# Changelog ## 3.0.6 **Assumption changes** - Updated gas supply projections to reflect Maui closure at the end of 2026. - Removed mining and construction from the list of subsectors where onsite anaerobic digestion is possible. - Updated some industrial sector ESD projection methods following review, most notably tying dairy processing demand to existing cattle stock projections. - Removed endogenous Methanex closure method, instead assuming closure when Maui closes end 2026. - Updated some heatpump costs following review. **Model fixes** - Corrected a bug in NDGHG implementation that allowed coal to still be used for cooking post-2037. - Updated GIC demand data to reflect historical Methanex use more accurately. - Adjusted geothermal heatrates to better align with official statistics on primary energy. - Corrected a bug in diesel emissions for electricity generation. - Delayed convergence from empirical 2023 transport utilisation to harmonised within-bucket utilisation assumptions until 2035 (previously 2024), avoiding an abrupt shift immediately after the base year for some vehicle classes. ## 3.0.5 **Assumption changes** - Solar generation capacity factors for different timeslices are now generated from NIWA EPW weather files using PVWatts, using MBIE's October 2024 TMY3 (present climate) release. - Allowed the electricity sector to build new diesel peakers, using cost and efficiency assumptions from existing OCGT data. - Assumed all grid-scale solar utilised single-axis tracking technology. - Added or tweaked several new industrial technologies, allowing more flexibility in the demand end, including some hydrogen uses. - Added biomass as a fuel option for high-temperature heat in cement production to reflect the Golden Bay Cement biomass transition. - Added existing/known grid-scale battery installations. - Increased battery lifespan assumptions. - Increased geothermal generation lifespan assumptions. - Updated base year data to use latest release of EEUD, revising some figures. **Model features** - Rebuilt hydrogen electrolysis methods to correctly utilise electricity demand. - Increased model detail of distributed solar, reflecting improved capacity factor from larger-scale rooftop installations. - Added a more robust demand-flex method. This means ripple control is reflected more accurately with minimal processing overhead. Scalable to other demand flex areas as needed. - Added more detail to near-term or historical modelled plant build timings, ensuring accurate partial generation when fully commissioned partway through a year. - Added new functionality to constrain capital uptake rates of selected technologies. **Model fixes** - Resolved an error which led to excessive technology inflexibility in transport utilisation constraints. This allows for more realistic purchasing behaviour across different utilisation categories. - Disabled excessive optimisation of energy service demand between timeslices, which was leading to unrealistic demand flex behaviour. Note that this change also disables ESD timeslice reporting. - Corrected several links in biogas topology. - Resolved an issue that was leading to unrealistic switching behaviour in Other Industry demand. ## 3.0.4 **Assumption changes** - Adjusted lightbulb lifetimes in residential and commercial sectors according to usage assumptions. - Ensured realistic minimum utilisation of peaking plants. - Slightly increased conservative hydro peaking contribution assummption to 90%. - Create "two-tier" LNG market, limiting its use to electricity generation. **Model fixes** - Corrected implementation of new technologies for industrial applications. - Resolved a bug in implementation of 2037 coal boiler ban for intermediate process heat. - Resolved an issue that made industrial petrol engines more efficient than intended. - Resolved a bug leading to infeasible coal and biogas usage. - Added unallocated onsite generation to total system demand, ensuring system balance. - Add missing residential petrol emissions. - Allowed some industrial subsectors to electrify water heating using technologies already found in other subsectors. ## 3.0.3 **Features** - Changed distributed solar process to use exogenous forecasts, providing more realistic results. - Rebuilt gas distribution network to allow for biomethane blending and more accurately represent maintenance costs as fixed, rather than per unit of throughput. **Assumption changes** - Adjusted LNG terminal to fixed installation date in 2027 for Steady scenario. LNG remains excluded from Shift. - Also removed MIP functionality as this is no longer required. - Minor adjustments to capital costs for agricultural heating technologies following review. - Decreased average geothermal generation assumption slightly to align better with base year data. - Removed contingent gas from Steady scenario. **Model fixes** - (Biofuel) Improved municipal waste production differences between scenarios. - (Gas supply) Fixed historical domestic field output levels. - (Electricity generation) Fixed incorrect wind availability curve assumptions. - (Electricity generation) Fixed dispatchable hydro and wind peak contribution factors not aligning with documentation. - (Agriculture) Improved modelling of geothermal heating for indoor cropping. - (Agriculture) Fixed a bug that meant agricultural energy emissions were counted incorrectly. - (Agriculture) Added greater distinction for refrigeration, offroad transport, and heating technologies. - (Agriculture) Resolved an issue that meant petrol-powered boats and farm bikes were more efficient than intended. - (Residential) Fixed an issue allowing too much flexibility in space heating technologies. - (Commercial) Added greater distinction for boilers, burners, and resistance heaters. - (Commercial) Fixed modelling of geothermal heating for education, healthcare, WSR, and other commercial sectors. - (Commercial) Resolved an error that led to some LPG, natural gas, diesel, and petrol technologies being cheaper than intended. - (Industry) Resolved an error that allowed Methanol and Urea to close different parts of their business at different rates. ## 3.0.2 **Features** - Added more detail to residential load curves, allowing for different load profiles per use type. - Added an additional biofuel supply forecast scenario with AD feedstock growth and higher recoverability of biomass. - Separated biogas and biomethane modelling commodities, including a specific process for carbon scrubbing. - Added demand flex options for residential water heating, including splits between scenarios. **Model fixes** - (Residential) Resolved an error that undercounted space heating demand in detached dwellings. - (Natural gas) Fixed a methodological error that double-counted domestic natural gas costs. - (Commercial) Resolved an error that left commercial heatpumps more expensive and with worse efficiency than intended. - (Residential, Commercial) Fixed a methodological error that overly restricted demand device time flexibility. - (Electricity generation) Ensured geothermal plants operate as baseload. - (Electricity generation) Resolved an error that limited hydro dam output flexibility. **Assumption changes** - (Electricity generation) Adjusted Steady and Shift NREL scenarios from Conservative and Moderate to Moderate and Advanced, respectively. - (Commercial, Residential) Increased heatpump lifetimes slightly. - (Industrial) Removed some load curve assumptions. ## 3.0.1 **Features** - Incorporated documentation into sphinx-based site for hosting on readthedocs. - Explorer now includes detailed transport demand, service, and capacity chart options. **Model fixes** - (Biofuel) Ensured electricity demand was integrated into biofuel transformation processes. - (Commercial) Resolved an error that led to some demand technologies being cheaper than intended. - (Electricity) Ensured biogas and biomethane were made available as options for existing gas-fired plants. - (Electricity) Allowed Huntly Rankines to use black pellets. - (Transport) Ensured utilisation band constraints were properly fixed across model horizon. - (Transport) Resolved an error that led to electric heavy truck costs decreasing faster than intended in some scenarios. - (Transport) Resolved an error that led to some vehicle types having unrealistic utilisation rates. ## 3.0.0 **Features** - Initial build of `PREPARE-TIMES-NZ` module, which creates TIMES-NZ model files based on hosted data and code. - All preprocessing methods migrated to python and open-sourced, to ensure replicability and tracability. - Initial build of `TIMES-NZ-INTERNAL-QA` module, which allows for post-processing and interrogation of model results. - Results hosted publicly and can be explored at the highest possible level of detail to ensure transparency. - Rebuilt and updated model for base year 2023 and new Steady and Shift scenarios.