End-point impacts of ERW supply chain
End-point impacts per hectare (hectare), which are a combination of 18 mid-point environmental impacts category results across all supply chains, show the relative sustainability of ERW in each of the twelve countries included in our analysis (Fig.1, Tab.1). To assess the potential resource depletion (RE), ecosystems destruction (EC), and human health (HH), impacts are scaled from 0 – 120 points per hectare (points/ha).1 CDR. CDR.
Plots depict the three end points AdDepletion natural resources ehEcosystems and ilEnd-points on human health in 12 countries that have the potential to produce net 0.52.0Gt CO.2Yr1CDR under business as usual (BAU), and 2C energy policy scenarios. End-point impacts were calculated using 18 mid-point indicators and the LCA ReCiPe methodology based on 0120 weighted unit and points/ha functional units. 120 is the highest impact per hectare. Zero indicates no impacts.
Resource depletion (RE)
Resource depletion impacts per unit area quantify the degree to which abiotic resources are exhausted due to increased extraction and mining of metals and minerals and the depletion or primary energy carriers, such as oil, coal, and gas. They also include water consumption in the ERW supply chain processes (Table1). No matter what energy scenario, China, India, Poland and Germany have the highest RE values. Brazil, France and Canada have the lowest values (Fig.1ad). This pattern is consistent with the CDR goal increasing from 0.5 Gt CO to 0.5 Gt.2yr1To 2.0 Gt CO2yr1This coincides with the country’s electricity sector carbon emissions (Supplementary Figures.4 & 5) because metal and mineral depletion are higher in electricity grids that rely more on coal and gas. RE is decreasing in most countries, except China, as the transition from undertaking ERW and BAU to the 2C Energy scenario reduces it. Low-carbon electricity grids and cleaner future energy lead to reductions in RE. China, however, is given a higher contribution to the global CDR goal under the 2C scenario due to its cropland area, and the electricity capacity it has for grinding.5. This higher CDR goal will be met by increasing the grinding energy for smaller, more weathering-efficient particles, at the cost of greater RE.
Ecosystem loss (EC)
The ecosystem loss indicator aggregates impacts categories related to climate change and land use. It also includes ecotoxicity, ecotoxicity, and eutrophication (Table1). Our analysis shows that EC impacts are generally less than those for RE. There is less variation between countries because the majority EC midpoints (i.e. The constituent processes of EW are not likely to have any significant impact on acidification, ecotoxicity, or eutrophication (Fig.1eh). The EC impacts of ERW supply chains are similar in both energy scenarios. They range from 24 to 61 points/ha (BAU) and 25 to 65 for the 2C scenarios. Average impacts are 42 to 45 points/ha (BAU) and 2C scenarios. These results indicate that transitioning to clean energy has minimal impact on ERW supply chain impacts. This is due to the fact that Global Warming Potential (GWP100) is the only scenario dependent mid-point that contributes towards the EC end point.
Human health (HH)
Combining mid-point categories such as climate change, human toxic, ionising radiation and ozone depletion to produce health end-point impacts on the human body (Table1) results in human health (Table1).22. HH stands for the cumulative negative effects of the ERW-supply chain. (Fig.1il shows that HH impacts are highest in India, USA and China, Poland, and Germany, while they are lowest in France, Brazil and Spain, Indonesia and Italy. All HH impacts scores are significantly reduced in the 2C scenarios, especially for the USA, India and Germany, Mexico, Canada, Indonesia, and Canada, after decarbonisation. On average, HH values decrease from 22106 points/ha (BAU) to 1880 point/ha (2C scenario), with average impacts of 62 & 47 points/ha, respectively, for theBAU & 2C scenarios. As with RE, higher HH impact points to countries that rely on fossilfuels (Supplementary Figs.4 & 5) because of the inclusion of GWP100 middle-point. Despite being second in fossil fuel emissions per unit energy behind India and Indonesia, Indonesia ranks low for HH impacts. This is due to its tropical climate, perennial crops, and higher weathering rate. This means that less grinding is required in order to achieve an equal amount ofCDR.
Synthesis
End-point LCA results for nation-by-nation show a consistent grouping (Poland Germany, India, USA, China) that has the highest ERW supply-chain environmental impacts per unit area. There is another group (France Spain, Canada, Brazil) that has much lower impacts. Nations that use cleaner, lower-carbon energy to power ERW deployment are likely to reduce resource depletion and human health supply chain impacts. Ecosystem loss impacts, however, are less variable between countries and less affected by the choiceofenergy scenario.
Analysis of mid-point drivers
Next, we’ll look at the midpoint LCA indicators that quantify the environmental effects of ERW supply-chain processes per unit area (SupplementaryFig.7) and their response during the transition from business-as usual (BAU), to the 2C energy scenario. (Table1). These potential impacts, which are expressed as mid-point indicators in LCA, are contributory drivers to three end-point indicators (Resource Loss, RD, Ecosystem Loss and Human Health, EC), that provide relative indicators of sustainability for the ERW supply chains of nations (Fig.
Mid-point drivers for resource depletion (RE) impacts
FDP (Fossil fuel Depletion Potential) is highest in Germany and Poland, which contributes to high RE scores (Supplementary Figure7; Supplementary Table8). FDP scores are generally lower in the 2C than in the BAU energy scenario because of lower carbon emissions from mining and rock grinding. France has the lowest FDP impacts of all twelve nations. This is because nuclear energy reduces the depletion in fossil fuel resources along the ERW supply chain. The 2C energy scenario has a higher Metal Depletion Potential (MDP), which is due to the increased demand for raw material for electrification.24. The USA, Canada and Mexico have the highest MDP impacts per hectare, and Indonesia, China, Italy and France the lowest impacts(Supplementary Fig.7; Supplementary Table9). France has the highest Water Depletion Potential values (WDP) (Supplementary Tab.10). This is due to the large amount of nuclear energy in its electricity mix. (Supplementary Figure.4). It is used for comminution processes.
Mid-point drivers of ecosystem loss (EC)
The EC endpoint indicator is generally lower than the RE and HH indicators and more variable between countries. Three land-use indicators are responsible for determining EC environmental impacts (Supplementary Fig.7, Supplementary Tables1113). These are Natural Land Transformation Potential(NLTP), Agricultural Land Occupation Potential(ALOP), and Urban Land Occupation Potentials (ULOP), whose impacts tend to be dominated by mining (Supplementary Fig.7). This analysis considers the mining impact on hydrology, land-use change, and mine tailings. These mid-point indicators of environmental impact are generally low with small increases in energy scenarios BAU and 2C.
Moreover, ERW supply chain processes also contribute to EC by acidification freshwater and marine ecosystems (Supplementary table14), and ecotoxicity (FETPinf), of freshwater (METPinf ecosystems), (Supplementary tables1517). TAP100 values drop by 38%, METP decreases by 29%, and FETP drops by 30% when we move from BAU to 2C (Supplementary Figure 7), as a result of average CDR goals and countries. The transition from fossil fuels to freshwater causes a 30% decrease in the eutrophication of freshwater and marine ecosystems (SupplementaryFig.7, Supplementary Tables1819). GWP100, which is ERW supply chain CO, also has an impact on ecosystems.2Equivalent emissions. Averaged across all CDR goals (Supplementary Fig.7), GWP100 is cut by 25% during the transition from theBAU scenario to the2C scenario. However, GWP100 is significantly higher in India than in any other country, and even lower in France or Brazil. These patterns are indicative of energy demand and mix, as well as relative rock dust transport distances.
Mid-point drivers of human health (HH) impacts
The Human Health (HH), end-point indicator, is based on four mid-point impact categories. The Human Toxicity Potential is (HTPinf), which is a measure of human health, decreases by 34% between BAU and 2C on average (Supplementary Figure7). However, the greatest impact is in India (SupplementaryTables2025). We predict an average 32.4% increase in Ionising Radiation Potential from BAU to 2C scenarios due to increased comminution, nuclear energy adoption under 2C, and increased comminution. France has the highest IRP_HE due to nuclear power. Photochemical oxygen formation potential (POFP), is reduced by 21% from BAU to 2C across CDR targets (Supplementary Figure 7.7) and linked to transport and comminution. The greatest impacts are found in Poland, while Brazil and France have the lowest. Ozone Depletion Potential, (ODP), increases by 30% when OAU is converted to 2C. BRA scenarios are also possible due to increased comminution requirements.
Analysis of the contribution of supply chain processes to the impact on the supply chain
Mid-point environmental impacts from ERW supplychain processes are summed to determine their cumulative contribution. The percentage is expressed as a percentage of each category in each ofthetwelve nations or two energy scenarios (Supplementary Fig.6). Because values are relative, however, the contribution analysis is not sensitiveto the energy scenario.
We focus on the countries with the greatest CDR potential: China, India, and the USA. China’s mid-point indicators categories are dominated by comminution, which accounts for 4080% of total impacts, except for ALOP/NLTP, where mining accounts to 90100%. India and the USA have similar process contribution fingerprints to all mid-point indicator categories, with a larger contribution of 2040% by road and railway transportation for 15 indicators (compared with China).
Canada’s contribution to the second bloc of nations is up to 60% for certain categories of rail and road transportation (e.g. MEP, ULOP and POFP are all indicative of the long distances involved with ERW deployment. A similar rise in transportation is also seen in India and the USA where the distances between quarry and field are large (Supplementary Fig.3). Brazil’s total impact potensios is dominated by mining and comminution. Indonesia is a country where the distances between road and rail rock dust transport are short. Comminution processes dominate impact categories. They account for 6080% of total impacts. However, there are important relative contributions from rock powder spreading to total MDP or ODP impacts. Mexico and Germany share similar supply chain process contribution fingerprints, with impact category totals dominated by comminution. Spain, France and Italy share similar contributions to supply chain processes to potential environmental effects. However, there is a greater proportion of road and rail transport to the mid-point indicators.
The contribution analysis suggests that comminution (i.e. The major contributors to potential health and environmental impacts for all 12 countries are rock grinding and transport of rock dust from the mines to the fields. Transportation impacts and rock grinding are related to energy and fuel needs, respectively. These findings support and extend previous LCA studies regarding regional ERW deployment in Sao Paulo State (Brazil).16. However, mining is a consistent contributor for all nations to the NLTP and ALOP categories.
The ERW supply chain is affected by the national LCA
CDR technologies will be evaluated first on their potential effects on the overall carbon budget, and then weighed against their critical land and water resource requirements, which could potentially conflict with food production.25. We therefore examine the country-level area-integrated total impacts for Global Warming Potential and Natural Land Transformation Potential (NLTP), as well as Water Depletion Potential and (WDP) at the country level, compared to their respective CDR/cropland contribution (Fig.2).
Results are shown AcGlobal Warming Potential (GWP100), relative net CDR for each of 12 countries; dfNatural Land Transformation (NLTP). giWater Depletion (WDP), NLTP, and WDP are shown relative to the cropland area used in ERW. Results are shown for 12 countries that have the potential to collectively produce net 0.52.0 Gt CO.2Yr1Carbon dioxide removal under business as usual (BAU), and 2C energy policy scenarios5. There are three levels of countries: high (USA, IND. CHN), medium(MEX, CAN. IDN. BRA), and low fossilfuel CO.2 emitters (POL, ITA, ESP, DEU, FRA). IND India, MEX Mexico, CAN Canada, IDN Indonesia, BRA Brazil.
Global warming potential (GWP100).
We report national GWP100 numbers calculated as the increase in radiative forcing over a 100 years time horizon for CO2Non-CO and CO2Greenhouse gas emissions related to the ERW supply chains. GWP100 represents the CO2Similar footprint for cradle to grave captured CO2. Figure2 scales GWP1002eqyr1) against the corresponding ERW net CO2To facilitate a comparative understanding on the relative impacts LCA has on ERW net CDR, each country must be removed. Supply chain CO2The LCA methodology reveals that emissions from rock dust spreading, mining, and grinding are comparable to those previously accounted for5.
The highest GWP100 value (i.e. Supply chain CO2emissions) are associated to China, India, and the USA, i.e. These are the nations with the highest CDR potential via ERW (Fig.2ac). These nations have a lower GWP100 value than the BAU scenario. This is due to the shift away from fossil fuels (Supplementary Fig.7) and subsequent reductions in comminution, and transport emissions. The GWP100 values of the second bloc of countries (Mexico Canada, Indonesia, Brazil) and European countries Poland, Italy, Spain and Germany are lower than the BAU energy scenario. Poland and Germany have the highest electricity sector carbon emissions and therefore the largest supply chain emission. However, they also have the greatest potential for improvement based upon 2C energy decarbonisation pathways for 2050.
Natural land transformation poten (NLTP)
NLTP is the conversion of land cover from one type into another.26,27ERW is mainly due to the opening of new mines. Consequently, NLTPs are roughly proportional the corresponding total rocks demand for each country, which scales with land area (rock application rate remains constant at 40 t ha).1Fig.2df), with China, India, and the USA having the highest values. The NLTPs of the second bloc of countries decrease by a factor three and by a factor ten in European countries due to the smaller cropland areas. As with GWP100 NLTP is decreased in the largest nations when switching from the BAU to 2C energy scenario. Cleaner electrical energy reduces emission penalties by grinding smaller particle sizes. This improves weathering efficiency, and allows for equivalent levels CDR with reduced rock requirements. This does not apply to countries such as Spain or Poland, which are already limited in electricity available for grinding.5. We find that ERW-linked NLTP could result in a maximum of 0.05% agricultural land loss for any one of the twelve countries considered. This supports its sustainability.
Water depletion potential (WDP)
WDP is the water consumed by ERW supply chain processes. It does not include water from the source of origin. WDP increases roughly in proportion to CDR potential across all twelve countries. WDP impacts in the USA, India, and China are simulated as being 10-fold greater than those of the other nations (Fig.2g and h). This is due to the large cropland area and large amounts of rock needed to mine and crush these countries for ERW deployment. The results show that there is potential to reduce WDP impact through a transition towards clean energy and a reduction of rock demand as weathering improves. Overall, however, we find that water depletion from ERW supply chain processes accounts for a very small proportion of freshwater resources (0.25%), in each of 12 nations (Supplementary Tab27).
Synthesis
Two general results of area-integrated national environmental effects of ERW are consistent across countries, energy supply scenarios, and global CDR goals. First, the national GWP and NLTP impacts scale by CDR. The nations that conduct ERW practices at a large spatial scale are simulated to have the highest environmental impacts. These impacts are reduced by switching from BAU energy scenarios to 2C energy. Lower electricity emissions enable additionalrock grinding to produce smaller particle sizes. This improves weathering efficiency and reduces rock/land consumption. This allows for lower impacts in all three categories. This is a pathway to increasing the sustainability and scale of ERW.
Carbon capture efficiency of ERW
We calculate the carbon removal effectiveness (CO2) to quantify the performance and economics of ERW.CO2), defined by net CDR as a percentage of gross CDR, where net CDR is gross CDR minus supply chain CO2 emissions. The results show that the ERW/CDR efficiency scales linearly to the carbon footprint of the projected electricity supply for 2050 across twelve nations (Fig.3). Nations with high ERW/CDR potential and high CO2There is potential for rising emissions from electricity generation (China and India, USA). CO2By switching from the BAU scenario to the 2C scenario with an aless carbon-intensive energie mix, you can increase your GDP by 1020% These nations would not be able to make this transition without it. CO2Values of around 0.75 are consistent with earlier estimates for the carbon penalty from supply chains processes5,11,17. Countries with low CO2 levels2Already high emissions have been recorded from power generation using the BAU energy blend CO2There is little scope for improvement. These countries include Canada, France, Brazil and France, each with a dominant low-carbon power source (hydro-, nuclear, and natural gas), (Supplementary Figure.4)
Each nation’s efficiency of ERW–CDR is shown under business-as usual (BAU) and 2C energy policies for 2050 (vector tail). Shaded areas indicate CDR efficiency of direct-air capture plants powered either by waste heat or geothermal heat.28.
Over a wide range national electricity CO2 emissions, CO2Regardless of whether electricity to power DAC comes from geothermal energy, or from waste heat from incineration plant, the ERW rate is generally higher than direct air capture.28 (Fig.3). Despite having a small electricity footprint, however the efficiency of both CDR strategies converges. The DAC plants in Hellisheii or Hinwil have the potential for reaching CO2Values of 93.1%, 85.4% and respectively28.
Comparison of ERW and other NETs
Analyzing the ERW supply chains potential impacts allows for a wider comparison to other NETs deployed in the same range (0.52.0 GT CO)2yr1)29. The lack of consistency between the methods and impact categories can make direct comparisons difficult.30However, it is possible to use standardised metrics for energy, land-use, and water-use (Fig.4). Literature values are linear functions (or CDR) across all three metrics. Our analysis is different because it shows the diminishing CDR returns with increasing geographical deployment of EW and prime cropland areas for high ERW.
A Energy, bLand and cWater requirements for a range CDR strategies across a similar CDR range. Enhanced rock weathering*) this study, ERW prior work11, Afforestation (AR), Direct Air Capture(DAC), Bioenergy (BECCS), and Bioenergy with Carbon Capture & Storage (BECCS).29,31. In bCropland area is only shown for comparison. ERW deploys on existing lands so no additional land is necessary.
Our energy requirements estimates are consistent in the mid-range of previous ERW studies.11, rising from 3-10 EJ yr1Over the 0.52 Gt CO2Yr1CDR range (Fig.4a). Direct Air Capture, on the other hand (DAC) requires twice as much energy to reach the same CDR range.29. Other NETs, like Bioenergy with Carbon Capture and Storage(BECCS), are also positive in terms of net energy balance.29.
The land-use requirements (cropland), to ERW can be calculated by adding the NLTP (mainly resulting from mining activities) and the cropland required for deployment (Fig.4b). Only the former should be considered an additional requirement. ERW has a significant advantage over other land-based CDR strategies in that carbon sequestration does not compete with other land uses (e.g. crop production and the loss of biodiversity and natural habitats.6,25,27. ERW is smaller than any other NETs because it focuses on the NLTP. DAC is not a land-based CDR tech, but it does require land to site the CO at industrial scale.2removal units, and for additional power to drive the DAC system and storage. Large-scale deployment of DAC for CO capture2yr1Required for 29km2To site the 3683 DAC units, and additionally between 445km2(wind) and 4450km2(Photovoltaics), of land for electricity generation depending on the energy source28. As with Biochar,31Unlike BECCS or DAC, ERW doesn’t require complex industrial infrastructure development to capture and transport the CO.2Sub-surface reservoirs can have environmental impacts on supply chains.28.
Our results show that ERW deployments resulting from LCA are more water-intensive than previously estimated, even though they were not quantified.11,29. Nevertheless, our estimates are significantly lower than those for AF/RF/BECCS and DAC across similar CDR goals (Fig.4c). These alternativeNETs require large amounts of water in order to support tree and bioenergy crop production. DAC however has a moderate-to–high water requirement with significant uncertainty. Other NETs such as soil carbon sequestration and biochar have very low water requirements.
Our comparison of the energy, water, and land requirements of ERW to CDR shows that ERW is sustainable and competitive. ERW requires half the energy demand of DACS.It needs more land than other land-based technologies (BECCS/AF/RF/biochar), but avoids land-use competition. ERW also has a 10100 times lower water consumption than other CDR strategies.