DC Carbon Footprint Calculator

Facility Size (MW IT) ?

Facility IT Capacity

Total IT power capacity in megawatts. This is linked to the IT Load value from the Operational tab and determines the building footprint, structural steel, and concrete quantities for embodied carbon calculations per EN 15978 LCA stages A1-A5.

Building Type ?

Construction Type

The type of building construction affects embodied carbon significantly. Purpose-built data centers use optimized designs. Converted warehouses reuse existing structures (lower embodied carbon). Modular/prefab facilities are manufactured off-site with standardized components.

Purpose-built Converted warehouse Modular / prefab Concrete Grade ?

Concrete Specification

Concrete grade determines embodied carbon intensity. Standard C30 emits ~150 kgCO2e per ton. High-strength C40 emits ~250 kgCO2e/ton but uses less volume. Low-carbon mixes with 50% GGBS (Ground Granulated Blast-furnace Slag) can reduce embodied carbon to ~100 kgCO2e/ton.

C30 standard (150 kgCO₂e/ton) C40 high-strength (250 kgCO₂e/ton) Low-carbon 50% GGBS (100 kgCO₂e/ton) Steel Sourcing ?

Structural Steel Source

Steel production method significantly impacts embodied carbon. World average (30% recycled) emits 1850 kgCO2e/ton. Electric Arc Furnace (EAF) with 90%+ recycled content emits only 500 kgCO2e/ton. Virgin Basic Oxygen Furnace (BOF) steel emits 2200 kgCO2e/ton. Data from ICE Database v4.1.

World average 30% recycled (1850 kgCO₂e/ton) EAF recycled 90%+ (500 kgCO₂e/ton) Virgin BOF (2200 kgCO₂e/ton) Seismic Zone ?

Seismic Design Category

Higher seismic zones require more structural reinforcement, increasing steel and concrete quantities by 10-40%. Zone 0 has no seismic requirements. Zone 4 (very high) requires base isolation, moment frames, and additional structural mass, significantly increasing embodied carbon.

0 - None 1 - Low 2 - Moderate 3 - High 4 - Very High

UPS Type ?

UPS Topology

Uninterruptible Power Supply type affects both embodied carbon (manufacturing) and operational efficiency. Modular UPS systems are lighter and can be right-sized. Standalone double-conversion UPS units are heavier with more copper and steel per kW.

Modular Standalone double-conversion Battery Type ?

Battery Chemistry

Battery chemistry determines embodied carbon per kWh of storage. VRLA Lead-Acid emits 68 kgCO2e/kWh with a 5-year life. Li-ion LFP emits 62 kgCO2e/kWh with a 10-year life. Li-ion NMC emits 74 kgCO2e/kWh but offers higher energy density.

VRLA Lead-Acid (68 kgCO₂e/kWh) Li-ion LFP (62 kgCO₂e/kWh) Li-ion NMC (74 kgCO₂e/kWh) Battery Runtime ?

UPS Runtime Duration

Duration the UPS battery system can support the full IT load during a power outage. Longer runtime requires more battery capacity, increasing embodied carbon. 5 minutes is typical for generator-backed sites; 15-30 minutes for sites without generators.

5 minutes 10 minutes 15 minutes 30 minutes Redundancy ?

Power Redundancy Level

Infrastructure redundancy level per Uptime Institute tiers. N = no redundancy. N+1 = one extra unit. 2N = fully duplicated. 2N+1 = duplicated plus one extra. Higher redundancy multiplies equipment quantities and embodied carbon proportionally.

N N+1 2N 2N+1

Workload Type ?

IT Workload Profile

Determines the embodied carbon per server. Traditional enterprise servers emit ~1200 kgCO2e each. AI/GPU servers (e.g., NVIDIA H100) have higher embodied carbon at ~4000 kgCO2e due to complex GPU manufacturing. Newer B200+ GPUs improve to ~3500 kgCO2e per server.

Traditional Enterprise (1200 kgCO₂e/server) Cloud / Hyperscale (1200 kgCO₂e/server) AI / GPU H100 (4000 kgCO₂e/server) AI / GPU B200+ (3500 kgCO₂e/server) Servers per MW ?

Server Density

Number of servers that fit in 1 MW of IT load, auto-calculated based on workload type. Traditional servers average ~5 kW each (~200/MW). AI GPU servers average ~10 kW each (~100/MW). This determines total embodied carbon from IT equipment manufacturing.

Server Refresh Cycle: 4 years ?

Server Replacement Cycle

How often servers are replaced with new hardware. Shorter cycles increase total embodied carbon over the project lifetime. Industry standard is 3-5 years. Extending to 6-7 years reduces embodied carbon but may increase operational energy use from less efficient hardware.

Network Refresh Cycle: 5 years ?

Network Equipment Cycle

Replacement cycle for switches, routers, and network infrastructure. Network gear typically has a longer useful life than servers. Embodied carbon per network device is lower than servers but the total count can be significant in large deployments.

Project Lifetime: 20 years ?

Facility Operational Lifetime

Total expected operational life of the data center facility. Determines how many IT refresh cycles occur and amortizes building embodied carbon. Typical range is 15-25 years for purpose-built facilities, though some operate 30+ years with major refurbishments.

Equipment Origin ?

Supply Chain Distance

Average distance that major equipment (UPS, generators, cooling units, switchgear) travels from manufacturer to site. Local sourcing under 500 km has minimal transport emissions. International shipments over 5000 km can add 2-5% to total embodied carbon.

Local <500km Regional 500-2000km International >5000km Transport Mode ?

Primary Transport Method

Mode of transport for heavy equipment delivery. Truck-only is typical for local. Truck+Ship is standard for international heavy cargo (lowest per-ton-km emissions). Truck+Air is used for urgent or specialized components but has ~50x higher emissions than sea freight.

Truck only Truck + Ship Truck + Air Construction Duration ?

Build Timeline

Total construction duration from ground-breaking to commissioning. Longer construction periods mean more on-site diesel consumption (cranes, excavators, generators), worker transport emissions, and temporary facility energy use. Modular builds typically complete faster.

12 months 18 months 24 months 36 months

Decommission Plan ?

End-of-Life Strategy

Approach to facility decommissioning at end of operational life. Full demolition has the highest end-of-life carbon. Partial refit preserves the shell structure. Building reuse retains the entire structure for a new purpose, earning carbon credits under LCA stage D.

Full demolition Partial refit Building reuse E-waste Program ?

IT Equipment Recycling

E-waste recycling and recovery program for decommissioned IT equipment. Standard recycling recovers ~80% of materials. Advanced recovery reaches 90% with precious metal extraction. Circular programs achieve 95% through refurbishment, component reuse, and certified recycling partners.

Standard recycling 80% Advanced recovery 90% Circular program 95% Steel Recycling Rate: 92% ?

Structural Steel Recovery

Percentage of structural steel recovered and recycled at end of life. Steel is highly recyclable with typical recovery rates of 85-98%. Higher rates reduce the net embodied carbon through avoided virgin production credits in LCA stage D (benefits beyond system boundary).

Copper Recycling Rate: 85% ?

Copper Cable Recovery

Percentage of copper from cabling, busbars, and transformers recovered at end of life. Copper recycling saves ~4 kgCO2e per kg compared to virgin production. Rates of 85-95% are achievable with dedicated cable stripping and separation processes.

Country ?

NDC Country Reference

This mirrors the country selected in the Location & Grid tab. Each country has Nationally Determined Contributions (NDCs) under the Paris Agreement with specific reduction targets and timelines. The NDC data auto-populates based on your selected location.

NDC Target Year ?

NDC Commitment Year

The year by which the country pledges to achieve its Nationally Determined Contribution target under the Paris Agreement. Most countries have targets for 2030, with some extending to 2035 or 2050 for net-zero commitments.

NDC Reduction Target ?

Emissions Reduction Pledge

The country's pledged percentage reduction in greenhouse gas emissions compared to a baseline year (varies by country). For example, the US targets 50-52% below 2005 levels by 2030. The EU targets 55% below 1990 levels by 2030.

Current Total Emissions (MtCO₂) ?

National GHG Emissions

The country's total annual greenhouse gas emissions in megatons of CO2 equivalent. This provides context for the data center sector's proportional share of national emissions and helps assess sector-level carbon budget allocation.

DC Sector Share: 3.5% ?

Data Center Energy Share

Estimated percentage of national electricity consumption attributed to data centers. Globally this ranges from 1-4%, but in data center hubs like Ireland (18%), Singapore (7%), or Virginia (25%+ of local grid), the share is much higher. This determines the DC sector's fair share of the national carbon budget.

DC Sector Growth Rate: 15%/yr ?

Annual DC Growth Rate

Projected annual growth rate of data center capacity in the selected country. AI-driven demand is pushing growth rates to 15-30% in major markets. Higher growth means the sector's share of the carbon budget must be split among more facilities, reducing per-MW allocation.

Budget Period ?

Assessment Timeframe

The time period over which the cumulative carbon budget is calculated. Shorter periods (2025-2030) align with near-term NDC targets. Longer periods (2025-2050) align with net-zero commitments. The budget methodology determines how emissions are distributed across this period.

2025-2030 2025-2035 2025-2040 2025-2050 Budget Methodology ?

Reduction Pathway Shape

How the carbon budget is distributed over the budget period. Linear decline reduces emissions evenly each year. Front-loaded requires steeper initial cuts (more ambitious). Back-loaded allows higher near-term emissions but requires rapid reduction later (higher cumulative risk).

Linear decline Front-loaded reduction Back-loaded Cumulative DC Budget (MtCO₂) ?

Sector Carbon Budget

Total cumulative CO2 emissions allowed for the data center sector in this country over the budget period. Calculated from the national carbon budget, DC sector share, and growth projections. This is the "carbon pie" that all data centers in the country must share.

Annual Budget per MW (tCO₂/MW/yr) ?

Per-MW Carbon Allocation

Your facility's fair share of the national DC sector carbon budget, expressed as tons of CO2 per MW of IT load per year. If your actual emissions exceed this budget, you are consuming more than your fair share of the national carbon budget. Values below 1000 tCO2/MW/yr typically require significant renewable energy procurement.

AI Training: 15% ?

AI Model Training

Percentage of IT load dedicated to AI/ML model training workloads. Training is the most energy-intensive workload per unit of economic output but is typically a transient burst workload. Carbon allocation per MWh for training is debated due to its high societal value potential.

AI Inference: 20% ?

AI Inference Serving

Percentage of IT load for AI inference (serving predictions/responses). Inference typically uses 60-90% of AI compute but at lower per-request energy than training. Rapidly growing workload segment driven by LLM adoption, autonomous systems, and real-time AI applications.

Cloud SaaS/IaaS: 30% ?

Cloud Computing

Percentage of IT load running cloud services including SaaS applications and IaaS virtual machines. Cloud workloads benefit from high utilization rates and elastic scaling, typically achieving better carbon efficiency per unit of compute than on-premises enterprise IT.

Enterprise IT: 20% ?

Enterprise Applications

Percentage of IT load for traditional enterprise workloads such as ERP, databases, email, file storage, and internal applications. These workloads often have lower server utilization rates (15-30%) compared to cloud-native applications, resulting in higher carbon per useful compute unit.

Government / Critical: 15% ?

Critical Infrastructure

Percentage of IT load serving government, healthcare, defense, financial, and other critical national infrastructure. These workloads may receive preferential carbon budget allocation due to their essential nature, but they also face stricter regulatory reporting requirements.

Current Renewable %: 40% ?

Current Renewable Procurement

Your facility's current percentage of electricity from renewable sources (PPAs, RECs, on-site generation). This baseline is used to model the pathway to your net-zero target year and calculate the required annual increase in renewable procurement.

Target Year Net-Zero ?

Net-Zero Target Year

The year by which your facility aims to achieve net-zero carbon emissions. Earlier targets (2030) require aggressive renewable procurement and efficiency measures. This is compared against the national NDC timeline to assess Paris Agreement alignment.

2030 2035 2040 2050 Carbon Credit Strategy ?

Offsetting Approach

Strategy for addressing residual emissions that cannot be eliminated through efficiency and renewables. Voluntary offsets (e.g., Verra VCS) are purchased on voluntary markets. Compliance credits (e.g., EU ETS allowances) are mandatory in regulated markets. Using "Both" provides regulatory coverage and voluntary leadership positioning.

None Voluntary offsets Compliance credits Both