE STEAM
2-PIPE STEAM
HYDRONIC
VACUUM STEAM
21
The violin plot representation of the distribution of water use intensity (WUI) is shown in gallons per square foot per year for each
distribution system, and is truncated at the maximum and minimum reported values. Medians and quartiles are shown.
22
To independently verify that a majority of the buildings with high water consumption are being billed on flat-rate water and sewer accounts,
Ashokan Water Services linked the Borough, Block, and Lot (BBL) number found in LL84 and LL87 data to its records in DEPs Customer
Information System.
| 17
AUGUST 2016
hours (kWh) per square foot (Figure 10). 2 3 Audited hospitals follow as the
second-largest user of plug and miscellaneous loads, at a median of 5 kWh
per square foot, due to their intensive use of medical equipment. Audited
offices are the third-largest plug load users, with a median of 2.7 kWh per
square foot. 2 4 Absent other changes, that number will likely continue to rise
as worker density in office spaces increases. In fact, the average amount of
space per office worker in North America dropped by 22 percent from 2010
to 2012. 2 5 Should this trend continue, other measures of energy use, such as
energy per employee, may become more relevant.
4.0
3.0
2.0
1.0
A
B
SP
IT
A
L
K
IN /FI
ST NA
IT N
U C
TI IA
O L
N
E
IC
FF
O
IO
FA R
C CA
IL R
IT E
Y
M
E
O DI
FF C
IC AL
E
IL
TA
SE
H K
O -1
O 2
L
SC
TH
E
R
E
F
W RI
A G
R ER
E A
H T
O E
U D
SE
-R
H
W OU
O SE
R
SH O
IP F
M
U
LT
H IF
O A
U M
SI IL
N Y
G
H RE
A S
LL ID
/D EN
O C
R E
M
SE
0.0
W
A
5.0
23
The loads used by data centers and servers differ from typical plug loads, as they operate with an independent power supply. However, the
current auditing tool is not clear on how to report these loads and should be clarified.
24
This value is consistent with the plug load density found in a detailed 2012 office equipment study. Plug load use density ranged from 2.18
kWh/sf to 10.5 kWh/sf annually: Acker, B.; Duarte, C.; & Van Den Wymelenberg, K. (2012). ACEEE Summer Study on Energy Efficiency in
Buildings: Office Space Plug Load Profiles and Energy Saving Interventions. Retrieved from http://aceee.org/files/proceedings/2012/data/
papers/0193-000277.pdf
25
Barron, J. (2015). As Office Space Shrinks so Does Privacy for Workers. Retrieved from www.nytimes.com/2015/02/23/nyregion/as-officespace-shrinks-so-does-privacy-for-workers.html
18 |
lighting quality. The multifamily sector, which has the most inefficient
lighting installed, has the largest opportunity to save energy through
lighting improvements. In fact, 40 percent of the sectors audited area
is likely lit by older, lower-efficiency fluorescent lamps and incandescent
bulbs (Figure 11). 2 6 These low-efficiency lamps are used in nearly half of
low-rise multifamily audited buildings. And while almost 60 percent of the
office-sector audited area is already using higher-efficiency fluorescents,
there are still many opportunities for upgrades in high-rise and very large
office buildings. For instance, LED lights, the most energy efficient form of
interior lighting available, have not yet achieved widespread deployment
in the audited area, despite making inroads in every sector.
LIGHTING CONTROLS
Lighting systems with automated controls, including timers, occupancy
sensors, and daylight sensors, use substantially less electricity than
less-efficient, uncontrolled lighting does. These systems limit lighting to
times when it is needed. Most lighting systems in audited New York City
properties have no automatic controls (Figure 12). In fact, auditing data
show that only 10 percent of buildings have adopted automatic controls,
despite the substantial energy-saving potential they offer. Audit reports
for buildings did not distinguish manual switches, which must be turned
on and off, from lighting that is always on. Timed switches and occupancy
sensors make up the majority of the audited control systems currently in
use, so there remain many opportunities to use more advanced, energysaving systems that control lighting based on measured brightness and
daylight levels. These more advanced control systems improve lighting
quality and can reduce lighting energy use.
This lighting information comes with two important caveats. Though
LL87 was written so that energy audits include evaluations of lighting
technologies for fixtures and controls in both building common areas and
tenant spaces, the data represent mostly building common areas, the most
basic requirement for LL87, with not all audits sampling lighting fixtures in
tenant spaces. Therefore, the data show both kinds of lighting but focus
on common area lighting. Where auditors did not report lighting controls,
controls were assumed to be manual switches or lights were assumed to
be always on and included this way in the chart.
There are great opportunities for savings from lights that are always
on in common areas, hallways, stairwells, and other means of egress.
The low level of adoption of occupancy sensors in audited multifamily
buildings shows a significant opportunity for energy savings, as the use
of occupancy sensors in hallways and stairwells is widely regarded as
cost-effective. 2 7 Requirements for lighting controls in recent energy
codes and in Local Law 88 of 2009, which requires commercial spaces
in large buildings and common spaces in residential buildings to upgrade
their lighting to meet code by 2025, will likely lead to wider deployment
of these controls in both new and existing buildings.
26
Certain anomalies in the data indicate different auditors interpreted the forms used to collect lighting and other energy use information
differently, affecting the accuracy of the lighting data. For example, the low percentage of incandescent lighting recorded in multifamily
properties may indicate that many auditors did not include information on lighting use in apartments and, rather, recorded only commonarea lighting use. A process for improving these forms is proposed in the last section of this report.
27
McKinsey & Company. (2009). Unlocking Energy Efficiency in the U.S. Economy. Retrieved from https://www.energystar.gov/buildings/
tools-and-resources/unlocking-energy-efficiency-u-s-economy
| 19
AUGUST 2016
28
Offices are outpacing multifamily buildings in lighting retrofits. Lighting efficiency varies widely, with the most efficient
systems shown at the top of these bar graphs and the least efficient systems at the bottom. (Urban Green Council)
30%
Other
LOW
RISE
High-Intensity
Discharge
LED
HIGH
RISE
25%
OTHER
20%
Higher Efficiency
Linear
Fluorescent
(T-8 & T-5)
Compact
Fluorescent
Old Linear
Fluorescent
(T12)
VERY
LARGE
15%
Incandescent
VERY
LARGE
10%
HIGH
RISE
5%
LOW
RISE
0%
MULTIFAMILY HOUSING
OFFICE
OTHER
Photo/Daylight
Sensors
LOW
RISE
HIGH
RISE
20%
VERY
VERY
LARGE
LARGE
OTHER
Occupancy/
Vacancy
Sensors
Timer
Manual/Other
15%
VERY
LARGE
10%
HIGH
RISE
5%
LOW
RISE
0%
MULTIFAMILY HOUSING
28
OFFICE
OTHER
20 |
COOLING TECHNOLOGIES
Cooling uses the fourth-most energy in New York
Citys audited buildings, accounting for 11 percent of
the total source energy reported. Nearly 40 percent
of audited building area use window air conditioners,
through-the-wall cooling units, or packaged terminal
air conditioners (PTACs). These are self-contained
units with vents and other equipment that go through
walls (Figure 13). These systems serve individual
rooms and are often poorly sealed, meaning the
systems leak heat in the winter and allow heat into
conditioned spaces in summer. These systems serve
the majority of low-rise multifamily building audited
area and half the high-rise multifamily audited area.
Many audited office buildings and some very large
multifamily audited buildings use central systems that
serve either zones or entire buildings. These systems
are built into the building infrastructure and serve the
spaces they cool at substantially higher efficiency.
Central plant chillers are used for cooling in half
of the audited area in very large office buildings.
These plants chill water for distribution throughout
a building. The water then enters air-handling units
to absorb heat from ventilation air, thus cooling the
30%
Other
HIGH
HIGH
RISE
RISE
Absorption
Chiller
Central Chiller
25%
DX Units
OTHER
Split System
Packaged
System
20%
HIGH
HIGH
RISE
RISE
15%
LOW
RISE
P TAC/Through
Wall A/C
VERY
LARGE
Window A/C
VERY
LARGE
10%
5%
LOW
RISE
0%
MULTIFAMILY HOUSING
OFFICE
OTHER
29
The areas shown in Figure 13 represent the area cooled from the audit data rather than the whole building area. Often, auditors did not
report area cooled if the primary cooling system was tenant-based, such as window or through-wall air conditioners, underestimating
their prevalence in the data presented.
30
Absorption chillers have reached a coefficient of performance (COP, a measurement of efficiency that compares the amount of power
input to a system to the amount of power output) of 1.6: Murakami, S.; Levine, M.; Yoshino, H.; Inoue, T.; Ikaga, T.; Shimoda, Y.; et al. (2006).
Energy Consumption, Efficiency, Conservation, and Greenhouse Gas Mitigation in Japans Building Sector. Lawrence Berkeley National
Laboratory. Retrieved from http://escholarship.org/uc/item/6gp873s1
| 21
AUGUST 2016
25%
LOW
RISE
Other
Instantaneous
(Point of Use)
HIGH
RISE
20%
VERY
LARGE
Separate Hot
Water Boiler with
Storage Tank
OTHER
Direct Fired
Storage Tank
Heat Exchanger
from Heating
System (Indirect)
15%
VERY
LARGE
Tankless Coil
10%
HIGH
RISE
5%
LOW
RISE
0%
MULTIFAMILY HOUSING
OFFICE
OTHER
31
Ben-Nathan, O. (2015). Daikin, Mitsubishi, Toshiba VRV / VRF / Split Air- Conditioning Systems - Seamless Integration and Remote Control
Operation. Retrieved from http://homeenergypros.lbl.gov/profiles/blogs/daikin-mitsubishi-toshiba-vrv-vrf-split-air-conditioning-systems
32
The intermittent cycling of boilers during summer months wastes energy: Intermittent Combustion and Boiler Efficiency. Retrieved from
www.engineeringtoolbox.com/intermittent-boiler-efficieny-d_1133.html
22 |
VENTILATION
Ventilation systems consume 6 percent of audited energy. However, most of this
energy consumption occurs in office buildings, as audits of 46 percent of the
multifamily building area did not report a ventilation system or declared Other.
In residential buildings, building code requires ventilation only in both kitchens
and bathrooms that have no windows.
Most modern offices are not designed to be ventilated naturally and therefore
require mechanical ventilation. 33 The audit data show that 98 percent of office
area uses ventilation equipment, while nearly half of this area is served by airhandling units (AHU), which move, filter, and heat or cool air. In 29 percent of
the total ventilated office area, AHUs were combined with an economizer, which,
when outdoor temperatures allow, turns off air conditioning and blows cool air
directly into a buildings interior. This type of ventilation typically occurs most
often in the spring and autumn, when buildings can use outside air to dissipate
the internal heat that people, lights, and electrical equipment generate.
Less than 1 percent of multifamily area is ventilated using an energy-saving
device called an energy- or heat recovery ventilator (ERV/HRV), though this
number is growing. 3 4 HRVs recover heat from exhaust air before it is released
outdoors, while ERVs recover both heat and moisture from exhaust air. Energyefficient dedicated outdoor air systems (DOAS), which bring in outdoor air but
do not condition the space, often incorporate ERVs. These systems appear in
2 percent of commercial building audited area and 5 percent of multifamily
building audited area.
33
Wood, A.; Salib, R.. (2012). Natural Ventilation in High-Rise Office Buildings: An output of the CTBUH Sustainability Working Group.
Retrieved from https://store.ctbuh.org/PDF_Previews/Reports/2012_CTBUHNaturalVentilationGuide_Preview.pdf
34
There were some inconsistencies in whether the ventilation system was described as one or multiple system types.
| 23
AUGUST 2016
25%
Other
LOW
LOW
RISE
RISE
Wood-Framed
Metal Building
HIGH
RISE
20%
VERY
VERY
LARGE
LARGE
15%
OTHER
OTHER
Steel-Framed
Mass
VERY
VERY
LARGE
LARGE
10%
HIGH
RISE
5%
LOW
RISE
0%
OFFICE
MULTIFAMILY HOUSING
OTHER
50%
Other
Single Pane
VERY
LARGE
Double Pane
40%
HIGH
RISE
HIGH
RISE
30%
VERY
LARGE
LOW
RISE
LOW
RISE
OTHER
20%
10%
0%
MULTIFAMILY HOUSING
OFFICE
OTHER
24 |
Figure 17: Relationship Between Energy Use Intensity and Glazing Fraction (LL84 and LL87 data)
Most of the audited buildings that have a glazing ratio larger than 50 percent were built after 1990. (Urban Green Council)
Before 1990
300
After 1990
250
200
150
100
50
0
0.0
0.2
0.4
0.6
0.8
1.0
Glazing Fraction
35
Wilson, A. (2010). Rethinking the All-Glass Building. BuildingGreen. Retrieved from https://www.buildinggreen.com/feature/rethinkingall-glass-building
| 25
AUGUST 2016
No Data
LOW
RISE
20%
15%
M aster
Meter Only
HIGH
RISE
Sub-Metering
VERY
LARGE
OTHER
Direct Metering
VERY
LARGE
10%
HIGH
RISE
5%
LOW
RISE
0%
MULTIFAMILY HOUSING
OFFICE
OTHER
with the predictions of models, further study using more LL87 data or data from other
cities to explore cooling energy use and other factors such as age or effective energy
code at time of construction, will be helpful.
36
New York State Energy Research and Development Authority (NYSERDA). (1997 revised 2001). Residential Electric Submetering Manual.
Retrieved from www.submeteronline.com/pdf/subman2001.pdf
M ultifamily Housing
O ffice
9.3%
ENVELOPE
6.6%
17.2%
LIGHTING
18%
2.6%
MOTORS
26 |
6.9%
1.3%
VENTILATION
4.2%
8.2%
DISTRIBUTION
SYSTEM
6%
0.6%
COOLING
SYSTEM
7.8%
7%
HVAC CONTROLS
& SENSORS
8.4%
14.2%
DOMESTIC
HOT WATER
4%
4.6%
HEATING
SYSTEM
5.5%
2.1%
FUEL
SWITCHING
2.6%
0%
5%
10%
15%
37
Universities participating in the NYC Carbon Challenge have already achieved a 17% reduction in carbon emissions across hundreds of
buildings: New York City Mayors Office of Sustainability. The NYC Carbon Challenge for Universities Retrieved from www.nyc.gov/html/
gbee/html/challenge/universities.shtml
20%
AUGUST 2016
38
Deutsche Bank CSR, Living Cities, HR&A Advisors, & Steven Winter Associates. (2012). The Benefits of Energy Efficiency in Multifamily
Affordable Housing. Retrieved from http://energyefficiencyforall.org/sites/default/files/DBLC _Recognizing_the_Benefits_of_Efficiency_
Part_B_1.10%20%281%29.pdf
39
SF Environment & ULI Greenprint Center for Building Performance. San Francisco Existing Commercial Buildings Performance Report
2010-2014. Retrieved from http://sfenvironment.org/sites/default/files/fliers/files/sfe_gb_ecb_performancereport.pdf
| 27
28 |
| 29
AUGUST 2016
Figure 21: Estimated Site Energy Savings per Square Foot for Each Energy Conservation Measure (LL87 data)
Audited offices could realize the greatest savings per square foot from on-site generation and ventilation, while audited multifamily
buildings greatest savings could come from HVAC controls. (NYU CUSP)
0.2
0.3
0.2
COOLING SYSTEM
M=55 O=95
0.8
PROCESS &
PLUG LOADS
M=56 O=4
0.3
0.3
DISTRIBUTION
SYSTEM
M=715 O=56
O ffice
0.2
MOTORS
M=245 O=74
M ultifamily Housing
0.1
LIGHTING
M=3001 O=232
0.5
0.2
VENTILATION
M=112 O=41
0.5
CONVEYING
SYSTEMS
M=87 O=24
0.5
2.7
1.5
1.0
OTHER
M=33 O=6
0.2
1.4
ENVELOPE
M=1081 O=69
0.6
ON-SITE
GENERATION
M=137 O=9
1.5
4.1
2.1
FUEL SWITCHING
M=175 O=21
1.3
DOMESTIC
HOT WATER
M=1453 O=35
2.3
0.3
2.4
HEATING SYSTEM
M=403 O=46
1.6
3.3
SUB-METERING
M=28 O=0
HVAC CONTROLS
& SENSORS
M=627 O=95
3.9
1.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
is not influenced by the expected life of the ECM, financing and other business
requirements, expected changes in ownership, confidence in ECM cost- and
energy-saving projections, and other factors. This data, like other data reported
here, may be of more value to government and private utility program planners
than to building operators, although it does offer a general guide to areas that
offer the quickest return on investment.
40
Deutsche Bank/Living Cities, HR&A Advisors, & Steven Winter Associates. (2011). Deutsche Bank/Living Cities: Building Energy Efficiency
Data Report. Retrieved from www.swinter.com/8fbf625d-309f-4d7c-ba33-a390399aec68/resources-research-guidelines-research.htm
30 |
SUB-METERING
M=28 O=0
O ffice
LIGHTING
M=3001 O=232
DISTRIBUTION SYSTEM
M=715 O=56
FUEL SWITCHING
M=175 O=21
VENTILATION
M=112 O=41
HEATING SYSTEM
M=403 O=46
MOTORS
M=245 O=74
PROCESS & PLUG LOADS
M=56 O=4
COOLING SYSTEM
M=55 O=95
ENVELOPE
M=1081 O=69
ON-SITE GENERATION
M=137 O=9
0
10
15
20
25
41
At $1.00 per therm of gas and $0.20 per kWh of electricity, the breakeven cost for a five-year payback is $0.05/kBtu for gas and $0.29/
kBtu for electricity: New York State Energy Research and Development Authority (NYSERDA). Energy Prices and Weather Data. Retrieved
from https://www.nyserda.ny.gov/Cleantech-and-Innovation/Energy-Prices
| 31
AUGUST 2016
Figure 23: Cost per Amount of Energy Saved for Each Energy
Conservation Measure (LL87 data)
At current prices, Energy Conservation Measures (ECMs) that save electricity are
more financially attractive than those that save natural gas. (NYU CUSP)
GAS
COST
DISTRIBUTION SYSTEM
M=715 O=56
M ultifamily Housing
ELECTRICITY
COST
O ffice
HEATING SYSTEM
M=403 O=46
VENTILATION
M=112 O=41
OTHER
M=33 O=6
LIGHTING
M=3001 O=232
SUB-METERING
M=28 O=0
ENVELOPE
M=1081 O=69
MOTORS
M=245 O=74
PROCESS & PLUG LOADS
M=56 O=4
COOLING SYSTEM
M=55 O=95
FUEL SWITCHING
M=175 O=21
ON-SITE GENERATION
M=137 O=9
CONVEYING SYSTEMS
M=87 O=24
0.0
0.5
1.0
1.5
2.0
2.5
The fact that only a small number of ECM categories achieve the break-even
point for natural gas savings is the result of current, low natural gas prices. 42
Residential gas prices are no higher than prices in 1980 if the cost comparison
is adjusted for inflation. As with all measures, some of the ECMs will be more
energy and cost-effective than the category average, and some will be less. The
values displayed are general summaries over several ECMs; individual buildings
will require analyses of their options.
42
U.S. Energy Information Administration. (2016). Natural Gas Prices. Retrieved from https://www.eia.gov/dnav/ng/ng_pri_sum_dcu_
nus_m.htm
3.0
32 |
BUILDING CHARACTERISTICS
100%
10%
90%
100%
Other
Office
90%
M ultifamily
Housing
80%
24%
70%
60%
50%
40%
67%
30%
70%
College/University
5%
Mixed Use
Property
5%
Hospital (General
Medical & Surgical)
8%
Retail Store
50%
14%
40%
10%
0%
0%
Financial Office
Senior Care
Community
Non-Refrigerated
Warehouse
All Other Types
30%
10%
Self-Storage
Facility
8%
8%
20%
K-12 School
5%
60%
20%
43
4%
5%
80%
38%
| 33
AUGUST 2016
SIZE
The majority of benchmarked multifamily and office properties have a floor area
between 50,000 square feet, which is the benchmarking minimum, and 100,000
square feet (Figure 26). This is particularly true for multifamily buildings. More than
half of these properties are smaller than 100,000 square feet. Office buildings,
by contrast, are often very large, with more than one in three office buildings
occupying more than 300,000 square feet. (Fewer than one in 12 multifamily
buildings encompass 300,000 square feet.) Additionally, almost 10 percent of
office properties encompass floor areas greater than 1,000,000 square feet.
Figure 26: Number of Benchmarked Properties by Gross Floor Area (LL84 data)
M ultifamily Housing
Most benchmarked buildings are smaller than 100,000 square feet. (NYU CUSP)
O ffice
3885
100 AND
LESS
243
1893
100-200
218
519
200-300
102
198
300-400
74
120
400-500
45
59
500-600
34
33
600-700
30
13
700-800
22
17
800-900
23
8
900-1,000
15
25
OVER
1,000
83
500
1000
1500
2000
2500
Number of Properties
3000
3500
4000
M ultifamily
Housing
M
ultifamily
Housing Median
Building booms occurred in the 1920s, 1960s, and 2000s. (NYU CUSP)
O ffice
O
ffice Median
180
350
160
300
140
250
120
200
100
80
150
60
100
40
50
20
0
Pre1900
1900
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
2010
Decade Built
AGE
A buildings age is often a useful indicator of its energy use, as shown in
Section 5. To that end, construction trends help show when much of New York
Citys benchmarked building stock was built. For example, both the office and
multifamily building sectors experienced booms in the 1920s and 1950s-60s
(Figure 27). More recently, in the 2000s, multifamily construction has been the
primary driver of growth in floor area and number of buildings. Multifamily and
office properties built in the 1960s and 1970s also appear to be larger than those
built earlier in the 1920s. Though a similar amount of floor area was built during
both eras, the number of buildings built in the 1960s and 1970s is nearly half that
built in the 1920s.
BUILDING USE
Another common characteristic of some New York City buildings that can
impact energy use is mixed building use. For example, some multifamily
buildings include retail stores or banks on the first floor, while offices and
industrial spaces sometimes share buildings. In 2013, owners of nearly 75
percent of multifamily buildings and 40 percent of office buildings reported
their buildings were single-use (Figure 28). And though some large buildings
housed as many as five or more uses, the majority of the remainder were only
used for one additional purpose.
34 |
| 35
AUGUST 2016
100%
4 Uses
90%
Percentage of Properties
80%
3 Uses
2 Uses
1 Use
70%
60%
50%
40%
30%
20%
10%
0%
MULTIFAMILY
HOUSING
OFFICE
36 |
Figure 29: Median Energy Use Intensity by Property Type (LL84 data)
(area of circle is proportional to energy consumed). Benchmarked multifamily buildings consume the most
energy overall, while benchmarked supermarkets use the most energy per square foot. (NYU CUSP)
# = Median
600
Supermarket / Grocery Store 599
Hospital (General Medical & Surgical) 547
500
Hotel 225
300
200
195
OFFICE
MULTIFAMILY
HOUSING
125
Non-refrigerated Warehouse 57
Distribution Center 56
Self-storage Facility 47
0
0
1000
2000
3000
4000
5000
6000
7000
8000
Number of Properties
M ultifamily Housing
Large offices tend to be more energy intensive than large multifamily properties. (NYU CUSP)
O ffice
Median
EUI
125
Median
EUI
125
Median
EUI
195
Median
EUI
195
1600
1600
120
120
1400
1400
100
100
Frequency of Properties
1200
1200
8080
1000
1000
6060
800
800
600
600
40
40
400
400
2020
200
200
00
00
200
200
400
400
600
600
800
800
1000
1000
00
00
200
200
400
400
600
600
800
800
1000
1000
| 37
AUGUST 2016
Figure 31: Distribution of New York City 2013 Office ENERGY STAR Scores (LL84 data)
New York Citys median ENERGY STAR score, 74, surpasses the national median of 50. (NYU CUSP)
Percentage
of Offices
Reporting: 84%
25
Median Score 74
20
15
10
0
0
20
40
60
80
100
44
U.S. Energy Information Administration. Commercial Buildings Energy Consumption Survey (CBECS). Retrieved from www.eia.gov/
consumption/commercial/
45
U.S. Energy Information Administration. (2009). Residential Energy Consumption Survey (RECS). Retrieved from www.eia.gov/
consumption/residential/
46
ENERGY STAR scores for multifamily properties will be reported in the 2014 New York City benchmarking data and will significantly
increase the number of properties that receive ENERGY STAR scores: U.S. Environmental Protection Agency. ENERGY STAR Score for
Multifamily Housing in the United States. Retrieved from https://www.energystar.gov/buildings/tools-and-resources/energy_star_score_
multifamily_housing_united_states.
38 |
Figure 32: Distribution of Water Use Intensity by Benchmarked Property Type (LL84 data) 5 0
The hospital sector is the most water intensive, while the multifamily sector is the largest user of water overall. (Urban Green Council)
350
Median WUI
250
200
150
100
50
U
N
/C IV
O ER
LL S
EGIT Y
E
E
IC
FF
O
R
ST ET
O AI
R L
E
SI
D
/D E
O NC
R E
M H
IT A
O L
R L
Y
LT
H IF
O A
U M
SI IL
N Y
G
M
E
O DI
FF C
IC AL
E
L
TE
O
H
SE
C NI
O O
M R
M C
U A
N R
IT E
Y
SP
IT
A
300
performing well above the national median of 50 (Figure 31). New York City
has a lower median score than other major cities such as San Francisco (87),
Washington, D.C. (79), and Boston (78), though its median score is higher
than Philadelphias (64). 47 It is possible that the methodology that ENERGY
STAR uses to calculate this score does not account for New York Citys unique
building size and density. Therefore ENERGY STAR has potential limitations in
predicting building energy use in New York City. 4 8
47
SF Environment & ULI Greenprint Center for Building Performance. San Francisco Existing Commercial Buildings Performance Report
2010-2014. Retrieved from http://sfenvironment.org/sites/default/files/fliers/files/sfe_gb_ecb_performancereport.pdf. Greenovate
Boston. (2015). Energy and Water Use in Bostons Large Buildings, 2013Retrieved from http://www.cityofboston.gov/images_documents/
BERDO_rprt_webfinal_tcm3-52025.pdf
48
Kontokosta, C. E. (2014). A Market-Specific Methodology for a Commercial Building Energy Performance Index. The Journal of Real Estate
Finance and Economics J Real Estate Finan Econ, 51(2), 288-316. Retrieved from http://link.springer.com/article/10.1007/s11146-014-9481-0
49
50
The width of each violin plot approximates the number of properties with that water use intensity in that sector.
AUGUST 2016
Of the two largest water consumers, the multifamily sector should remain the
focus for efforts to reduce potable water use. In addition to consuming almost
eight times more water in total than office buildings, multifamily properties are
also more water intensive, with a per-square-foot water footprint three times
greater than that of office buildings. The multifamily sector also includes the
most water-intensive benchmarked properties, as evidenced by the tall tail
in Figure 32, below. Possible explanations for these large water users include
operation and maintenance issues, building owners paying a flat rate for water
and thus not being aware of their buildings usage, or other causes.
| 39
40 |
MULTIFAMILY
HOUSING
1.2
2013
2012
1.0
2011
0.8
2010
0.6
Percentage of Properties
0.4
0.2
0
0
100
200
300
400
500
600
0.7
OFFICE
2013
0.6
2012
0.5
2011
2010
0.4
0.3
0.2
0.1
0
0
100
200
300
400
500
600
| 41
AUGUST 2016
in multifamily energy use, but the majority of the properties are still within
the same range of EUIs they first reported in 2010 (Figure 33). By contrast,
the distribution for office properties shows more significant reductions. The
longer tail in the office distribution in Figure 33, below, indicates that there are
buildings with higher energy intensities that might result from longer operating
hours, higher densities of workers or technology, or other causes. Yet even this
tail is gradually flattening, suggesting that high users of energy have also been
reducing their energy use.
Overall energy use for these sectors illustrates these trends more clearly
(Figure 34). The median energy use intensity for multifamily properties has
dropped 5 percent over four years, while the median for offices has dropped
by 11 percent. In 2013, the median ENERGY STAR score for offices reporting
consistently in all four years was 78, four points higher than the score for office
properties reporting only in 2013 (Figure 31). This represents a 12 percent
increase from their 2010 median score and shows that half of the consistently
reporting office buildings qualified for ENERGY STAR certification, which
requires a score of 75 or above.
In addition to more permanent improvements such as operator training and
ECMs, temporary factors could have contributed to this reduction in energy
use. The most notable was Superstorm Sandy in late October of 2012, which
caused many buildings in the inundation areas to be unoccupied for periods
of time. Among benchmarked properties, this had a greater impact on office
properties and may partially account for the greater reductions seen in that
sector. An analysis in the previous benchmarking report that looked at a
100
80
200
60
150
40
100
20
50
0
2010
2011
2012
Data Year
2013
250
42 |
Natural Gas
Fuel Oil #2
Fuel Oil #4
Fuel Oil #5 & 6
80%
26%
30%
35%
73%
68%
60%
58%
49%
40%
35%
20%
0%
2010
2011
2012
2013
Data Year
51
The City of New York. (2014). New York City Local Law 84 Benchmarking Report, September 2014. Retrieved from www.nyc.gov/html/
planyc/downloads/pdf/publications/2014_nyc_ll84_benchmarking_report.pdf
52
For more information on the NYC Clean Heat Program and the NYC Retrofit Accelerator, visit the Citys website: www.nyc.gov/
retrofitaccelerator.
53
New York City Mayor's Office of Sustainability. (2016). Heating Oil Regulations. Retrieved from http://www.nyc.gov/html/gbee/html/
codes/heating.shtml
| 43
AUGUST 2016
Figure 36: Median Energy Use Intensity by Gross Floor Area for Consistently
Benchmarked Properties (LL84 data)
Larger benchmarked office buildings use more energy per square foot. However, building
size does not appear to have an impact on year-to-year performance. (NYU CUSP)
350
MULTIFAMILY
HOUSING
2013
300
2012
2011
250
2010
OFFICE
200
2013
2012
150
2011
2010
100
50
0
100 AND
LESS
100200
200300
300400
400500
500600
600700
700800
800900
9001,000
OVER
1,000
54
Durst, A. Council on Tall Buildings and Urban Habitat (CTBUH). (2015). Efficient Energy Production for High-Demand Tenants of Tall
Buildings. Retrieved from http://global.ctbuh.org/resources/papers/download/2436-efficient-energy-production-for-high-demandtenants-of-tall-buildings.pdf
large buildings use more energy per square foot than mid-size buildings. More
analysis needs to be done on large office buildings and additional metrics are
needed to account for their increased size and energy usage.
The office sector also shows a more dramatic incline in energy use intensity
in buildings built in the 1960s and 1970s. Given that these buildings use
comparable lighting and equipment, such as computers, this dramatic change
in EUI might result from the type of cooling system used, as central cooling
became more common then; from changes in envelope construction; or,
from other reasons. In addition, buildings constructed more recently might
have characteristics such as higher worker densities or longer occupancy
hours. 55 The decline in EUIs after the 1990s may reflect the impact of more
stringent energy codes and greater enforcement efforts by the Department
of Buildings (DOB). These possibilities may become substantiated as more
data on equipment and building characteristics continues to be gathered
under LL87. Since buildings built after the 1960s use approximately 25
percent more energy per square foot than those built before, this question
deserves further study.
Another factor related to a buildings age is its fuel mix (Figure 4). Properties
built in the 1980s and before show greater use of heavy fuel oils#4, #5, and
#6 (Figures 38, 39). This is particularly true in multifamily properties, for which
fuel represents up to 43 percent of energy use. (That compares with only 18
Figure 37: Median Weather Normalized, Source
Energy Use Intensity by Decade Built (LL84 data)
M
ultifamily
Housing Median
O
ffice Median
350
44 |
300
250
200
150
100
Pre1900
1900
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
Decade Built
55
Durst, A. Council on Tall Buildings and Urban Habitat (CTBUH). (2015). Efficient Energy Production for High-Demand Tenants of Tall
Buildings. Retrieved from http://global.ctbuh.org/resources/papers/download/2436-efficient-energy-production-for-high-demandtenants-of-tall-buildings.pdf
2010
| 45
AUGUST 2016
Electricity
District Steam
Natural Gas
Fuel Oil #2
80
Fuel Oil #4
Fuel Oil #5 & 6
60
40
20
0
Pre1900
1900
1910
1920
1930
1940
1950
1960
1980
1970
Decade Built
Electricity
District Steam
Natural Gas
Fuel Oil #2
80
Fuel Oil #4
Fuel Oil #5 & 6
60
40
20
0
Pre1900
1900
1910
1920
1930
1940
1950
1960
Decade Built
1970
1980
1990 2000
2010
46 |
percent for office buildings.) This fact reinforces the findings from the LL87
data presented earlier in Figure 3, as higher oil and natural gas demand in
the multifamily sector reflects the sectors substantial space heating needs.
Meanwhile, electricity dominates office properties fuel use, due to their
higher demand for plug and process loads and cooling.
A buildings date of construction is also an indicator of its water use. Here
again, trends in multifamily and office properties differ. Compared to multifamily
buildings built before 1900, the median water use intensity in those buildings is
56 percent higher in buildings built between the 1940s and the 1970s. Water use
steadily drops in more recently constructed, benchmarked multifamily buildings,
with buildings built in the 2010s using 53 percent less water than those buildings
constructed in the 1970s (Figure 40). This may be due to increased stringency
in federal standards, as well as increased awareness of water conservation. 5 6 In
contrast, the WUI values of office buildings have been relatively stable over time,
at around 20 gal/sf except for buildings built during the 1990s, for which water use
intensity spikes to 36 gal/sf. One possible explanation for this increase is the use of
once-through cooling systems that use potable water to cool condensers and then
dispose of it. As further data is collected under LL87, this should be investigated.
Number of Properties
M ultifamily Housing
M
ultifamily Housing Median
O ffice
O
ffice Median
700
70
600
60
500
50
400
40
300
30
200
20
100
10
0
Pre1900
1900
1910
1920
1930
1940
1950
1960
1970
1980
1990
2000
2010
Decade Built
56
Gleick, P. H. (2013). The Worlds Water: The biennial report on freshwater resources (Vol. 7). Chapter 7: U.S. Water Policy Reform.
Retrieved from http://worldwater.org/wp-content/uploads/sites/22/2013/07/chapter_7_us_water_policy_reform.pdf
The similarities between the chart shown here and that in the Citys 2014 water data
benchmarking report suggest that water data reporting is fairly consistent, bearing
out DEPs efforts to improve water data accuracy. (See sidebar in Section 4).
48 |
Since 73 percent of New York Citys carbon emissions come from energy used in
buildings, benchmarking and audit data are crucial to helping the City craft impactful,
data-driven programs and policies that will help achieve its goal of reducing citywide
carbon emissions by 80 percent by 2050 (80 x 50). 5 7 Other cities will also look to
New York City as a model for initiating studies at this level. Next steps toward reducing
building energy use, improving the collection and quality of building energy data, and
facilitating the use of that data include the following:
57
The City of New York. (2016). Inventory of New York City Greenhouse Gas Emissions in 2014. Retrieved from www.nyc.gov/html/planyc/
downloads/pdf/NYC _GHG_Inventory_ 2014.pdf
AUGUST 2016
58
This is an open source software application created to validate, clean, and manipulate building performance data. For more information,
please see DOEs website: http://energy.gov/eere/buildings/standard-energy-efficiency-data-platform.
59
This percentage was calculated using the Department of City Planning (DCP)s Primary Land Use Tax Lot Output (PLUTO) database.
| 49
50 |
IMPROVEMENTS TO DATE
New York City has made great strides over the past few years to improve data
quality, accessibility, and building energy use, including:
DATA QUALITY
Department of Buildings Data Quality Enforcement
In March 2016, in an effort to improve data reporting, DOB issued a service notice
informing building owners that benchmarking submissions will be reviewed for
completeness and accuracy using a nine point review process. Building owners
who submit data before the May deadline will have an opportunity to have their
data reviewed by DOB and receive confirmation of whether their reports need
to be corrected and re-submitted.
ACCESSIBILITY
Two web-based tools have been developed to make it easier to access building
level benchmarking data.
60
Kontokosta, Constantine E. and Christopher Tull. 2016. EnergyViz: Web-Based Eco-Visualization of Urban Energy Use from Building
Benchmarking Data, Computing in Civil and Building Engineering
AUGUST 2016
Brooklyn office properties built before 1950 are the bestor worstperformers.
More than just a repository of performance data, Metered is designed to educate
users about the factors underlying property statisticsand more importantly,
how to improve them through training and incentives. Metered also provides
tips and resources alongside relevant property report cards, so that users can
see which upgrades are available at the moment they are discovering these
buildings need them. Metered is available at www.metered.nyc.
BUILDING PERFORMANCE
NYC Retrofit Accelerator
In September 2015, Mayor de Blasio launched the NYC Retrofit Accelerator,
a program that provides free technical assistance and advisory services to
building owners undertaking critical energy efficiency, water conservation,
and clean energy upgrades. The program is anticipated to reduce citywide
greenhouse gas emissions by roughly one million metric tons per year by
2025the equivalent of taking almost 200,000 passenger vehicles off the
roadsby accelerating retrofits in as many as 1,000 properties a year by
2025. These retrofits can save New Yorkers an estimated $350 million a year
in utility costs, all while generating more than 400 local construction-related
jobs. In addition to helping building owners reduce operating costs, raise asset
values, and improve occupant comfort, the Retrofit Accelerator assists in LL87
compliance by clarifying the requirements, providing the updates, suggesting
compliance timelines, and more. Additional information is available at www.
nyc.gov/retrofitaccelerator.
| 51
52 |
APPENDIX A
OVERVIEW OF METHODOLOGY
This section presents the basic terms and approaches used in this report
to analyze energy use and characteristics in New York Citys buildings, as
reported under Local Law 84 and Local Law 87. The terms referenced in
this report are as follows:
61
U.S. Environmental Protection Agency. (2013). ENERGY STAR Portfolio Manager Technical Reference: Source Energy. Retrieved from
https://portfoliomanager.energystar.gov/pdf/reference/Source%20Energy.pdf
62
The City of New York. (2016). Inventory of New York City Greenhouse Gas Emissions in 2014. Retrieved from www.nyc.gov/html/planyc/
downloads/pdf/NYC _GHG_Inventory_ 2014.pdf
63
U.S. Environmental Protection Agency. ENERGY STAR Portfolio Manager. Retrieved from https://www.energystar.gov/buildings/facilityowners-and-managers/existing-buildings/use-portfolio-manager
| 53
64
LL84 reports fuel oils #2, #4, and #5/#6 separately, while LL87 reports all fuel oils in a combined field. The individually reported LL84 fuel
oil uses were summed together in order to apply the space heating fraction from LL87.
54 |
differences among low-rise, mid-rise, and very large properties. Data about
these distinctions can lead to the development of more targeted approaches
and to recommendations for building system improvements. Future reports will
continue to explore the many individual factors that affect buildings energy use.
Medians
The descriptions and comparisons of the energy use and the characteristics
of various classes of buildings have been carried out using standard statisticalanalysis tools. Most of these descriptions and comparisons were generated
using the programming language Python. 6 5
The median, the point at which half of the rank ordered data points falls above
or below, and the quartiles, the four equally distributed groups of data points,
provide a clear picture of energy use within a single type of building. These
tools help building owners and managers assess how their properties use
energy relative to others in the same building sector. Medians work well in
data sets that can be easily divided into smaller groups for detailed analysis.
This is usually true in historical comparisons of different existing buildings
with known locations, construction types, construction years and systems.
Medians do not take building size into account, so each building is given equal
weight. This means that medians cannot be used to find the total energy use
of a sector, even if the sectors total area is known.
65
AUGUST 2016
APPENDIX B
| 55
56 |
| 57
NYU CUSP
Each record in the compiled LL87 audit dataset consisted of a row of entries,
where features represented the fields from the Energy Audit Data Collection
Tool. For proper analysis, data needed to be extracted, cleaned, and
transformed. Because energy consultants entered numeric and categorical data
manually, features were cleaned to identify and correct improper and missing
entries. Non-numeric records that could not be directly converted to numbers
were stripped of spaces, commas, and appropriate units (e.g. kBtu for Energy
Savings). The remaining non-numeric records, which consisted of symbols,
comments, and indications that the data was unavailable, were identified as
missing data for the purpose of analysis. Categorical data was also cleaned
and manipulated for input in models. For certain categorical inputs, specifically
Heating System Type and Exterior Wall Type, there were significant differences
between how individual auditors entered data. For example, in the Heating
System Type field, where the Fuel Source was district steam, some auditors
entered the Heating System Type as Steam Boiler while others listed it as Other.
For Exterior Wall Type, many auditors with Mass walls listed wall type as Other
and described the wall as masonry, concrete, or brick in the comments field.
For all exterior walls that had one of these words in the comments, the wall
type was changed to Mass. Additionally, the year of construction of the building
was recoded to categorical bins of Before 1901, 1901 1920, 1921 1946, 1947
1970, 1971 1990, and After 1990. The Build Period divide in 1947 represents
the observed separation of Pre-War and Post-War buildings. The year of
construction field was recoded to account for observed non-linearity in the
effects of build year on energy use and efficiency (Kontokosta, 2015, 2012).
ECM recommendations were converted in a similar manner to systems data.
For each property (identified by the Borough Block Lot number or BBL),
the number of recommendations in each category and the sum of energy
savings, cost savings, and implementation cost are calculated. Finally, for
each BBL, the presence of an ECM recommendation in each category is
determined for each building in the dataset.
58 |
| 59
AUGUST 2016
Figure 41: LL84 Reported Floor Area Vs. LL87 Reported Floor Area
Properties reporting under both LL84 and LL87 tend to have different floor areas
reported in each data set. (Urban Green Council)
2.5M
Y=X
2.0M
1.5M
TREND LINE
1.0M
0.5M
0
0
0.5M
1.0M
1.5M
2.0M
2.5M
Figure 42: LL84 Reported Energy Use Intensity Vs. LL87 Reported Energy Use Intensity
Source EUIs for buildings listed in both the LL84 and the LL87 datasets often do not match. (Urban Green Council)
1000
Y=X
800
TREND LINE
600
400
200
0
0
200
400
600
800
1000
80%
84%
75%
Compliance Rate
60 |
87%
75%
60%
83%
87%
75%
40%
55%
20%
0%
2011
2012
2013
Reporting Year
2014
| 61
AUGUST 2016
Covered Buildings List (CBL). This effort resulted in a new, more accurate list
that will be used moving forward. Figure 43 reflects the compliance rate based
on the revised Covered Buildings List for 2014 only.
Every property type has improved compliance over time (Figure 44). The
multifamily and office sectors have had the highest and most consistent
Figure44: LL84 Compliance by Property Type Over Time
Multifamily and office buildings are taking the lead in compliance.
(Urban Green Council)
2011
MULTIFAMILY
HOUSING
2012
2013
2014
OFFICE
THEATERS
HOTELS
HOSPITALS
AND HEALTH
RETAIL
WAREHOUSES
EDUCATION
INDUSTRIAL
CULTURAL
GARAGES
OTHER
RELIGIOUS
0%
20%
40%
60%
Compliance
80%
100%
62 |
compliance rates, while warehouses and the education sector have seen the
largest discrete percentage improvement during the four years of reporting.
Recently, DOB announced a standard for the data quality of submissions
regarding missing or incorrect values in key fields. 66 In light of this
announcement, this report examined the overall data quality of each sector
for the 2013 benchmarking data, based on whether a submission provided
valid energy data as defined in Appendix B. For each sector, the percentage
of properties in compliance exceeded the percentage remaining after data
cleaning. For example, while 90 percent of multifamily properties submitted
data, only 78 percent submitted valid energy data (Figure 45). DOBs new
requirements for compliance will likely call attention to the issue of data
quality, and the expectation is this number will improve beginning with
data submitted in 2016.
Figure 45: LL84 Compliance Compared to Validity of Energy Data
Analysis of data quality shows that the majority of compliant properties provide
valid data. (Urban Green Council)
100%
Compliance
with Usable
%
Energy Data
90%
80%
Compliance Rate
70%
60%
50%
40%
30%
20%
10%
0%
IN
66
T
US
RI
AL
ED
TI
CA
ON
AR
W
U
HO
SE
S
RE
I
TA
LS
ITALT H
P
S A
H O& H E
T
HO
EL
S
OF
FI
CE
I LY
AM ING
F
I
S
LT O U
MU H
For more information about DOBs latest changes to data compliance see their notice here: www1.nyc.gov/assets/buildings/pdf/ll84_
change.pdf.
| 63
AUGUST 2016
ADVISORS
Michael Bobker
Building Performance
Lab at City University
of New York
Michael Colgrove
New York State
Energy Research and
Development Authority
Luke Falk
Related Companies
Jonathan Flaherty
Tishman Speyer
Adam Hinge
Sustainable Energy
Partnerships
Cliff Majersik
Institute for Market
Transformation
David Penick
Hines Holdings, Inc.
Jeffrey Perlman
Bright Power, Inc.
Peter Lampen
Douglas Elliman
Philip Skalaski
The Durst Organization
Conor Laver
Bright Power, Inc.
Marc Zuluaga
Steven Winter Associates
John Lee
New York City Mayors
Office of Sustainability
Ross MacWhinney
New York City Mayors
Office of Sustainability
Stacy Lee
New York City Mayors
Office of Sustainability
Daniel Marasco
Center for Urban Science
and Progress, New York
University
Vadhil Amadiz
Ashokan Services, Inc.
Liz Galst
Editorial Consultant
Hershel Weiss
Ashokan Services, Inc.
Hilary Ford
Graphic Design
Scott Moe
Scott Moe Design
Leslie Cook
Environmental Protection
Agency
Alexander Durst
The Durst Organization
PARTNERS
Bartosz Bonczak
Center for Urban Science
and Progress, New York
University
Constantine Kontokosta
Center for Urban Science
and Progress, New York
University
CONSULTANTS
Jamie Kleinberg
Policy and Research
Assistant Manager
Richard Leigh
Research Director
Cecil Scheib
Chief Program Officer
Russell Unger
Executive Director
Jonathan Walsh
Communications Manager
Anna Weingord
Research Associate