diff --git a/.github/workflows/pytest.yaml b/.github/workflows/pytest.yaml index 2dfead8729..f5ba0d2145 100644 --- a/.github/workflows/pytest.yaml +++ b/.github/workflows/pytest.yaml @@ -2,11 +2,11 @@ name: Test on: # To debug the workflow, uncomment this entry AND comment pull_request_target - # pull_request: - # branches: [ main ] - pull_request_target: - branches: [ main, "migrate**" ] - types: [ labeled, opened, reopened, synchronize ] + pull_request: + branches: [ main ] + # pull_request_target: + # branches: [ main, "migrate**" ] + # types: [ labeled, opened, reopened, synchronize ] schedule: - cron: "0 5 * * *" # = 06:00 CET = 07:00 CEST @@ -25,8 +25,20 @@ env: python-version: "3.13" jobs: + skip: + name: Skip workflow on PR branch + runs-on: ubuntu-latest + steps: + - name: Early exit + run: | + gh run cancel ${{ github.run_id }} + gh run watch ${{ github.run_id }} + env: + GH_TOKEN: ${{ secrets.GITHUB_TOKEN }} + check: name: Check permissions, identify ref to test + needs: skip runs-on: ubuntu-latest steps: - if: > diff --git a/doc/_static/global_r12_basin_map.png b/doc/_static/global_r12_basin_map.png new file mode 100644 index 0000000000..a57f75bcc3 Binary files /dev/null and b/doc/_static/global_r12_basin_map.png differ diff --git a/doc/_static/message_nexus_structure.png b/doc/_static/message_nexus_structure.png new file mode 100644 index 0000000000..1ef2fa62ac Binary files /dev/null and b/doc/_static/message_nexus_structure.png differ diff --git a/doc/_static/water_reference_system.png b/doc/_static/water_reference_system.png new file mode 100644 index 0000000000..b697ff84ac Binary files /dev/null and b/doc/_static/water_reference_system.png differ diff --git a/doc/global/index.rst b/doc/global/index.rst index c579ae9c25..61985291b2 100644 --- a/doc/global/index.rst +++ b/doc/global/index.rst @@ -1,6 +1,15 @@ MESSAGEix-GLOBIOM global model ============================== +.. caution:: This section of the documentation is under revision + on this branch. + See :issue:`424` for details. + + Text on these pages may mix old and revised material; + links may be broken; + and in general these pages **should not** be used as reference + until changes are reviewed and merged via :pull:`425`. + These pages document the IIASA Integrated Assessment Modeling (IAM) framework, also referred to as **MESSAGEix-GLOBIOM**, owing to the fact that the energy model |MESSAGEix| and the land use model GLOBIOM are its most important components. |MESSAGEix|-GLOBIOM was developed for the quantification of the so-called Shared Socio-economic Pathways (SSPs) which are the first application of the IAM framework. diff --git a/doc/global/water/climate_impacts.rst b/doc/global/water/climate_impacts.rst new file mode 100644 index 0000000000..ae09907525 --- /dev/null +++ b/doc/global/water/climate_impacts.rst @@ -0,0 +1,584 @@ +.. _water-climate-impacts: + +Climate Change Impacts +====================== + +A key innovation of the MESSAGEix-Nexus module is the explicit representation of climate change impacts on both water availability and energy systems (Awais et al., 2024 :cite:`awais_2024_nexus`). This enables analysis of climate change adaptation strategies, compound risks at the water-energy nexus, and interactions between climate impacts and mitigation policies. + +Overview +-------- + +Climate change affects the water-energy-land nexus through multiple pathways: + +**Direct Impacts on Water**: + +* Changes in precipitation patterns and amounts +* Shifts in snowmelt timing and magnitude +* Altered groundwater recharge rates +* Increased evapotranspiration +* More frequent and severe droughts +* Changes in seasonal water availability + +**Direct Impacts on Energy**: + +* Reduced thermal power plant efficiency due to higher ambient temperatures +* Increased cooling water requirements +* Higher cooling water temperatures constraining once-through cooling +* Changes in hydropower generation potential +* Shifts in electricity demand (heating vs. cooling) +* Impacts on renewable energy resources (wind, solar) + +**Nexus Interactions**: + +* Water scarcity constrains thermal power plant operation +* Temperature and water stress compound during heat waves +* Competing demands for limited water resources intensify +* Adaptation measures in one sector affect the other + +MESSAGEix-Nexus represents these impacts through: + +* Time-varying water availability from climate-driven hydrological models +* Temperature-dependent power plant efficiency +* Cooling technology performance degradation +* Changed sectoral water demands + +Climate Forcing and Scenarios +------------------------------ + +Climate impacts are derived from climate model projections and impact models forced by these projections. + +Representative Concentration Pathways (RCPs) +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Climate forcing is characterized by Representative Concentration Pathways (van Vuuren et al., 2011 :cite:`vanvuuren_2011_rcp`): + +* **RCP 2.6**: Strong mitigation, ~2°C warming by 2100 +* **RCP 4.5**: Moderate mitigation, ~2.5°C warming by 2100 +* **RCP 6.0**: Moderate-high emissions, ~3°C warming by 2100 +* **RCP 8.5**: High emissions, ~4-5°C warming by 2100 + +Each RCP implies different: + +* Global mean temperature increase +* Regional temperature patterns +* Precipitation changes (regional and seasonal) +* Extreme event frequency and intensity + +Shared Socioeconomic Pathways (SSPs) +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +The MESSAGEix-Nexus module is typically applied within the SSP-RCP scenario framework (O'Neill et al., 2014 :cite:`oneill_2014_ssp`; Riahi et al., 2017 :cite:`riahi_2017_ssp`): + +* **SSP1-2.6**: Sustainability pathway with strong mitigation +* **SSP2-4.5**: Middle-of-the-road with moderate mitigation +* **SSP3-7.0**: Regional rivalry with weak mitigation +* **SSP5-8.5**: Fossil-fueled development with no mitigation + +This framework enables exploration of how socioeconomic development pathways interact with climate change impacts on the water-energy nexus. + +Climate Model Ensembles +^^^^^^^^^^^^^^^^^^^^^^^^ + +To account for climate model uncertainty, impacts are derived from ensembles of global climate models (GCMs): + +* **CMIP5**: Coupled Model Intercomparison Project Phase 5 (used in IPCC AR5) +* **CMIP6**: CMIP Phase 6 (used in IPCC AR6) +* **Ensemble median**: Typical approach to represent central tendency +* **Ensemble spread**: Can be used to explore uncertainty + +Typically, 5-10 GCMs are used to force hydrological models, and ensemble statistics (median, quantiles) are calculated at the basin scale. + +Hydrological Impacts +-------------------- + +Changes in water availability are the most direct climate impact on the water-energy nexus and are represented through outputs from global hydrological models. + +Hydrological Model Framework +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Two global hydrological models are primarily used: + +**PCR-GLOBWB 2** (Sutanudjaja et al., 2018 :cite:`sutanudjaja_2018_pcrglobwb`): + +* 5 arcmin spatial resolution (~10 km at equator) +* Simulates full terrestrial water cycle +* Includes surface water, soil moisture, groundwater +* Forced by climate model outputs (temperature, precipitation, etc.) +* Provides runoff, groundwater recharge, and river discharge + +**CWatM** - Community Water Model (Burek et al., 2020 :cite:`burek_2020_cwatm`): + +* Variable resolution (typically 5 arcmin) +* Represents water availability, demand, and allocation +* Includes reservoirs and water management +* Can simulate environmental flows +* Provides similar outputs to PCR-GLOBWB + +Both models are forced by bias-corrected climate model outputs to simulate historical (1971-2000) and future (2020-2100) water availability. + +Spatial Aggregation +^^^^^^^^^^^^^^^^^^^ + +Hydrological model outputs are spatially aggregated to MESSAGE basins: + +1. **Grid cell outputs** (5 arcmin resolution) +2. **Basin delineation** using HydroSHEDS +3. **Area-weighted aggregation** to ~200 MESSAGE basins +4. **Mapping to MESSAGE regions** (R12) for consistency + +This multi-scale approach preserves spatial heterogeneity while enabling computational tractability. + +Temporal Aggregation +^^^^^^^^^^^^^^^^^^^^ + +Hydrological model outputs are temporally aggregated: + +* **Native resolution**: Daily or monthly +* **Sub-annual MESSAGE**: Seasonal or monthly averages +* **Annual MESSAGE**: Annual mean with optional reliability constraints + +For climate impact studies, sub-annual resolution is critical to capture seasonal dynamics. + +Key Hydrological Impact Patterns +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Climate change impacts on water availability exhibit strong regional patterns (Awais et al., 2024 :cite:`awais_2024_nexus`): + +**Wetting Regions** (increased water availability): + +* High northern latitudes (more precipitation, earlier snowmelt) +* Parts of East Africa (intensified monsoons) +* Some tropical regions (increased convective precipitation) + +**Drying Regions** (decreased water availability): + +* Mediterranean basin (reduced precipitation, increased evaporation) +* Middle East and North Africa (lower precipitation) +* Southern Africa (decreased precipitation) +* Parts of South America (Amazon, Northeast Brazil) +* Southwestern USA (reduced snowpack, increased evaporation) + +**Seasonal Shifts** (changed timing of availability): + +* Snow-dominated basins (earlier snowmelt peak, lower summer flows) +* Monsoon regions (potential shifts in monsoon timing and intensity) +* Mediterranean climate regions (drier summers, wetter winters) + +**Increased Variability**: + +* More frequent and intense droughts +* Increased interannual variability +* Higher flood risks (not directly represented in MESSAGE) + +Drought Representation +^^^^^^^^^^^^^^^^^^^^^^^ + +Droughts are represented through: + +* **Low flow quantiles**: 10th or 20th percentile of flow distribution +* **Multi-year sequences**: Persistent dry periods from climate model runs +* **Statistical characterization**: Changes in drought frequency, duration, and severity + +The model can use low-flow quantiles to represent water availability under drought conditions, testing system resilience. + +Energy System Impacts +--------------------- + +Climate change directly affects energy system performance through temperature-dependent efficiency and cooling constraints. + +Thermal Power Plant Efficiency +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Thermal power plant efficiency declines with higher ambient temperature through: + +1. **Thermodynamic efficiency**: Carnot efficiency ∝ (T_hot - T_cold); higher T_cold reduces efficiency +2. **Cooling system performance**: Less effective heat rejection at high ambient temperatures +3. **Auxiliary loads**: Increased cooling system energy requirements + +The efficiency penalty is represented as: + +:math:`\eta(T) = \eta_0 \cdot \left(1 - \alpha \cdot (T - T_0)\right)` + +where: + +* :math:`\eta(T)` is efficiency at ambient temperature :math:`T` +* :math:`\eta_0` is reference efficiency at reference temperature :math:`T_0` +* :math:`\alpha` is temperature sensitivity coefficient (~0.2-0.5% per °C) + +Typical efficiency penalties: + +* **+1°C ambient temperature**: 0.2-0.5% efficiency loss +* **+3°C (RCP 4.5 by 2100)**: 0.6-1.5% efficiency loss +* **+5°C (RCP 8.5 by 2100)**: 1.0-2.5% efficiency loss + +This translates to increased fuel consumption and emissions for the same electricity output. + +Cooling Technology Performance +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Different cooling technologies respond differently to temperature increases: + +**Once-Through Cooling**: + +* Moderate efficiency penalty from higher ambient/water temperature +* Potentially binding constraints on intake/discharge water temperature +* Forced curtailment or shutdown during extreme heat events +* Affected by low flow conditions (reduced dilution capacity) + +**Recirculating (Wet Tower) Cooling**: + +* Moderate efficiency penalty +* Performance degrades at high wet-bulb temperature (limiting evaporation) +* Can operate at higher ambient temperatures than once-through +* Increased water consumption due to higher evaporation rates + +**Dry Cooling**: + +* Severe efficiency penalty at high temperatures (5-10% at 40°C) +* No water availability constraint +* Performance critical during heat waves when electricity demand peaks +* May require capacity derating at extreme temperatures + +Climate change thus creates differential impacts, making dry cooling relatively less attractive in hot climates despite eliminating water use. + +Cooling Water Temperature Limits +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Environmental regulations often limit: + +* **Intake temperature**: Maximum temperature of water that can be withdrawn +* **Discharge temperature**: Maximum temperature of water returned to environment +* **Delta-T**: Maximum temperature increase between intake and discharge + +Typical limits: + +* Discharge temperature: 30-35°C (varies by jurisdiction and water body) +* Delta-T: 3-5°C for once-through cooling + +As river and lake temperatures increase with climate change: + +* More frequent violations of discharge limits +* Required curtailment or shutdown during hot periods +* Economic incentive to retrofit to recirculating or dry cooling + +This is represented through temperature-dependent availability constraints on once-through cooling. + +Hydropower Generation +^^^^^^^^^^^^^^^^^^^^^^ + +Hydropower is affected by changes in: + +* **Annual runoff**: Determines total generation potential +* **Seasonal patterns**: Affects capacity factor and firm capacity +* **Reservoir inflows**: Impacts storage and regulation capability +* **Competing water uses**: Irrigation, municipal, environmental flows + +Regional hydropower impacts: + +* **Increases**: High latitudes, some tropical regions with increased precipitation +* **Decreases**: Snow-dominated basins (reduced summer flows), drying regions +* **Seasonal shifts**: Earlier spring peak, lower summer generation in snow basins + +Hydropower impacts are implicitly represented through changed water availability in basins with hydropower resources. + +Electricity Demand +^^^^^^^^^^^^^^^^^^ + +Climate change shifts electricity demand patterns through: + +* **Increased cooling demand**: Higher temperatures increase air conditioning loads +* **Decreased heating demand**: Milder winters reduce heating loads +* **Peak demand shifts**: More summer peaks, fewer winter peaks in many regions + +The net effect varies by region: + +* **Hot regions**: Large increase in cooling demand +* **Cold regions**: Decreased heating demand may offset cooling increases +* **Temperate regions**: Mixed effects depending on baseline climate + +Demand impacts are represented through temperature-dependent demand adjustments in MESSAGE. + +Compound Events and Cascading Impacts +-------------------------------------- + +A critical insight from MESSAGEix-Nexus is that climate impacts at the water-energy nexus can compound and cascade, creating risks greater than the sum of individual impacts. + +Heat-Drought Compound Events +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Heat waves and droughts often co-occur, creating compounding stresses: + +**Simultaneous Impacts**: + +* High electricity demand (cooling loads) +* Reduced power plant efficiency (high temperature) +* Low water availability (drought) +* High water temperature (constrains once-through cooling) +* Competing water demands (irrigation for stressed crops) + +**Cascading Effects**: + +* Water scarcity forces power plant curtailment +* Reduced electricity supply during peak demand +* Higher electricity prices and potential shortages +* Reduced economic output from energy-intensive industries +* Water allocation conflicts between sectors + +Historical examples: + +* **2003 European heat wave**: Nuclear plants curtailed due to high river temperatures +* **2012 US drought**: Thermal plants constrained by low water availability and high temperatures +* **2010 Russian heat wave**: Energy and water systems both severely stressed + +Climate change increases the frequency and severity of such events (Satoh et al., 2022 :cite:`satoh_2022_drought`). + +Nexus Stress Indicators +^^^^^^^^^^^^^^^^^^^^^^^^ + +MESSAGEix-Nexus can quantify nexus stress through indicators: + +* **Water scarcity index**: Ratio of demand to availability +* **Energy-water stress**: Frequency of water constraints on energy generation +* **Compound event frequency**: Co-occurrence of heat, drought, and high demand +* **Adaptation costs**: Investment required to maintain energy and water security + +These indicators show nonlinear increases under high-emission scenarios, with stress intensifying after mid-century (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Regional Vulnerability +^^^^^^^^^^^^^^^^^^^^^^ + +Regions with high vulnerability to compound water-energy-climate risks: + +* **Middle East and North Africa**: Already water-scarce, extreme heat, high cooling demands +* **South Asia**: High population, monsoon variability, irrigation demands +* **Mediterranean**: Drying trend, summer heat, tourism-driven peak demand +* **Southwestern USA**: Declining Colorado River, heat waves, competing demands +* **Australia**: Droughts, heat, limited water resources + +These regions show the largest impacts in MESSAGEix-Nexus scenarios and require substantial adaptation investment. + +Adaptation Strategies +--------------------- + +MESSAGEix-Nexus represents multiple adaptation strategies that can be endogenously selected or exogenously imposed. + +Water Supply Adaptation +^^^^^^^^^^^^^^^^^^^^^^^^ + +Expanding water supply through: + +* **Desalination**: Coastal regions can invest in seawater desalination (energy-intensive) +* **Groundwater expansion**: Where sustainable reserves exist (depth-dependent costs) +* **Wastewater reuse**: Treated effluent for non-potable uses +* **Inter-basin transfers**: Where infrastructure exists or can be built + +The model selects the least-cost portfolio of supply options based on: + +* Resource availability and costs +* Energy requirements and availability +* Competing demands +* Infrastructure constraints + +Energy System Adaptation +^^^^^^^^^^^^^^^^^^^^^^^^^ + +Adapting energy systems to water constraints: + +* **Cooling technology shifts**: Move from water-intensive to dry cooling +* **Generation technology shifts**: Increase wind, solar PV (no cooling water) +* **Plant siting**: Locate new thermal plants near reliable water sources +* **Operational flexibility**: Dispatch based on water availability and temperature + +The model endogenously optimizes technology choice and dispatch. + +Demand-Side Adaptation +^^^^^^^^^^^^^^^^^^^^^^^ + +Reducing water demands through: + +* **Irrigation efficiency**: Drip irrigation, scheduling optimization +* **Industrial water recycling**: Closed-loop systems +* **Municipal efficiency**: Leak reduction, efficient appliances +* **Energy efficiency**: Reduces cooling water requirements + +Some efficiency improvements are represented through exogenous technology improvement; others can be investment options. + +Integrated Nexus Adaptation +^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Optimal adaptation often involves coordinated strategies: + +* **Renewable energy + desalination**: Use solar PV to power desalination +* **Wastewater reuse for cooling**: Close water loop in industrial areas +* **Seasonal coordination**: Align energy maintenance with low water periods +* **Portfolio diversification**: Mix of generation and water supply options + +The integrated optimization in MESSAGEix-Nexus can identify such synergistic solutions. + +Adaptation Costs and Limits +^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Adaptation is not costless: + +* **Desalination**: 0.50-1.50 USD/m³ + energy costs +* **Dry cooling**: 8-15% capital cost increase + 3-8% efficiency penalty +* **Renewable energy**: Capital cost differential (though declining) +* **Efficiency improvements**: Upfront investment requirements + +At high levels of climate change (RCP 8.5), adaptation costs can be substantial: + +* 10-30% increase in water supply costs in water-stressed regions +* 5-15% increase in electricity generation costs +* Trade-offs with other investment priorities (development, mitigation) + +There may also be **adaptation limits** where physical or economic constraints prevent full adaptation: + +* Finite desalination capacity expansion rates +* Thermodynamic limits on dry cooling in extreme heat +* Competing uses for limited renewable energy +* Social and institutional barriers to demand reduction + +Results and Insights +--------------------- + +Application of MESSAGEix-Nexus with climate impacts provides key insights (Awais et al., 2024 :cite:`awais_2024_nexus`): + +Baseline Climate Impacts +^^^^^^^^^^^^^^^^^^^^^^^^^ + +In baseline (no climate policy) scenarios with climate change (RCP 4.5 or 8.5): + +* **Water availability** declines in 40-50% of global basins by 2050-2100 +* **Thermal power efficiency** reduced by 0.5-2% globally by 2100 +* **Cooling water constraints** become binding in 20-30% of basins with thermal generation +* **Adaptation costs** reach 50-150 billion USD/year globally by 2050 + +Regional impacts vary dramatically: + +* **MENA, South Asia**: Severe water scarcity, high adaptation costs +* **Europe, North America**: Moderate impacts, mostly manageable through adaptation +* **Sub-Saharan Africa**: Heterogeneous impacts, limited adaptation capacity + +Climate Mitigation Co-Benefits +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Stringent climate mitigation (RCP 2.6) substantially reduces nexus stress: + +* **Reduced temperature impacts**: Limits ambient temperature increases +* **Smaller hydrological changes**: Moderates precipitation and runoff changes +* **Lower thermal generation**: Rapid coal and gas phase-out reduces cooling water demand +* **Renewable expansion**: Wind and solar eliminate most cooling water requirements + +Co-benefits of mitigation for water resources: + +* 30-60% reduction in adaptation costs by 2050 (mitigation vs. baseline) +* Avoided water scarcity in 10-20% of basins +* Reduced compound event frequency by 40-70% + +This demonstrates that climate mitigation provides substantial co-benefits for water-energy security. + +SDG Interactions Under Climate Change +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Climate change exacerbates trade-offs between SDGs: + +* **SDG6 vs. SDG7**: Water access competes with energy access in water-scarce regions +* **SDG13 (climate action) supports both**: Mitigation reduces nexus stress +* **Costs of achieving SDGs**: 20-50% higher under RCP 8.5 vs. RCP 2.6 + +Regional variation is critical: + +* **MENA, South Asia**: Difficult to achieve both SDG6 and SDG7 under high climate change without substantial investment +* **Other regions**: Generally feasible but at higher cost + +The integrated framework allows quantification of these trade-offs and identification of investment priorities. + +Uncertainty and Robustness +^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Key uncertainties in climate impact assessment: + +* **Climate model spread**: ±30-50% uncertainty in regional precipitation changes +* **Hydrological model differences**: Different models show different sensitivities +* **Socioeconomic assumptions**: SSP pathway affects vulnerability and adaptive capacity +* **Technology development**: Uncertain costs and performance of adaptation options + +Robust adaptation strategies that perform well across scenarios: + +* **Renewable energy expansion**: Reduces cooling water needs across all scenarios +* **Water use efficiency**: Low-regret option with multiple benefits +* **Diversified supply portfolio**: Reduces vulnerability to single source failures +* **Flexible infrastructure**: Can adjust to different future conditions + +Sensitivity analysis and scenario exploration help identify robust strategies. + +Model Validation and Evaluation +-------------------------------- + +The climate impact representation in MESSAGEix-Nexus has been evaluated through: + +Historical Validation +^^^^^^^^^^^^^^^^^^^^^ + +Comparison of historical simulations (1971-2000) with observations: + +* **Water availability**: Hydrological models reproduce observed runoff patterns +* **Power plant performance**: Temperature-efficiency relationships match empirical data +* **Heat wave impacts**: Historical events (2003 Europe, 2012 USA) can be reproduced + +Intercomparison with Other Models +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Comparison with other integrated assessment and water-energy models: + +* **Qualitative agreement** on direction and magnitude of major impacts +* **Regional patterns** consistent across models +* **Quantitative differences** due to spatial resolution, representation details + +The multi-model comparison builds confidence in key findings while highlighting uncertainties. + +Stakeholder Engagement +^^^^^^^^^^^^^^^^^^^^^^ + +Results have been presented to and evaluated by: + +* Water resource managers +* Energy system planners +* Climate adaptation practitioners +* Policy makers + +Feedback has validated the relevance of modeled impacts and adaptation options while highlighting additional considerations (institutional barriers, equity, etc.) not fully represented in the model. + +Future Development +------------------ + +Ongoing and planned enhancements to climate impact representation: + +**Enhanced Hydrology**: + +* More detailed reservoir and water management representation +* Groundwater-surface water interactions +* Water quality and temperature tracking + +**Energy System Details**: + +* Sub-daily electricity demand and generation +* Transmission constraints affected by temperature +* Renewable energy resource climate sensitivities (wind, solar) + +**Extremes and Risks**: + +* Explicit flood representation +* Cascading infrastructure failures +* Financial and economic risk metrics + +**Socioeconomic Impacts**: + +* Health impacts of compound heat-water stress +* Migration and displacement from water scarcity +* Inequality in climate impact distribution + +These developments will further enhance the capability of MESSAGEix-Nexus to inform climate adaptation and resilience planning for water-energy systems. + +.. footbibliography:: + diff --git a/doc/global/water/cooling.rst b/doc/global/water/cooling.rst new file mode 100644 index 0000000000..a1cb8bec8b --- /dev/null +++ b/doc/global/water/cooling.rst @@ -0,0 +1,435 @@ +.. _water-cooling: + +Power Plant Cooling Technologies +================================= + +Power plant cooling technologies are a critical component of the water-energy nexus in MESSAGEix-Nexus. Thermal power plants (coal, gas, nuclear, concentrated solar power, geothermal) require cooling to dissipate waste heat from the thermodynamic cycle. The implementation of cooling technologies in MESSAGE explicitly represents the tradeoffs between water use, energy efficiency, and capital costs (Fricko et al., 2016 :cite:`fricko_2016`; Parkinson et al., 2019 :cite:`parkinson_2019`; Awais et al., 2024 :cite:`awais_2024_nexus`). + +Thermodynamic Basis +------------------- + +The water requirements and thermal pollution from power plant cooling are fundamentally linked to the plant's thermodynamic efficiency through the energy balance. + +Energy Balance +^^^^^^^^^^^^^^ + +Looking at a simplified thermal energy balance at the power plant (:numref:`fig-ppl_energy_balance`), total combustion energy (:math:`E_{comb}`) is converted into: + +* Electricity (:math:`E_{elec}`) +* Emissions and stack losses (:math:`E_{emis}`) +* Waste heat absorbed by cooling system (:math:`E_{cool}`) + +:math:`E_{comb} = E_{elec} + E_{emis} + E_{cool}` + +.. _fig-ppl_energy_balance: +.. figure:: /_static/ppl_energy_balance.png + :width: 400px + :align: center + + Simplified power plant energy balance. + +Converting to per unit electricity generation, we can estimate the cooling requirement per unit of electricity (:math:`\phi_{cool}`) based on average heat rate (:math:`\phi_{comb}`) and heat lost to emissions (:math:`\phi_{emis}`): + +:math:`\phi_{cool} = \phi_{comb} - \phi_{emis} - 1` + +where all quantities are expressed per unit of electricity output (e.g., MJ thermal per MWh electric). + +Time-Varying Heat Rates +^^^^^^^^^^^^^^^^^^^^^^^^ + +With time-varying heat rates (i.e., :math:`t = 0,1,2,...`) representing efficiency improvements, and assuming a constant share of energy to emissions and electricity: + +:math:`\phi_{cool}[t] = \phi_{comb}[t] \cdot \left( 1 - \dfrac{\phi_{emis}}{\phi_{comb}[0]} \right) - 1` + +This formulation enables heat rate improvements for power plants represented in MESSAGE to be automatically translated into improvements (reductions) in cooling water intensity. As plants become more efficient (lower heat rate), less waste heat must be dissipated per unit of electricity generated. + +For example: + +* **Coal plant**: Heat rate improvement from 10,000 MJ/MWh (36% efficient) to 8,500 MJ/MWh (42% efficient) reduces cooling requirement by ~15% +* **Gas combined cycle**: Heat rate improvement from 6,500 MJ/MWh (55% efficient) to 5,800 MJ/MWh (62% efficient) reduces cooling requirement by ~11% + +Cooling Water Intensities +^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Water withdrawal and consumption intensities for power plant cooling technologies are calibrated to ranges reported in the literature (Meldrum et al., 2013 :cite:`meldrum_2013`; Macknick et al., 2012 :cite:`macknick_2012`). The intensities account for: + +* Waste heat to be dissipated (from heat rate) +* Cooling technology efficiency +* Ambient conditions (temperature, humidity) +* Water temperature limits for discharge + +Representative water intensities are provided in :ref:`water-demand` and vary by plant type and cooling technology. + +Cooling Technology Options +--------------------------- + +Three main cooling technology categories are represented in MESSAGEix-Nexus, each with distinct characteristics regarding water use, energy penalties, and costs. + +Once-Through Cooling +^^^^^^^^^^^^^^^^^^^^ + +Once-through (open-loop) cooling draws water from a surface water body, passes it through the condenser to absorb waste heat, and returns the warmed water to the source. + +**Characteristics**: + +* **Very high water withdrawal**: 100-200 m³/MWh depending on plant type +* **Low water consumption**: 1-2 m³/MWh (only evaporation from source due to heating) +* **Minimal energy penalty**: Small pumping requirement (<0.3% of generation) +* **Low capital cost**: Simplest cooling system +* **Requires large water body**: River, lake, or ocean with adequate flow +* **Thermal pollution**: Discharged water is 8-15°C warmer than intake + +**Advantages**: + +* Lowest cost option +* Minimal parasitic energy loss +* Simple operation and maintenance + +**Disadvantages**: + +* Very high water withdrawal (though mostly returned) +* Thermal pollution impacts aquatic ecosystems +* Restricted by environmental regulations in many regions +* Requires proximity to large, reliable water source +* Vulnerable to water temperature constraints during heat waves + +**Availability**: + +* Primarily for coastal plants (seawater cooling) +* Large rivers with high minimum flows +* Great Lakes and similar large water bodies +* Increasingly restricted by environmental regulations (EU, USA) + +Recirculating (Wet Tower) Cooling +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Recirculating cooling uses a closed-loop system where water circulates between the condenser and a cooling tower. Heat is dissipated by evaporation in the cooling tower. + +**Characteristics**: + +* **Low water withdrawal**: 2-4 m³/MWh (to replace evaporation and blowdown) +* **Moderate-high water consumption**: 2-4 m³/MWh (mostly evaporation) +* **Small energy penalty**: 1-2% of generation (pumps, fans) +* **Moderate capital cost**: Cooling tower construction +* **Independent of large water bodies**: Can be located anywhere with adequate water supply +* **Minimal thermal pollution**: Water recirculates; only blowdown is discharged + +**Advantages**: + +* Much lower withdrawal than once-through +* Can be sited inland without large water body +* Minimal thermal discharge to environment +* Widely accepted technology + +**Disadvantages**: + +* Moderate-high water consumption (comparable to or higher than once-through) +* Energy penalty reduces net generation +* Higher capital and operating costs than once-through +* Visible water vapor plumes +* Still vulnerable to water scarcity during droughts + +**Availability**: + +* Standard technology for inland plants +* Can be retrofitted to existing once-through plants +* Suitable for most locations with adequate water supply + +Dry (Air) Cooling +^^^^^^^^^^^^^^^^^ + +Dry cooling uses air instead of water to dissipate heat, eliminating water consumption. Heat is transferred via air-cooled condensers or air-cooled heat exchangers. + +**Characteristics**: + +* **Minimal water withdrawal**: 0.05-0.15 m³/MWh (only for auxiliary systems) +* **Minimal water consumption**: 0.05-0.15 m³/MWh (>95% reduction vs. wet cooling) +* **Significant energy penalty**: 3-8% of generation depending on climate +* **High capital cost**: Large air-cooled condenser surface area +* **Climate-dependent performance**: Efficiency loss greater in hot climates +* **No thermal water pollution**: All heat dissipated to atmosphere + +**Advantages**: + +* Eliminates water use for cooling (~95-99% reduction) +* Enables plant siting in water-scarce regions +* No thermal water pollution +* No water availability risk to plant operations + +**Disadvantages**: + +* Significant efficiency penalty (especially in hot weather) +* Much higher capital cost (2-3× cooling system cost) +* Larger physical footprint +* Performance degradation during heat waves (when power demand peaks) +* Higher operating costs due to energy penalty + +**Availability**: + +* Increasingly used in water-scarce regions +* Required in some jurisdictions with limited water +* Growing market share for new plants in arid regions + +Hybrid Cooling +^^^^^^^^^^^^^^ + +Hybrid systems combine wet and dry cooling to balance water use and performance: + +* **Parallel hybrid**: Wet and dry systems operate in parallel; can shift load seasonally +* **Series hybrid**: Dry pre-cooling with wet trim cooling +* **Wet operation in peak demand**: Use wet cooling when electricity value is highest +* **Dry operation in water scarcity**: Save water when scarce + +Hybrid systems offer flexibility but add complexity and cost. They are represented in MESSAGEix-Nexus as a distinct technology option for some plant types. + +Implementation in MESSAGEix-Nexus +---------------------------------- + +The cooling technology representation in MESSAGEix-Nexus allows the model to endogenously select the optimal cooling technology for each power plant type in each region and time period (Parkinson et al., 2019 :cite:`parkinson_2019`; Awais et al., 2024 :cite:`awais_2024_nexus`). + +Technology Structure +^^^^^^^^^^^^^^^^^^^^ + +Each thermal power plant type that requires cooling is connected to multiple cooling technology options (:numref:`fig-cooling_implement1`). The investment and operation of cooling technologies are explicit decision variables in the optimization. + +.. _fig-cooling_implement1: +.. figure:: /_static/cooling_implement1.png + :width: 800px + :align: center + + Implementation of cooling technologies in the MESSAGE IAM (Fricko et al., 2016 :cite:`fricko_2016`). + +For example, a coal power plant can be built with: + +* Coal plant + once-through cooling +* Coal plant + recirculating cooling +* Coal plant + dry cooling + +Each combination has specific: + +* **Capital costs**: Plant cost + cooling system cost +* **Efficiency**: Plant efficiency - cooling energy penalty +* **Water withdrawal/consumption**: Technology-specific intensities +* **Operational constraints**: Water availability, thermal limits + +The model simultaneously optimizes: + +* Which power plants to build +* Which cooling technology to pair with each plant +* Operational dispatch considering water and energy constraints + +Cost Representation +^^^^^^^^^^^^^^^^^^^ + +Cooling technology costs are represented as: + +* **Capital cost differential**: Additional investment for cooling system relative to reference + + * Once-through: Reference (lowest cost) + * Recirculating: +5-10% of plant cost + * Dry cooling: +8-15% of plant cost + +* **Efficiency penalty**: Parasitic load reducing net electricity output + + * Once-through: 0.2-0.5% reduction + * Recirculating: 1-2% reduction + * Dry cooling: 3-8% reduction (climate-dependent) + +* **Operating costs**: Maintenance and additional fuel consumption + +Cost assumptions are derived from technology assessments (Zhai and Rubin, 2010 :cite:`zhai_2010`; Zhang et al., 2014 :cite:`zhang_2014`; Loew et al., 2016 :cite:`loew_2016`). + +Initial Cooling Technology Distribution +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +The base year (2020) distribution of cooling technologies for existing power plants is estimated using the dataset from Raptis and Pfister (2016) :cite:`Raptis_2016_powerplant_data`, which provides plant-level cooling technology data. + +Basin-scale shares of cooling technologies across all power plant types are shown in :numref:`fig-cooling_implement2`. The historical distribution shows: + +* **Coastal regions**: Predominantly once-through cooling +* **Inland rivers**: Mix of once-through and recirculating +* **Arid inland regions**: Higher share of dry and recirculating cooling +* **Developed regions**: Shift toward recirculating due to environmental regulations + +.. _fig-cooling_implement2: +.. figure:: /_static/cooling_implement2.png + :width: 800px + :align: center + + Average cooling technology shares across all power plant types at the river basin-scale (Fricko et al., 2016 :cite:`fricko_2016`). + +Future cooling technology choices are endogenous based on: + +* Water availability and scarcity +* Regulatory constraints (thermal pollution limits) +* Technology costs and performance +* Competition with other water demands + +Water-Energy Tradeoffs +---------------------- + +The explicit cooling technology representation enables MESSAGEix-Nexus to capture key water-energy tradeoffs. + +Water Scarcity Drives Technology Choice +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +In water-scarce regions or time periods, the model faces a choice: + +1. **Build thermal plants with water-intensive cooling**: Requires water allocation from other uses or new water supply +2. **Build thermal plants with dry cooling**: Higher cost and efficiency penalty +3. **Build alternative generation technologies**: Renewables (wind, solar PV) that don't require cooling water + +The optimal choice depends on: + +* Relative costs of water supply vs. efficiency penalty +* Availability and cost of alternative generation +* Value of water in competing uses + +Example: In a water-scarce basin, if groundwater costs 0.20 USD/m³ and a gas combined cycle plant requires 2.5 m³/MWh with wet cooling, the water cost is 0.50 USD/MWh. Dry cooling eliminates this water cost but has a ~4% efficiency penalty. At gas prices of 5 USD/GJ and 6,000 MJ/MWh heat rate, the efficiency penalty costs ~1.20 USD/MWh. If capital cost differential is small, wet cooling remains attractive despite water costs. + +Climate Change Amplification +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Climate change affects cooling technology performance through: + +1. **Higher ambient temperatures**: + + * Reduce efficiency of all cooling technologies + * Particularly severe for dry cooling (larger penalty) + * Can force output derating during heat waves + +2. **Higher water temperatures**: + + * Once-through cooling constrained by discharge temperature limits + * Recirculating cooling less affected (evaporative cooling) + +3. **Reduced water availability**: + + * Increases water scarcity and costs + * Incentivizes shift to dry cooling or alternative generation + +4. **Increased electricity demand**: + + * More cooling demand for buildings + * Increases value of generation, making efficiency penalties more costly + +These interactions can create "compound events" where heat waves simultaneously: + +* Increase electricity demand (cooling loads) +* Reduce power plant efficiency (high ambient temperature) +* Constrain water availability (drought) +* Limit once-through cooling (high water temperature) + +MESSAGEix-Nexus captures these dynamics, showing that climate impacts on the energy-water nexus can be more severe than impacts on either sector individually (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Regional Patterns +^^^^^^^^^^^^^^^^^ + +Cooling technology evolution varies by region: + +**Water-Abundant Regions** (Northern Europe, Canada, parts of South America): + +* Continued use of once-through cooling where environmentally acceptable +* Recirculating cooling as standard inland +* Limited dry cooling adoption + +**Water-Stressed Regions** (Middle East, North Africa, Central Asia, Australia): + +* Rapid shift to dry cooling for new thermal plants +* Reduced overall thermal generation share +* Increased solar PV and wind (no cooling water requirements) + +**Developing Regions** (South Asia, Southeast Asia, Sub-Saharan Africa): + +* Initial expansion with recirculating cooling (standard technology) +* Potential shift to dry cooling if water scarcity intensifies +* Competition between energy access and water access goals + +**Transition Regions** (China, India, Western USA): + +* Mix of technologies depending on local water availability +* Retrofits of once-through to recirculating +* New plants increasingly using dry or hybrid cooling in water-scarce areas + +Scenario Results +---------------- + +Results from MESSAGEix-Nexus scenarios illustrate the cooling technology dynamics (Awais et al., 2024 :cite:`awais_2024_nexus`): + +Baseline Scenarios (No Climate Policy) +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +In baseline scenarios without climate policy: + +* **Thermal generation** continues to dominate (40-50% of global generation) +* **Recirculating cooling** becomes dominant technology (60-70% of new thermal capacity) +* **Dry cooling** grows in water-scarce regions (10-20% of new thermal capacity) +* **Once-through cooling** declines due to environmental regulations (20-30% of capacity by 2100, down from ~50% in 2020) + +Water consumption from power generation increases by 50-100% by 2050 despite efficiency improvements, driven by: + +* Generation growth in developing regions +* Shift from once-through (low consumption) to recirculating (high consumption) + +Climate Change Impacts +^^^^^^^^^^^^^^^^^^^^^^^ + +Adding climate change impacts (no adaptation): + +* **Thermal generation efficiency** declines by 1-3% due to higher ambient temperatures +* **Water scarcity** intensifies, particularly in already water-stressed regions +* **Compound heat-drought events** force generation curtailments +* **Energy-water nexus stress** increases costs of electricity generation + +With endogenous adaptation: + +* **Dry cooling share** increases to 30-40% of new thermal capacity in hot, water-scarce regions +* **Renewable generation** (solar PV, wind) expands faster due to cooling water constraints on thermal +* **Thermal generation declines** more rapidly than in scenarios without water-energy nexus constraints + +Climate Mitigation Scenarios +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +In stringent climate mitigation scenarios (1.5-2°C): + +* **Thermal generation** declines rapidly (coal phase-out, reduced gas) +* **Cooling water demand** peaks around 2030-2040 and then declines +* **Cooling technology choice** matters less for new capacity (less thermal being built) +* **Existing capacity** may see retrofits to dry cooling in water-scarce regions +* **Renewable generation** eliminates most cooling water demand by 2070-2100 + +Climate mitigation substantially reduces water-energy nexus stress by reducing thermal generation. + +SDG Interactions +^^^^^^^^^^^^^^^^ + +When SDG6 (water access) constraints are enforced: + +* **Municipal water demand** increases due to infrastructure for universal access +* **Competition for water** intensifies between municipal and energy sectors +* **Dry cooling adoption** accelerates in regions with SDG-driven water stress +* **Trade-offs** emerge between energy access (SDG7) and water access (SDG6) in water-scarce regions + +The model can quantify these trade-offs and identify least-cost pathways to achieve both SDGs (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Key Insights +------------ + +The cooling technology representation in MESSAGEix-Nexus provides several key insights: + +1. **Water availability is an important constraint** on energy system development in water-scarce regions, affecting technology choice and generation dispatch. + +2. **Endogenous cooling technology choice** enables the model to find cost-effective adaptation strategies to water scarcity, including shifts to dry cooling and alternative generation technologies. + +3. **Climate change creates compound risks** at the water-energy nexus, with simultaneous temperature, water availability, and demand stresses. + +4. **Mitigation reduces nexus stress**: Climate mitigation scenarios reduce cooling water demand by phasing out thermal generation, providing a co-benefit for water resources. + +5. **SDG interactions are complex**: Achieving universal water access can constrain energy system choices in water-scarce regions, requiring careful planning and investment. + +6. **Regional heterogeneity matters**: Global average trends obscure important regional dynamics where water-energy constraints are binding. + +The implementation demonstrates the value of integrated water-energy modeling for understanding nexus interactions, identifying vulnerabilities, and evaluating policy and technology options. + +.. footbibliography:: + diff --git a/doc/global/water/demand.rst b/doc/global/water/demand.rst new file mode 100644 index 0000000000..41876aecaf --- /dev/null +++ b/doc/global/water/demand.rst @@ -0,0 +1,422 @@ +.. _water-demand: + +Water Demand +============ + +Water demand in MESSAGEix-Nexus is represented across four major sectors: energy, municipal, industrial manufacturing, and agriculture (Awais et al., 2024 :cite:`awais_2024_nexus`). Demands are specified at the basin scale and evolve over time based on socioeconomic drivers (population, GDP, urbanization) and technological change. Competition between sectors for limited water resources is explicitly resolved through the optimization. + +Energy Sector Water Demand +--------------------------- + +The energy sector is the most explicitly represented water demand in MESSAGEix-Nexus, with water requirements emerging from technology-specific intensities rather than exogenous demand trajectories. + +Power Plant Cooling +^^^^^^^^^^^^^^^^^^^ + +Thermal power plants (coal, gas, nuclear, concentrated solar power, geothermal) require cooling to dissipate waste heat. Cooling water requirements are the largest energy sector water demand in most regions. The cooling technology implementation is described in detail in :ref:`water-cooling`. + +Water withdrawal and consumption intensities vary by: + +* **Power plant type**: Different heat rates and cooling requirements +* **Cooling technology**: Once-through, recirculating (wet tower), dry cooling +* **Climate conditions**: Ambient temperature affects cooling requirements + +Typical water intensities (Meldrum et al., 2013 :cite:`meldrum_2013`): + +.. list-table:: Power plant cooling water intensities + :widths: 30 25 25 + :header-rows: 1 + + * - Technology + - Withdrawal (m³/MWh) + - Consumption (m³/MWh) + * - Coal - once-through + - 100-150 + - 1-2 + * - Coal - recirculating + - 2-3 + - 2-3 + * - Coal - dry cooling + - 0.05-0.10 + - 0.05-0.10 + * - Gas combined cycle - once-through + - 40-80 + - 0.5-1 + * - Gas combined cycle - recirculating + - 0.5-1.5 + - 0.5-1.5 + * - Nuclear - once-through + - 100-200 + - 1.5-2.5 + * - Nuclear - recirculating + - 2.5-4 + - 2.5-4 + * - Concentrated solar power - recirculating + - 2.5-3.5 + - 2.5-3.5 + +Once-through cooling withdraws large volumes but returns most water to the source (albeit warmer). Recirculating cooling withdraws less but consumes most of what is withdrawn through evaporation. Dry cooling eliminates water use but has efficiency penalties and higher capital costs. + +The model endogenously chooses cooling technologies based on water availability, costs, and performance impacts (see :ref:`water-cooling`). + +Fuel Extraction and Processing +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Water is required for fossil fuel extraction and processing: + +* **Coal mining**: 0.05-0.30 m³/GJ (washing, dust suppression) +* **Conventional oil and gas**: 0.02-0.10 m³/GJ (drilling, processing) +* **Unconventional oil (oil sands, shale)**: 0.50-2.00 m³/GJ (steam injection, hydraulic fracturing) +* **Biofuel production**: 1-5 m³/GJ (crop irrigation, processing) - mainly captured through agricultural demand + +These demands are relatively small compared to cooling but can be significant in water-scarce regions with large extractive industries. + +Hydropower +^^^^^^^^^^ + +Hydropower generation does not consume water (it is non-consumptive) but affects water availability through: + +* **Reservoir evaporation**: Can be significant in arid regions with large reservoirs +* **Flow timing**: Alters seasonal patterns of water availability downstream +* **Environmental flows**: Minimum release requirements affect energy generation + +Reservoir evaporation is calculated based on: + +:math:`Evap = A_{reservoir} \cdot E_{rate} \cdot f_{exposure}` + +where :math:`A_{reservoir}` is surface area, :math:`E_{rate}` is evaporation rate (mm/year, climate-dependent), and :math:`f_{exposure}` is the fraction of time the reservoir is full. + +Typical evaporation from reservoirs ranges from 1-3 m/year in temperate climates to 2-4 m/year in arid regions. + +Municipal Water Demand +---------------------- + +Municipal water demand includes residential, commercial, and public sector water use in urban and rural areas. + +Demand Drivers +^^^^^^^^^^^^^^ + +Municipal water demand is driven by: + +* **Population**: Total population in each basin/region +* **Urbanization rate**: Urban populations have higher per-capita demand +* **Income level**: Water use increases with GDP per capita (up to saturation) +* **Water access rates**: Connection to piped water systems +* **Water use efficiency**: Technological change and policy-driven improvements + +Demand Estimation +^^^^^^^^^^^^^^^^^ + +Municipal water demand is projected using a regression-based approach: + +:math:`D_{municipal,b,t} = Pop_{b,t} \cdot \left( f_{urban,b,t} \cdot d_{urban}(GDP_{pc,t}) + (1-f_{urban,b,t}) \cdot d_{rural}(GDP_{pc,t}) \right) \cdot access_{b,t}` + +where: + +* :math:`D_{municipal,b,t}` is municipal demand in basin :math:`b`, time :math:`t` +* :math:`Pop_{b,t}` is population +* :math:`f_{urban,b,t}` is urbanization rate +* :math:`d_{urban}`, :math:`d_{rural}` are per-capita demand functions of GDP per capita +* :math:`access_{b,t}` is the fraction of population with access to improved water supply + +Per-Capita Demand Patterns +^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Per-capita municipal water demand follows an income-dependent pattern: + +* **Low income** (<5,000 USD/capita/year): 20-50 liters/capita/day +* **Middle income** (5,000-20,000 USD/capita/year): 100-200 liters/capita/day +* **High income** (>20,000 USD/capita/year): 150-300 liters/capita/day (saturates) + +The relationship is typically modeled as a logarithmic or logistic function that saturates at high income levels. Urban demand is typically 2-3 times rural demand at similar income levels due to: + +* Access to piped water systems +* Water-using appliances +* Commercial and public sector demands +* Landscape irrigation + +Regional variations exist based on climate (outdoor water use), culture, and water pricing. + +SDG6 Water Access Constraints +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +The Sustainable Development Goals framework includes targets for universal water access (SDG 6.1): + +* **Universal access** to safely managed drinking water by 2030 +* Requires infrastructure investment proportional to unserved population +* Creates minimum demand for municipal water infrastructure + +SDG constraints can be activated in MESSAGEix-Nexus scenarios: + +:math:`access_{b,t} \geq access_{target}(t)` + +where :math:`access_{target}(t)` is the target access rate trajectory (e.g., reaching 100% by 2030). + +Achieving universal access requires substantial investment in water supply infrastructure, particularly in sub-Saharan Africa and South Asia where current access rates are 50-70% (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Return Flows +^^^^^^^^^^^^ + +Municipal water use has significant return flows: + +* **Wastewater return rate**: 70-90% of withdrawals return as wastewater +* **Treatment level**: Determines usability for reuse or environmental release +* **Timing**: Return flows available in same period as withdrawal (no storage) + +Return flows can be: + +* Released to rivers (adding to downstream availability) +* Treated and reused locally +* Used for environmental flows + +Industrial Manufacturing Demand +-------------------------------- + +Industrial water demand includes manufacturing processes, cooling, and product incorporation. It is distinct from energy sector industrial demands (already counted in power generation). + +Demand Drivers +^^^^^^^^^^^^^^ + +Industrial water demand is driven by: + +* **Manufacturing output**: GDP from industrial sector +* **Industrial structure**: Heavy vs. light industry have different water intensities +* **Technology and efficiency**: Water recycling and process improvements +* **Water pricing**: Higher prices incentivize efficiency + +Demand Estimation +^^^^^^^^^^^^^^^^^ + +Industrial demand is estimated using a water intensity approach: + +:math:`D_{industrial,b,t} = GDP_{ind,b,t} \cdot I_{water}(t)` + +where: + +* :math:`GDP_{ind,b,t}` is industrial GDP in basin :math:`b`, time :math:`t` +* :math:`I_{water}(t)` is water intensity (m³ per USD of industrial output) + +Water intensity typically declines over time due to: + +* **Technological improvement**: More efficient processes and water recycling +* **Structural change**: Shift from heavy to light industry +* **Regulations**: Water use restrictions and pricing + +Historical trends show water intensity declining at 1-2% per year in developed economies. + +Sectoral Water Intensities +^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Different industrial sectors have very different water requirements: + +* **Chemicals and petrochemicals**: 10-50 m³/1000 USD +* **Paper and pulp**: 50-300 m³/1000 USD +* **Steel and metals**: 20-100 m³/1000 USD +* **Food and beverages**: 10-50 m³/1000 USD +* **Textiles**: 50-200 m³/1000 USD +* **Electronics**: 5-20 m³/1000 USD + +Aggregate industrial water intensity depends on the sectoral composition of manufacturing in each region. + +Return Flows and Recycling +^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Industrial water use has variable return flows: + +* **High-recycling industries** (steel, chemicals): 50-90% return rate +* **Low-recycling industries** (food, textiles): 20-40% return rate +* **Product incorporation** (beverages): 5-10% consumed in products + +Industrial wastewater may require treatment before reuse or environmental release, depending on: + +* Pollutant loads (organic, inorganic, thermal) +* Discharge regulations +* Reuse opportunities + +Industrial demand is relatively stable seasonally compared to agricultural demand. + +Agricultural Irrigation Demand +------------------------------- + +Agricultural irrigation is the largest water demand globally (~70% of total withdrawals) and exhibits strong seasonal variability. In MESSAGEix-Nexus, irrigation demand is derived from the GLOBIOM land-use model linkage. + +GLOBIOM Linkage +^^^^^^^^^^^^^^^ + +Irrigation water demand is calculated in GLOBIOM based on: + +* **Crop area**: Irrigated area for each crop type +* **Crop water requirements**: Climate-dependent evapotranspiration +* **Irrigation efficiency**: Technology-dependent water delivery and application efficiency +* **Rainfall**: Effective precipitation reduces irrigation needs + +GLOBIOM provides basin-scale irrigation demand to MESSAGEix-Nexus, which must be satisfied by available water resources. Water scarcity in MESSAGEix-Nexus can feed back to GLOBIOM by: + +* Increasing irrigation costs (water pricing) +* Constraining irrigated area expansion +* Incentivizing efficiency improvements + +Seasonal Patterns +^^^^^^^^^^^^^^^^^ + +Irrigation demand varies seasonally based on: + +* **Crop calendars**: Planting and growing season timing +* **Evapotranspiration**: Peak during warm, dry periods +* **Monsoon patterns**: Low irrigation during rainy seasons + +Example monthly demand pattern (Northern India): + +* **January-March**: High (wheat, vegetables) +* **April-June**: Very high (summer crops, pre-monsoon) +* **July-September**: Low (monsoon period) +* **October-December**: Moderate (post-monsoon crops) + +Seasonal variability creates critical periods when irrigation competes strongly with other demands and water availability is lowest (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Irrigation Technologies +^^^^^^^^^^^^^^^^^^^^^^^ + +Irrigation efficiency depends on technology: + +* **Flood/furrow irrigation**: 40-60% efficiency (large conveyance and field losses) +* **Sprinkler irrigation**: 60-75% efficiency +* **Drip/micro irrigation**: 75-90% efficiency + +Efficiency improvements reduce demand for the same crop production: + +:math:`D_{irrigation} = \dfrac{CWR \cdot Area}{Eff_{irrigation}}` + +where :math:`CWR` is crop water requirement, :math:`Area` is irrigated area, and :math:`Eff_{irrigation}` is irrigation efficiency. + +Higher efficiency technologies have higher capital costs but reduce water demand and can enable expansion of irrigated area in water-constrained regions. + +Climate Change Impacts +^^^^^^^^^^^^^^^^^^^^^^ + +Climate change affects irrigation demand through: + +* **Evapotranspiration changes**: Generally increases with temperature +* **Precipitation changes**: Regional increases or decreases affect irrigation needs +* **Crop calendar shifts**: Earlier springs, longer growing seasons +* **CO₂ fertilization**: Higher CO₂ can reduce crop water requirements + +In most regions, climate change increases net irrigation demand despite CO₂ effects (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Sectoral Competition and Allocation +------------------------------------ + +When water is scarce (demand exceeds availability), the model must allocate water across competing sectors. Allocation is determined by: + +Economic Value +^^^^^^^^^^^^^^ + +Sectors with higher economic value per unit water receive priority: + +* **Industrial/municipal**: High value (1-10 USD/m³) +* **Energy (cooling)**: Medium-high value (0.50-5 USD/m³) +* **Irrigation**: Variable value (0.01-1 USD/m³ depending on crop and productivity) + +The model balances marginal values across sectors to maximize total economic welfare. + +Infrastructure and Flexibility +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Existing infrastructure creates rigidities: + +* Power plants require cooling or must reduce generation +* Urban populations require minimum municipal supply +* Agricultural demands are flexible (can fallow fields, deficit irrigate) + +The model accounts for costs of: + +* Not meeting demand (scarcity costs, value of lost load for electricity) +* Adjusting to constraints (switching technologies, deficit irrigation) + +Temporal Flexibility +^^^^^^^^^^^^^^^^^^^^ + +Some demands are temporally flexible: + +* **Irrigation**: Can shift timing within crop growth period +* **Industrial**: Some processes can shift to wet season +* **Energy**: Flexible generation can be scheduled to water availability +* **Municipal**: Relatively inflexible, requires continuous supply + +Storage (reservoirs, aquifer storage) provides temporal flexibility to match seasonal supply and demand. + +Regional Differences +^^^^^^^^^^^^^^^^^^^^ + +Water scarcity and sectoral competition vary greatly by region: + +* **Arid regions** (Middle East, North Africa, Central Asia): Scarcity is norm, high competition +* **Monsoon regions** (South Asia, Southeast Asia): Seasonal scarcity, competition in dry season +* **Temperate regions** (Europe, North America): Generally abundant, localized scarcity +* **Tropical regions** (Sub-Saharan Africa, Latin America): Variable, infrastructure-limited + +Scenarios with stringent climate change and rapid development can increase water scarcity and sectoral competition significantly (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Demand Projections +------------------ + +Future water demand depends on scenario assumptions: + +Shared Socioeconomic Pathways (SSPs) +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Different SSPs imply different demand trajectories: + +* **SSP1 (Sustainability)**: + + * Lower population growth + * High efficiency and water productivity + * Strong environmental regulations + * Lowest demand growth + +* **SSP2 (Middle-of-the-road)**: + + * Medium population and economic growth + * Moderate efficiency improvements + * Continued irrigation expansion + * Medium demand growth + +* **SSP3 (Regional rivalry)**: + + * High population growth in developing regions + * Slow efficiency improvements + * Irrigation expansion constrained by water scarcity + * Highest demand growth but supply-limited + +* **SSP5 (Fossil-fueled development)**: + + * Rapid economic growth and urbanization + * High energy demands = high cooling water demand + * Efficient water use in high-income regions + * High total demand but technology-enabled supply + +Climate Change Impacts +^^^^^^^^^^^^^^^^^^^^^^ + +Climate change affects demands through: + +* **Temperature**: Higher cooling demands (energy, buildings) +* **Precipitation**: Changed irrigation requirements +* **Extremes**: Droughts increase marginal value of water + +Combined SSP-RCP scenarios (SSP2-4.5, SSP5-8.5, etc.) capture both socioeconomic and climate drivers (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Global Demand Outlook +^^^^^^^^^^^^^^^^^^^^^^ + +Baseline global water demand projections (2020-2100): + +* **Municipal**: 50-150% increase (driven by population and urbanization) +* **Industrial**: 100-300% increase (driven by economic growth) +* **Irrigation**: 10-70% increase (limited by water availability and efficiency) +* **Energy**: 50-200% increase (depends on generation mix and cooling choices) + +Regional patterns vary substantially, with largest growth in South Asia, Middle East, and Sub-Saharan Africa. + +.. footbibliography:: + diff --git a/doc/global/water/index.rst b/doc/global/water/index.rst index 0b4bfff01f..8e5d57a567 100644 --- a/doc/global/water/index.rst +++ b/doc/global/water/index.rst @@ -1,71 +1,213 @@ -Water -============ - -The water withdrawal and return flows from energy technologies are calculated in -MESSAGE following the approach described in Fricko et al., (2016) :cite:`fricko_2016`. -Each technology is prescribed a water withdrawal and consumption intensity (e.g., m3 per kWh) -that translates technology outputs optimized in MESSAGE into water requirements and return -flows. - -.. _fig-ppl_energy_balance: -.. figure:: /_static/ppl_energy_balance.png - :width: 320px - :align: right - - Simplified power plant energy balance. - -For power plant cooling technologies, the amount of water required and energy dissipated -to water bodies as heat is linked to the parameterized power plant fuel conversion efficiency (heat -rate). Looking at a simple thermal energy balance at the power plant (:numref:`fig-ppl_energy_balance`), total combustion -energy (:math:`E_{comb}`) is conveterted into electricity (:math:`E_{elec}`), emissions (:math:`E_{emis}`) -and additional thermal energy that must be absorbed by the cooling system (:math:`E_{cool}`): - -:math:`E_{comb} = E_{elec} + E_{emis} + E_{cool}` - -Converting to per unit electricity, we can estimate the cooling required per unit of electricity generation -(:math:`\phi_{cool}`) based on average heat-rate (:math:`\phi_{comb}`) and heat lost to emissions -(:math:`\phi_{emis}`), and this data is identified from the literature :cite:`fricko_2016`. - -:math:`\phi_{cool} = \phi_{comb} - \phi_{emis} - 1` - -With time-varying heat-rates (i.e., :math:`t =0,1,2,...`) and a constant share of energy to emissions and electricity: - -:math:`\phi_{cool}[t] = \phi_{comb}[t] \cdot \left( \, 1 - \dfrac{\phi_{emis}}{\phi_{comb}[0]} \, \right) - 1` - -Increased fuel efficiency (lower heat-rate) reduces the cooling requirement per unit of electricity generated. -This enables heat rate improvements for power plants represented in MESSAGE to be translated into -improvements in water intensity. Water withdrawal and consumption intensities for power plant -cooling technologies are calibrated to the range -reported in Meldrum et al., (2013) :cite:`meldrum_2013`. Additional parasitic electricity demands from recirculating -and dry cooling technologies are accounted for explicitly in the electricity balance calculation. All -other technologies follow the data reported in Fricko et al. -(2016) :cite:`fricko_2016`. - -A key feature of the implementation is the representation of power plant cooling -technology options for individual power plant types (:numref:`fig-cooling_implement1`). -Each power plant type that requires cooling in MESSAGE -is connected to a corresponding cooling technology option (once-through, recirculating or -air cooling), with the investment into and operation of the cooling technologies included in the -optimization decision variables :cite:`parkinson_2019`. This enables MESSAGE to choose the type of cooling technology -for each power plant type and track how the operation of the cooling technologies impact water -withdrawals, return flows, thermal pollution and parasitic electricity use. - -.. _fig-cooling_implement1: -.. figure:: /_static/cooling_implement1.png - :width: 820px +.. _water_nexus: + +Water-Energy-Land Nexus (MESSAGEix-Nexus) +****************************************** + +The MESSAGEix-GLOBIOM nexus module (MESSAGEix-Nexus) integrates water sector representation and climate impacts into the |MESSAGEix|-GLOBIOM integrated assessment modeling framework (Awais et al., 2024 :cite:`awais_2024_nexus`). This comprehensive nexus implementation enables consistent analysis of interdependencies between water, energy, and land systems under different climate and socioeconomic scenarios. + +MESSAGEix-Nexus builds upon earlier water-energy linkages (Parkinson et al., 2019 :cite:`parkinson_2019`; Vinca et al., 2020 :cite:`vinca_2020_nest`) by adding: + +* Basin-scale water resource representation (surface water and groundwater) +* Water demands from multiple sectors (energy, municipal, industrial, irrigation) +* Water supply technologies (surface water extraction, groundwater extraction, desalination, wastewater treatment) +* Power plant cooling technology options with explicit water-energy tradeoffs +* Climate change impacts on water availability and energy systems +* Sustainable Development Goal (SDG) constraints for water access + +Overview +======== + +The nexus module represents water resources and demands at the spatial scale of ~200 river basins globally, while maintaining consistency with the 12-region spatial resolution of MESSAGEix-GLOBIOM (R12). Water is explicitly tracked through the energy system via cooling technologies for thermal power plants, while also accounting for sectoral water demands that compete with energy sector water use. + +Basin-scale Spatial Resolution +------------------------------- + +Water resources are represented at the basin scale using a global delineation of river basins derived from HydroSHEDS (Lehner et al., 2008 :cite:`lehner_2008_hydrosheds`). Basins are mapped to MESSAGE regions through spatial intersection, enabling consistent aggregation of basin-level constraints to regional energy system decisions. This multi-scale approach captures: + +* Spatial heterogeneity in water availability within MESSAGE regions +* Local water scarcity constraints that affect technology choices +* Inter-basin water transfers where infrastructure exists +* Climate impacts on basin-specific hydrology + +:numref:`fig-global-basin-map` shows the global distribution of approximately 200 basins mapped to the 12 MESSAGE regions (R12), providing the spatial foundation for the nexus module. + +.. _fig-global-basin-map: +.. figure:: /_static/global_r12_basin_map.png + :width: 800px + :align: center + + Global basin delineation mapped to MESSAGE R12 regions. Basins are derived from HydroSHEDS and aggregated to provide spatially explicit water resource representation within the MESSAGEix-GLOBIOM framework (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Temporal Resolution +------------------- + +The nexus module can operate at annual or sub-annual (seasonal/monthly) temporal resolution. For climate impact studies, sub-annual resolution is critical to capture: + +* Seasonal variations in water availability (monsoons, snowmelt, dry seasons) +* Mismatches between seasonal water supply and demand +* Hydropower generation patterns and reservoir management +* Irrigation water requirements aligned with crop calendars + +When sub-annual time steps are defined in the MESSAGE model, the water module automatically generates water balance constraints at the corresponding temporal resolution (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Model Structure +=============== + +The water nexus implementation follows a resource-technology-demand structure analogous to the energy system representation in MESSAGEix. The conceptual framework integrates water resources, supply technologies, sectoral demands, and the linkages between water, energy, and land systems as illustrated in :numref:`fig-nexus-structure`. + +.. _fig-nexus-structure: +.. figure:: /_static/message_nexus_structure.png + :width: 800px :align: center - Implementation of cooling technologies in the MESSAGE IAM. - -Costs and efficiency for -cooling technologies are estimated following previous technology assessments :cite:`zhai_2010,zhang_2014,loew_2016`. -The initial distribution of cooling technologies in each region -and for each technology is estimated with the dataset described in Raptis and Pfister (2016) :cite:`Raptis_2016_powerplant_data`. -The shares estimated at the river basin-scale are depcited in :numref:`fig-cooling_implement2` . - -.. _fig-cooling_implement2: -.. figure:: /_static/cooling_implement2.png - :width: 820px + Conceptual structure of the MESSAGEix-GLOBIOM nexus module showing the integration of water resources, supply technologies, sectoral demands, and linkages with energy and land systems (Awais et al., 2024 :cite:`awais_2024_nexus`). + +The nexus module comprises three main components: + +**Resources**: Renewable surface water and groundwater availability in each basin and time period, derived from hydrological models. See :doc:`supply` for detailed description of water resources and supply technologies. + +**Technologies**: Water extraction, treatment, conveyance, and end-use technologies including: + +* Surface water extraction and distribution +* Groundwater extraction (with depth-dependent costs) +* Desalination (thermal and reverse osmosis) +* Wastewater treatment and reuse +* Power plant cooling technologies (once-through, recirculating, dry cooling) +* Irrigation technologies + +**Demands**: Sectoral water requirements including: + +* Energy sector (power plant cooling, fuel extraction and processing) +* Municipal and domestic water use +* Industrial manufacturing water use +* Agricultural irrigation (linked to GLOBIOM) + +See :doc:`demand` for comprehensive coverage of all sectoral water demands and allocation mechanisms. + +Water Reference Energy System +------------------------------ + +The water flows through the model are represented using a Reference Energy System (RES) structure, extending the MESSAGEix energy RES to include water commodities and technologies. :numref:`fig-water-res` shows the simplified water reference energy system structure that connects water resources to sectoral demands through various supply technologies. + +.. _fig-water-res: +.. figure:: /_static/water_reference_system.png + :width: 800px :align: center - - Average cooling technology shares across all power plant types at the river basin-scale. \ No newline at end of file + + Simplified water Reference Energy System (RES) showing the flow of water from resources (surface water, groundwater, desalination) through treatment and distribution to sectoral demands (municipal, industrial, agricultural, energy). The RES structure enables explicit tracking of water quantities, qualities, and associated costs and energy requirements (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Water balance equations ensure that total water extraction does not exceed renewable availability plus sustainable groundwater use, while meeting all sectoral demands and environmental flow requirements. The optimization simultaneously determines the least-cost portfolio of water supply technologies and the allocation of water across competing sectoral demands. + +Energy-Water-Land Linkages +========================== + +MESSAGEix-Nexus captures multiple nexus interactions: + +**Energy → Water**: + +* Cooling water requirements for thermal power plants (see :doc:`cooling` for detailed cooling technology representation) +* Water consumption in fuel extraction (coal mining, unconventional oil and gas) +* Hydropower production from surface water resources +* Energy requirements for water supply (pumping, treatment, desalination) + +**Water → Energy**: + +* Water availability constraints on thermal power plant siting and operation +* Cooling technology choices driven by water scarcity +* Hydropower generation governed by river flows and reservoir storage +* Groundwater pumping costs dependent on aquifer depth + +**Land → Water** (via GLOBIOM linkage): + +* Irrigation water demands for crop production +* Land use change impacts on runoff and water availability +* Water allocation between agriculture and other sectors + +**Climate → Water-Energy-Land**: + +* Temperature and precipitation changes affecting water availability +* Extreme events (droughts, floods) impacting all sectors +* Climate-driven changes in cooling water requirements and efficiency +* Shifts in crop water demands and irrigation needs + +Climate Change Impacts +====================== + +The nexus module incorporates climate change impacts on both water availability and energy systems (Awais et al., 2024 :cite:`awais_2024_nexus`): + +**Hydrological Impacts**: Basin-specific changes in renewable water availability derived from global hydrological models (PCR-GLOBWB, CWatM) forced by climate model outputs. Impacts include: + +* Changes in mean annual runoff and groundwater recharge +* Shifts in seasonal water availability patterns +* Increased frequency and severity of droughts +* Modified snowmelt timing in snow-dominated basins + +**Energy System Impacts**: + +* Reduced thermal power plant efficiency due to higher ambient temperatures +* Increased cooling water requirements from higher water temperatures +* Changes in hydropower generation potential +* Modified electricity demand patterns (cooling vs. heating) + +**Adaptation Measures**: The model can endogenously select adaptation measures such as: + +* Shifts to less water-intensive cooling technologies +* Investment in desalination and water reuse +* Inter-basin water transfers +* Changes in electricity generation technology mix + +For comprehensive discussion of climate change impacts on the water-energy nexus, see :doc:`climate_impacts`. + +Sustainable Development Goals +============================== + +The nexus module includes optional constraints to represent progress toward water-related Sustainable Development Goals (Awais et al., 2024 :cite:`awais_2024_nexus`): + +**SDG 6**: Clean water and sanitation + +* Targets for urban and rural water access rates +* Wastewater treatment coverage requirements +* Water use efficiency improvements + +**SDG 7**: Affordable and clean energy + +* Energy access targets requiring water for cooling and hydropower +* Trade-offs between water and energy access in water-scarce regions + +Implementation constraints enforce minimum investment in water supply infrastructure to achieve specified access targets in each region and time period, creating additional water demand and infrastructure requirements that compete with energy sector water use. + +Detailed Documentation +======================= + +For detailed technical documentation of the MESSAGEix-Nexus module components, please refer to the following sections: + +* :doc:`supply` - Water resources and supply technologies (surface water, groundwater, desalination, wastewater reuse) +* :doc:`demand` - Sectoral water demands (energy, municipal, industrial, agricultural) and allocation +* :doc:`cooling` - Power plant cooling technologies and water-energy tradeoffs +* :doc:`climate_impacts` - Climate change impacts on water availability and energy systems + +.. toctree:: + :maxdepth: 2 + :hidden: + + supply + demand + cooling + climate_impacts + +Reference +========= + +The MESSAGEix-Nexus module is described in detail in: + +Awais, M., Vinca, A., Byers, E., Frank, S., Fricko, O., Boere, E., Burek, P., Poblete Cazenave, M., Kishimoto, P.N., Mastrucci, A., Satoh, Y., Palazzo, A., McPherson, M., Riahi, K., and Krey, V. (2024). MESSAGEix-GLOBIOM nexus module: integrating water sector and climate impacts. *Geoscientific Model Development*, 17, 2447-2469. https://doi.org/10.5194/gmd-17-2447-2024 + +The NEST (Nexus Solutions Tool) framework that preceded this implementation is described in: + +Vinca, A., Parkinson, S., Byers, E., Burek, P., Khan, Z., Krey, V., Diuana, F.A., Wang, Y., Ilyas, A., Köberle, A.C., Staffell, I., Pfenninger, S., Muhammad, A., Rowe, A., Schaeffer, R., Rao, N.D., Wada, Y., Djilali, N., and Riahi, K. (2020). The NExus Solutions Tool (NEST) v1.0: an open platform for optimizing multi-scale energy–water–land system transformations. *Geoscientific Model Development*, 13, 1095-1121. https://doi.org/10.5194/gmd-13-1095-2020 + +Power plant cooling implementation is described in: + +Parkinson, S., Byers, E., Gidden, M., Krey, V., Burek, P., Vollmer, D.,Jalava, M., Palazzo, A., Graham, N., Fricko, O., Tracking the water-energy-land-food nexus: integrated assessment of sustainable development goal interactions. Submitted to *Nature Sustainability*. 2019. + +.. footbibliography:: diff --git a/doc/global/water/supply.rst b/doc/global/water/supply.rst new file mode 100644 index 0000000000..07f1c898f0 --- /dev/null +++ b/doc/global/water/supply.rst @@ -0,0 +1,268 @@ +.. _water-supply: + +Water Supply +============ + +Water supply in MESSAGEix-Nexus is represented through multiple technology options that extract, treat, and distribute freshwater from surface and groundwater sources, as well as non-conventional sources such as desalination and treated wastewater reuse (Awais et al., 2024 :cite:`awais_2024_nexus`). Each basin has specific renewable water availability derived from hydrological model outputs, which constrains total water extraction. + +Surface Water +------------- + +Surface water resources include runoff from precipitation, snowmelt, and glacier melt aggregated at the river basin scale. Surface water availability is represented as a time-varying resource potential for each basin. + +Hydrological Data Sources +^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Basin-scale surface water availability is derived from global hydrological models that simulate the terrestrial water cycle: + +* **PCR-GLOBWB 2** (Sutanudjaja et al., 2018 :cite:`sutanudjaja_2018_pcrglobwb`): A global hydrological model at 5 arcmin resolution (~10km at equator) that simulates river discharge, soil moisture, and groundwater recharge +* **CWatM** (Community Water Model; Burek et al., 2020 :cite:`burek_2020_cwatm`): A spatially distributed hydrological model representing water demand, supply, and environmental flows +* **Historical data** (1971-2000): Used for calibration and baseline water availability +* **Future projections** (2020-2100): Derived from hydrological models forced by climate model outputs from CMIP5/CMIP6 + +The hydrological model outputs provide monthly or seasonal water availability data that are spatially aggregated from grid cells to MESSAGE basins using area-weighted averages. For MESSAGE applications, seasonal or 5-yearly average values are typically used (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Temporal Variability +^^^^^^^^^^^^^^^^^^^^ + +Surface water availability exhibits strong seasonal and interannual variability: + +* **Seasonal patterns**: Monsoon regions show pronounced wet/dry seasons; snow-dominated basins have spring snowmelt peaks +* **Interannual variability**: Represented through statistical analysis of multi-year hydrological simulations +* **Climate trends**: Long-term changes in mean availability and variability under different climate scenarios +* **Extreme events**: Droughts represented as low quantiles (e.g., 10th percentile) of flow distributions + +For sub-annual MESSAGE implementations, seasonal water availability is explicitly represented. For annual implementations, average annual availability is used with optional constraints on reliability (e.g., water available in 90% of years). + +Environmental Flow Requirements +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Not all renewable surface water can be extracted for human use. Environmental flow requirements (EFRs) are subtracted from gross water availability to determine the extractable potential: + +:math:`SW_{extract,b,t} \leq SW_{available,b,t} - EFR_{b,t}` + +where :math:`SW_{extract,b,t}` is extractable surface water in basin :math:`b` and time period :math:`t`, :math:`SW_{available,b,t}` is total renewable surface water, and :math:`EFR_{b,t}` is the environmental flow requirement. + +Environmental flows are calculated using the Variable Monthly Flow (VMF) method (Pastor et al., 2014 :cite:`pastor_2014_efr`), which sets minimum flows as a percentage of mean monthly natural flow, with higher percentages for low-flow months to protect aquatic ecosystems. Typical EFR values range from 20-40% of mean annual flow depending on the basin and flow regime. + +Surface Water Extraction Technologies +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Surface water extraction is represented through technology archetypes with associated costs and infrastructure requirements: + +* **River/lake extraction**: Direct abstraction with intake structures, screening, and pumping +* **Small-scale reservoirs**: Storage for seasonal regulation and reliability +* **Large-scale reservoir storage**: Represented through hydropower technologies in MESSAGE +* **Inter-basin transfers**: Explicit connections between basins where infrastructure exists + +Extraction costs include: + +* Capital costs for intake structures, pumps, and basic treatment +* Operating costs for energy (pumping), maintenance, and operation +* Conveyance costs proportional to distance from source to demand location +* Treatment costs to achieve required water quality + +Typical costs range from 0.01-0.05 USD/m³ for surface water extraction and basic treatment (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Groundwater +----------- + +Groundwater provides a critical buffer against surface water variability and is explicitly represented in MESSAGEix-Nexus with depth-dependent extraction costs and sustainability constraints. + +Groundwater Resources +^^^^^^^^^^^^^^^^^^^^^^ + +Groundwater resources are characterized by: + +* **Renewable groundwater**: Annual recharge from precipitation infiltration and river seepage +* **Non-renewable (fossil) groundwater**: Deep aquifers with negligible recharge on human timescales +* **Groundwater storage**: Cumulative volume in aquifers (not fully represented in current implementation) + +Renewable groundwater recharge is derived from the same hydrological models as surface water (PCR-GLOBWB, CWatM), which simulate infiltration, percolation, and recharge processes. Basin-scale recharge rates are typically 10-30% of precipitation in humid regions and <5% in arid regions. + +Groundwater Extraction +^^^^^^^^^^^^^^^^^^^^^^ + +Groundwater extraction costs depend on: + +1. **Aquifer depth**: Pumping costs increase with depth (energy requirements) +2. **Extraction rate**: Higher rates require more/deeper wells +3. **Water quality**: Treatment requirements for brackish or contaminated groundwater + +The extraction cost function is represented as: + +:math:`Cost_{GW} = c_0 + c_1 \cdot d + c_2 \cdot d^2` + +where :math:`d` is the effective extraction depth and :math:`c_0`, :math:`c_1`, :math:`c_2` are cost parameters. Depths range from shallow (<50m) to deep (>500m) groundwater. + +Energy requirements for groundwater pumping create a water-energy feedback loop: + +:math:`E_{pump} = \dfrac{\rho \cdot g \cdot d \cdot V}{\eta}` + +where :math:`E_{pump}` is pumping energy, :math:`\rho` is water density, :math:`g` is gravitational acceleration, :math:`d` is depth, :math:`V` is volume pumped, and :math:`\eta` is pump efficiency (~0.6-0.8). + +Typical groundwater extraction costs range from 0.02 USD/m³ for shallow groundwater to 0.30 USD/m³ for deep groundwater (Awais et al., 2024 :cite:`awais_2024_nexus`), plus energy costs for pumping. + +Groundwater Sustainability Constraints +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Groundwater extraction is constrained to sustainable levels to prevent aquifer depletion: + +:math:`\sum_{t'=t_0}^{t} GW_{extract,b,t'} \leq \sum_{t'=t_0}^{t} GW_{recharge,b,t'} + GW_{buffer,b}` + +This ensures that cumulative extraction does not exceed cumulative recharge plus an allowable buffer representing accessible storage. This constraint prevents the model from mining groundwater unsustainably, which is a major concern in regions such as: + +* Northwest India and Pakistan (Indus-Ganges basin) +* North China Plain +* Arabian Peninsula +* High Plains Aquifer (USA) +* Mexico City basin + +Aquifer Storage and Recovery +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +In some basins, managed aquifer recharge (MAR) technologies are represented: + +* **Excess surface water** during wet periods can be used to recharge aquifers +* **Stored water** can be extracted during dry periods or drought +* Provides a form of inter-seasonal and inter-annual water storage + +This technology is particularly valuable in basins with strong seasonal variability and available aquifer storage capacity. + +Desalination +------------ + +Desalination technologies convert saline water (seawater or brackish groundwater) into freshwater, providing a climate-independent water source for coastal regions. Desalination is critical for water-scarce regions and is explicitly represented in MESSAGEix-Nexus (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Desalination Technologies +^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Two main desalination technology categories are represented: + +**Reverse Osmosis (RO)**: Membrane-based separation + +* Lower energy consumption: 3-4 kWh/m³ for seawater, 1-2 kWh/m³ for brackish water +* Requires electrical energy (high-quality energy) +* Modular and scalable +* Suitable for small to large plants +* Current technology of choice for new capacity + +**Thermal Desalination**: Evaporation-based processes (MSF, MED) + +* Higher energy consumption: 15-25 kWh/m³ thermal energy equivalent +* Can use waste heat from power plants (cogeneration) +* Historically dominant, now mostly in Middle East +* Often coupled with thermal power generation + +The technology choice depends on: + +* Availability of waste heat from power generation +* Cost of electricity vs. thermal energy +* Plant size and water demand patterns +* Feedwater salinity and quality + +Energy Requirements and Costs +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Desalination is energy-intensive, creating a water-energy nexus feedback: + +* **RO energy**: 3-4 kWh_e/m³ for seawater (~0.50-0.70 USD/m³ at typical electricity prices) +* **Thermal desalination**: 50-80 MJ_th/m³ heat (~0.30-0.50 USD/m³ with waste heat) +* **Additional costs**: Chemicals, membranes, maintenance, brine disposal + +Total levelized costs for desalinated water: + +* Seawater RO: 0.50-1.50 USD/m³ (decreasing with technology improvements) +* Brackish RO: 0.30-0.80 USD/m³ (lower salinity = lower costs) +* Thermal desalination: 1.00-2.50 USD/m³ (decreasing with scale) + +Costs have declined significantly (>50% reduction since 2000) due to: + +* Improved membrane technology and energy recovery devices +* Economies of scale in large plants +* Operational experience and optimization + +Regional Availability +^^^^^^^^^^^^^^^^^^^^^ + +Desalination is only available in basins with access to: + +* **Coastal regions**: Seawater desalination +* **Inland brackish groundwater**: Brackish water desalination + +The model includes geographical constraints limiting desalination to appropriate basins. Transport costs increase with distance from coast to demand centers. + +Current and Projected Capacity +^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ + +Historical desalination capacity (base year ~2020): + +* Global total: ~100 million m³/day +* Middle East and North Africa: ~70% of global capacity +* Growing rapidly in water-scarce regions + +Projected capacity expansion is endogenous in MESSAGEix-Nexus based on: + +* Water scarcity and availability of alternatives +* Energy costs and availability +* Economic development and water demands +* Climate change impacts on conventional water sources + +In water-stressed scenarios, desalination can grow to provide 10-20% of urban water supply in coastal MESSAGE regions by 2050-2100 (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Wastewater Treatment and Reuse +------------------------------- + +Treated wastewater provides an additional water source, particularly for non-potable uses such as industrial cooling, irrigation, and environmental flows. + +Treatment Technologies +^^^^^^^^^^^^^^^^^^^^^^ + +Multiple treatment levels are represented: + +* **Primary treatment**: Solids removal (~30% pollutant removal) +* **Secondary treatment**: Biological treatment (~85% pollutant removal) +* **Tertiary treatment**: Advanced treatment for reuse (~95% pollutant removal) + +Energy and cost requirements increase with treatment level: + +* Primary: 0.1-0.2 kWh/m³, 0.02-0.05 USD/m³ +* Secondary: 0.3-0.6 kWh/m³, 0.10-0.20 USD/m³ +* Tertiary: 0.5-1.0 kWh/m³, 0.30-0.60 USD/m³ + +Reuse Applications +^^^^^^^^^^^^^^^^^^ + +Treated wastewater can be used for: + +* **Industrial cooling**: Requires secondary treatment +* **Agricultural irrigation**: Requires secondary or tertiary treatment depending on crop type +* **Environmental flows**: Return to rivers with minimum treatment +* **Groundwater recharge**: Requires tertiary treatment +* **Potable reuse**: Requires advanced treatment (not currently represented) + +The economic attractiveness of wastewater reuse depends on: + +* Cost of alternative water sources +* Stringency of discharge regulations +* Proximity of treatment plant to reuse location +* Seasonal patterns of supply and demand + +Water reuse can provide 5-15% of total water supply in water-scarce urban regions (Awais et al., 2024 :cite:`awais_2024_nexus`). + +Water Supply Portfolio +----------------------- + +The model endogenously selects the optimal portfolio of water supply technologies based on: + +* Resource availability and variability +* Technology costs and energy requirements +* Water quality requirements for different demands +* Infrastructure constraints and existing capacity +* Climate change impacts on conventional sources +* Sustainability constraints on groundwater use + +In baseline scenarios, surface water typically provides 60-80% of total supply, groundwater 20-35%, and desalination/reuse 0-10% globally. In water-stressed climate scenarios, these shares shift substantially toward groundwater and desalination (Awais et al., 2024 :cite:`awais_2024_nexus`). + +.. footbibliography:: + diff --git a/doc/main.bib b/doc/main.bib index 81b8acf99d..0a6d5fe1cf 100644 --- a/doc/main.bib +++ b/doc/main.bib @@ -1607,6 +1607,89 @@ @article{Raptis_2016_powerplant_data year = {2016} } +@article{awais_2024_nexus, + author = {Awais, Muhammad and Vinca, Adriano and Byers, Edward and Frank, Stefan and Fricko, Oliver and Boere, Esther and Burek, Peter and Poblete Cazenave, Miguel and Kishimoto, Paul Natsuo and Mastrucci, Alessio and Satoh, Yusuke and Palazzo, Amanda and McPherson, Madeleine and Riahi, Keywan and Krey, Volker}, + doi = {10.5194/gmd-17-2447-2024}, + journal = {Geoscientific Model Development}, + pages = {2447--2469}, + title = {{MESSAGEix-GLOBIOM nexus module: integrating water sector and climate impacts}}, + volume = {17}, + year = {2024} +} + +@article{vinca_2020_nest, + author = {Vinca, Adriano and Parkinson, Simon and Byers, Edward and Burek, Peter and Khan, Zarrar and Krey, Volker and Diuana, Fabio A. and Wang, Yue and Ilyas, Asif and Köberle, Alexandre C. and Staffell, Iain and Pfenninger, Stefan and Muhammad, Awais and Rowe, Andrew and Schaeffer, Roberto and Rao, Narasimha D. and Wada, Yoshihide and Djilali, Ned and Riahi, Keywan}, + doi = {10.5194/gmd-13-1095-2020}, + journal = {Geoscientific Model Development}, + pages = {1095--1121}, + title = {{The NExus Solutions Tool (NEST) v1.0: an open platform for optimizing multi-scale energy–water–land system transformations}}, + volume = {13}, + year = {2020} +} + +@article{sutanudjaja_2018_pcrglobwb, + author = {Sutanudjaja, Edwin H. and van Beek, Rens and Wanders, Niko and Wada, Yoshihide and Bosmans, Joyce H. C. and Drost, Niels and van der Ent, Ruud J. and de Graaf, Inge E. M. and Hoch, Jannis M. and de Jong, Kor and Karssenberg, Derek and López López, Patricia and Peßenteiner, Stefanie and Schmitz, Oliver and Straatsma, Menno W. and Vannametee, Ekkamol and Wisser, Dominik and Bierkens, Marc F. P.}, + doi = {10.5194/gmd-11-2429-2018}, + journal = {Geoscientific Model Development}, + pages = {2429--2453}, + title = {{PCR-GLOBWB 2: a 5 arcmin global hydrological and water resources model}}, + volume = {11}, + year = {2018} +} + +@article{burek_2020_cwatm, + author = {Burek, Peter and Satoh, Yusuke and Kahil, Taher and Tang, Ting and Greve, Peter and Smilovic, Mikhail and Guillaumot, Luca and Zhao, Fang and Wada, Yoshihide}, + doi = {10.5194/gmd-13-3267-2020}, + journal = {Geoscientific Model Development}, + pages = {3267--3298}, + title = {{Development of the Community Water Model (CWatM v1.04) – a high-resolution hydrological model for global and regional assessment of integrated water resources management}}, + volume = {13}, + year = {2020} +} + +@article{lehner_2008_hydrosheds, + author = {Lehner, Bernhard and Verdin, Kristine and Jarvis, Andy}, + doi = {10.1029/2008EO100001}, + journal = {Eos, Transactions American Geophysical Union}, + number = {10}, + pages = {93--94}, + title = {{New global hydrography derived from spaceborne elevation data}}, + volume = {89}, + year = {2008} +} + +@article{pastor_2014_efr, + author = {Pastor, A. V. and Ludwig, F. and Biemans, H. and Hoff, H. and Kabat, P.}, + doi = {10.5194/hess-18-85-2014}, + journal = {Hydrology and Earth System Sciences}, + pages = {85--108}, + title = {{Accounting for environmental flow requirements in global water assessments}}, + volume = {18}, + year = {2014} +} + +@article{macknick_2012, + author = {Macknick, Jordan and Newmark, Robin and Heath, Garvin and Hallett, KC}, + doi = {10.1088/1748-9326/7/4/045802}, + journal = {Environmental Research Letters}, + number = {4}, + pages = {045802}, + title = {{Operational water consumption and withdrawal factors for electricity generating technologies: a review of existing literature}}, + volume = {7}, + year = {2012} +} + +@article{satoh_2022_drought, + author = {Satoh, Yusuke and Yoshimura, Kei and Pokhrel, Yadu and Kim, Hyungjun and Shiogama, Hideo and Yokohata, Tokuta and Hanasaki, Naota and Wada, Yoshihide and Burek, Peter and Byers, Edward and Schmied, Hannes Müller and Gerten, Dieter and Ostberg, Sebastian and Gosling, Simon N. and Boulange, Julien E. S. and Oki, Taikan}, + doi = {10.1038/s41467-022-30729-2}, + journal = {Nature Communications}, + pages = {3287}, + title = {{The timing of unprecedented hydrological drought under climate change}}, + volume = {13}, + year = {2022} +} + + @article{huppmann_2019_messageix, author = {Huppmann, Daniel and Gidden, Matthew and Fricko, Oliver and Kolp, Peter and Orthofer, Clara and Pimmer, Michael and Kushin, Nikolay and Vinca, Adriano and Mastrucci, Alessio and Riahi, Keywan and Krey, Volker}, doi = {10.1016/j.envsoft.2018.11.012},