2. Base year demand

The base year commercial data is intended to reflect the distribution of 2023 energy end use across all sectors in New Zealand. These commercial technologies should be available to the model to meet future demand, but with enough information (efficiency, lifetime, availability factors, capital costs, etc.) that the model will retire technologies at appropriate points and can make least-cost decisions on fuel switching and utilisation across demand segments.

TIMES-NZ requires detailed information on the existing commercial technologies, including:

• Energy efficiency

• Lifetime

• Investment cost ($/kW)

• O&M cost ($/kW)

• Availability factor

• Installed capacity (GW)

• Island splits

We have improved on TIMES-NZ 2.0 by adding data centre energy demand. This reflects the growth of New Zealand’s data centre energy demand due to rising digital demand, cloud services, and AI computing workloads. Also revisited other adjustments to the EEUD data, equipment lifetime, capital costs, and operating and maintenance costs.

2.1. Historic demand

Fuel energy demand for the 2023 year has primarily been sourced from the Energy Efficiency and Conservation Authority’s (EECA) Energy End Use Database (EEUD)[1], for the period ending 2023.

The EEUD’s stationary energy module uses a top-down approach: start with Ministry of Business, Innovation and Employment’s (MBIE) sector-by-fuel totals and then split them into finer categories using the best available information. In practice, for each combination of sector and fuel, EECA applies percentage splits (by subsector, end-use, and technology) to the MBIE total, yielding disaggregated estimates of delivered energy use. These percentage splits (the low-level detail) are based on data from either EECA’s internal research and modelling of energy profiles for given sectors (mainly 2020 estimates), or 2007 data estimates for given sectors. EECA scales these estimates to derive a current year estimate. The scaler for most sectors is sector-specific GDP values. The scalars act as a proxy for energy use growth over time.

Commercial categories in TIMES-NZ are defined combining a few EEUD categories together. Table 1 presents TIMES-NZ commercial sectors and EEUD sectors mapping. Some unallocated demand in EEUD was allocated to a TIMES-NZ “Other” sector.

Table 17 TIMES-NZ and EEUD sector mapping

TIMES-NZ commercial sector	EEUD commercial sector
Education	Education and Training: Pre-School, Primary and Secondary; Education and Training: Tertiary Education and Other Education
Healthcare	Health Care and Social Assistance
Office Blocks	Financing, Insurance, Real Estate and Business Services; Information Media and Telecommunications; Local Government Administration; Public Administration and Safety; Defence
Warehouses, Supermarkets and Retail (WSR)	Accommodation and Food Services; Retail Trade – Food; Transport, Postal and Warehousing (Commercial - Non-Transport); Wholesale and Retail Trade - Non-Food; Wholesale Trade - Food
Other	Arts, Recreational and Other Services; Building Cleaning, Pest Control and Other Support Services
Other (to be allocated to the above – assumptions applied)	NULL (unallocated)
Data centres	This is currently not defined in EEUD

Table 18 Commercial sector historic demand

Sector	Energy (PJ)	Energy share (%)
Education	3.17	6%
Healthcare	4.77	9%
Office Blocks	15.46	28%
Warehouses, Supermarkets and Retail (WSR)	21.24	38%
Other	3.79	7%
Other (to be allocated to the above)	7.35	13%
Total	55.78	100%

2.2. Data centre demand

MBIE/EEUD data does not recognise the energy use of Data Centers. The only known data available on data centre energy use is from NZTech[2] and MBIE’s Electricity Demand and Generation Scenarios (EDGS) 2024 model[3].

There are 56 operating facilities with a combined ≈ 104 MW of deployable capacity, drawing 238 GWh per year (0.62 % of national demand) in New Zealand[4]. While 58% operate with less than 1 MW each, most of the new capacity is being delivered through a small number of very large developments, particularly hyperscale (10–100+MW; Global cloud platforms, AI training) campuses for global cloud providers and large colocation (0.5–20 MW; Multi-tenant facilities) sites serving multiple clients[5].

TIMES-NZ 3.0 considered NZTech’s approach which is built based on a bottom-up inventory of every New Zealand datacentre, adjusting the numbers to reflect realistic deployable capacity rather than maximum design ratings. Where needed, it applied 89% deployable capacity. This excluded MWs already leased within other facilities to avoid double counting, and cross-checked results against public sources. This yielded a national total of 104 MW deployable capacity. To estimate electricity use we assume 80% of that space is occupied and servers draw 25% of name-plate power. We then apply the country’s median Power Usage Effectiveness (PUE)[6] of 1.3. This estimates about 238 GWh (or 0.856 PJ), or ~0.6 % of NZ’s 2024 demand. This total data centre demand 0.856 PJ is allocated in TIMES-NZ 3.0 as a new subsector, and an equivalent amount is deducted from TIMES-NZ Office Blocks to avoid double-counting.

Further disaggregation of base year capacity, based on the end use energy profile (IT equipment load, cooling system, power infrastructure, and building services) is conducted when implementing it into the TIMES-NZ technologies. Table 19 includes energy distribution in a more efficient hyperscale style centre with a PUE of 1.3.

Table 19 Energy distribution of typical vs hyperscale-style data centre

Component	Share of Total Energy (%) — PUE 1.3	Notes
IT Equipment (incl. servers, storage, networking)	~77%	Core computing load including storage and network gear
Cooling Systems	~15%	CRACs, chillers, air/liquid systems
Power Infrastructure	~6%	UPS inefficiency, PDU losses, power conversion
Building Services	<1%	Lighting, security, fire suppression

2.3. Unallocated EEUD demand

EEUD includes some unallocated amounts of biogas, diesel, geothermal, and petrol energy use. We have allocated these unallocated quantities of biogas, diesel, and petrol energy use to the TIMES-NZ “Other” sector. For petrol and diesel, the end use technology is assumed to be non-transport “Motive Power, Mobile”. EECA identifies this as a significant stationary energy end-use category, largely covering off-road, or recreational uses (i.e., forklifts, grounds-keeping equipment, and recreational marine vehicles for adventure tourism)[7].

For biogas, Nelson Hospital operates a 2.0 MW landfill gas–fired boiler that supplies over 60% of the hospital’s thermal energy needs for heating, hot water, and steam[8]. The Nelson Tasman Regional Landfill Business Unit (NTRLBU) sells landfill gas from York Valley to Health NZ, with an expected annual supply of about 2 million m³, equivalent to approximately 40% of the gas recovered from the landfill[9]. Typical landfill gas containing around 50% methane has a lower heating value of about 18–19 MJ per m³, giving the hospital’s annual landfill gas consumption an energy content of roughly 0.037 PJ[10]. This demand is captured by EECA Regional Energy Transition Accelerator (RETA) for process heating[11] applications at the hospital. Therefore, a 0.037 PJ of EEUD unallocated biogas was allocated to the “Healthcare” sector “Intermediate Heat (100-300 C), Process Requirements”.

The EEUD includes a large amount of unallocated geothermal, and this has been split across subsectors using the GNS New Zealand geothermal use database[12]. Percentages were derived based on the geothermal capacity by primary use and sector descriptions as given in the database and then applied these percentages to the EEUD unallocated geothermal demand.

EEUD also includes some amount of natural gas within Transport, Postal and Warehousing (Commercial - Non-Transport) WSR sector, which has not been allocated to any end use or technology. Given the operation of this sector, the total amount of natural gas is assumed to be use for motive power, mobile. Tracing back to EEUD original version in 2006-2007 further clarifies this. Table 4 includes those adjusted fuel demand by TIMES sector, end use, and technology. Within the office sector, natural gas motive power (mobile) was merged with LPG, since it accounted for only a small portion of total demand.

Table 20 TIMES-NZ commercial sector unallocated 2023 base year demand

TIMES Sector	TIMES Fuel	TIMES end use	TIMES Technology	Demand 2023 PJ	Share of fuel demand (%)
Other	Diesel	Motive Power, Mobiles	Internal Combustion Engine (Land Transport)	4.033	100%
Other	Petrol	Motive Power, Mobile	Internal Combustion Engine (Land Transport)	0.429	100%
Healthcare	Biogas	Process Heat	Boiler	0.037	6%
Other	Biogas	Process Heat	Boiler	0.223	94%
Education	Geothermal	Space Heating	Direct Heat	0.064	3%
Healthcare	Geothermal	Space Heating	Direct Heat	0.239	10%
WSR	Geothermal	Space Heating	Direct Heat	0.71	30%
WSR	Geothermal	Water Heating	Direct Heat	0.729	31%
Other	Geothermal	Space Heating	Direct Heat	0.544	23%
Other	Geothermal	Water Heating	Direct Heat	0.092	4%
WSR	Natural Gas	Motive Power, Mobile	Internal Combustion Engine (Land Transport)	0.254	100%

2.4. Island disaggregation

Energy demand across the North and South Islands was allocated by commercial subsector using appropriate proxy indicators:

• Education: Number of enrolled students by island[13]

• Healthcare: Number of hospital beds by island[14]

• Office Blocks: Regional GDP[15] of the relevant subsectors

• WSR: Regional GDP of the relevant subsectors

• Other: Population[16]

The resulting distribution of demand is shown in Table 21.

Table 21 Energy end-use share between North Island and South Island

Sector	NI Share	SI Share
Education	78.70%	21.30%
Healthcare	75.30%	24.70%
Office Blocks	78.50%	21.50%
WSR	81.60%	18.40%
Other	76.40%	23.60%

Natural gas and geothermal energy use are assumed to be 100% in the North Island, reflecting the fact that both the gas network and geothermal resources are located exclusively there.

2.5. Existing commercial end use

The following categories were used to represent end-use demand in the Commercial sector. These were adapted from the EEUD framework with several adjustments: cooking was divided into elements and ovens; lighting was disaggregated into incandescent, fluorescent, and LED; and pumping was combined with electronics, as it represented only a minor share when considered independently. TIMES 2.0 share assumptions were used for the items where further detail was added.

Table 22 Existing commercial end use technologies

End use	Found in
Intermediate Heat (100-300 C), Process Requirements	Healthcare, Other
Intermediate Heat (100-300 C), Cooking Elements	WSR
Intermediate Heat (100-300 C), Cooking Ovens	WSR
Electronics and Other Electrical Uses	All
Lighting	All
Low Temperature Heat (<100 C), Space Heating	All
Low Temperature Heat (<100 C), Water Heating	All
Motive Power, Mobile	All
Motive Power, Stationary	All
Refrigeration	Education, Healthcare, WSR
Space Cooling	All

---

[1]

EECA | EEUD: https://www.eeca.govt.nz/insights/data-tools/energy-end-use-database/

[2]

NZTech (2025). Empowering Aotearoa New Zealand’s Digital Future: Our national data centre infrastructure

[3]

MBIE | EDGS 2024: https://www.mbie.govt.nz/assets/electricity-demand-and-generation-scenarios-report-2024.pdf

[4]

New Zealand Data Centers - 55 Facilities from 46 Operators

[5]

NZTech 2025: Empowering Aotearoa New Zealand’s Digital Future: Our national data centre infrastructure

[6]

PUE = Total Facility Energy / IT Equipment Energy. An ideal PUE is 1.0, meaning all electricity goes to IT equipment (no overhead). In practice, some energy is always consumed by cooling, power conditioning, and other support systems.

[7]

EECA | Energy End Use Database Insights 2023 data

[8]

Bioenergy facilities directory | Nelson Hospital - landfill gas from York Valley Landfill

[9]

RNZ | Nelson Hospital ditches coal now heating fully powered by landfill gas

[10]

IEA | An introduction to biogas and biomethane

[11]

EECA | Regional Energy Transition Accelerator (RETA) Nelson, Marlborough, Tasman – Phase One Report

[12]

GNS Science | New Zealand Geothermal Use Database

[13]

Education Counts | School rolls

[14]

Ministry of Health | Certified Providers

[15]

Stats NZ | Infoshare | Regional Gross Domestic Product – RNA Gross domestic product, by region and industry (Annual-Mar)

[16]

Stats NZ | Infoshare | DPE Estimated Resident Population for Regional Council Areas, on 30 June (1996+) (Annual-Jun)