The common use of liming agents in agriculture (wherein liming with limestone/dolomite is typically a net CO₂ emitter (West and McBride, 2005)) is the most direct path for silicate amendments, considering the available supply chain, equipment for spreading, budget for soil amendment, scheduling opportunity, and long-term experience with mineral amendments. Notably, there are clear grower recommendations on the amendment rate of liming agents based on their liming index (Reid et al., 2006), yet the liming index of silicate amendments is unknown, and available studies on the use of silicates as liming agents are often inconclusive about their liming effects and effects on soil carbon (Amoakwah et al., 2023), wherein overliming with silicates (Yan et al., 2023) can cause the same short-term loss of soil organic carbon as for conventional liming (Paradelo et al., 2015). This calls for measured amendment approaches when considering the need for a sufficient silicate amendment to enable MRV and correct dosage to maximize agronomic benefits and total soil carbon storage.

The field of soil remineralization is being developed alongside the technology of ERW and has a longer existence, considering that rock dusts have historically been used worldwide as a locally available low-cost fertilizing amendment (van Straaten, 2007). Many silicate rocks contain appreciable amounts of plant nutrients such as P and K, which can be mobilized during weathering; hence, their use to reduce (at least partially) the use of artificial fertilizers can have additional emission reduction consequences. An often-overlooked non-essential plant nutrient is silica, which is also released in a reactive form (colloidal silica or silicic acid) during weathering and can be beneficial for crops that take up silica for tissue strength (Jariwala et al., 2022; Swoboda et al., 2022).

A continued concern about the scalability of ERW is the sourcing of suitable minerals and the added environmental impact of new mining operations. There are global stockpiles of mine waste, from quarrying activities, ore mining activities, and metallurgical activities, that have weathering potential and suitable safe composition (Paulo et al., 2021), and those are low-hanging fruits, apart from possible regulatory barriers on waste use in food production, to accelerate ERW deployment.

Despite representing a small fraction of global farming operations, organic farming calls for the use of natural soil amendments. Silicate minerals, as liming agents, fertilizing agents, or solely for carbon sequestration, can qualify for organic use. For example, wollastonite used in the GeoRewind work is organic-certified (Organic Materials Review Institute, 2015).

Many farmers are used to liming; hence, there is a low barrier to entry for ERW if the same conventional logistics are used to deliver and spread minerals. Notably, farming experience should trump scientific advice that, at times, propose very large amendment rates (Kelland et al., 2020). According to the Ontario Agricultural College Dean Prof. Rene Van Acker, “If the technology does not take into account the practical reality of farming, that’s a big mistake” (Grip Workshop, 2023).

It remains unclear if ERW verification should rely on soil water analysis (e.g., for alkalinity, electrical conductivity, isotope, or trace metal analysis), soil solid analysis (e.g., calcimetry, thermogravimetry, X-ray diffraction, and X-ray fluorescence), gas analysis (e.g., soil P _CO2 and eddy covariance), or even remote sensing (Almaraz et al., 2022). Scientifically, an all-of-the-above approach is useful for building confidence in the mechanisms of ERW and the fate of weathering products and residues, but practically, it remains to be determined what a feasible, scalable, and equitable form of verifying ERW rates and extent is. Eventually, as with other farming practices, ERW may become a guideline-based practice rather than verification-based technology.

Efforts have been made to model ERW from the scale of a single mineral grain (Georgakopoulos et al., 2016) to a finite portion of soil (column or a field) (Kelland et al., 2020), or to a watershed and beyond (Zhang et al., 2022). All of these have drawbacks, limitations, and uncertainties. Yet modeling is the backbone of MRV methodologies. As with verification, distinctions are needed for what models are meant for scientific use and what models serve to constrain carbon sequestration estimates (i.e., net CO₂ removal after accounting for carbonate precipitation (partial or complete), CO₂ off-gassing, loss of alkaline earth metals to sorption and plant uptake, etc.) and be connected to carbon markets.

Some studies have assumed that solid carbonates are the products of ERW (Manning et al., 2013; Haque et al., 2020), and some have assumed that soluble bicarbonates are the products (Amann and Hartmann, 2022; Buckingham et al., 2022). Either can be, and at any point in time, they depend on the soil, climate, soil saturation, and other factors (Khalidy et al., 2022). The carbonate assumption is conservative (and geologically long-term), while the bicarbonate assumption is maximal (and geologically short-term). It, thus, remains an MRV uncertainty as to what alkaline earth metal-to-CO₂ ratio to use in the carbon accounting of ERW. An equitable solution may be to use a constant global value of such ratios (within reason as to locations/conditions that make sense for ERW) as it is not at the farmer’s control if the ratio in one land is lower than in another, yet farmers incur similar costs for silicate spreading.

The availability of abundant minerals regionally for any given point of application is intricately tied to the life cycle carbon footprint of an ERW implementation (Eufrasio et al., 2022). Mineral reserves of suitable silicates for ERW are not well-recorded as geological surveys normally focus on ores of commercial value for other applications. As such, a renewed consideration of minerals stocks is needed to inform LCA models of ERW deployment in different regions of the world. Some efforts have been made for mine tailings (Bullock et al., 2021; Bullock et al., 2022).

Government agencies and industrial organizations (e.g., grower associations) have been slow to notice and take up interest in ERW. Examples of this include the absence of ERW mention in agricultural funding calls, the usage of the term “soil carbon sequestration” to signify only those approaches that lead to organic carbon accumulation (Bai and Cotrufo, 2022), and the lack of ERW research occurring within such agencies and organizations or with involvement of their members (from the bibliometric data on Figure 1A, the few exceptions include Haque et al. (2019), Gomez-Casanovas et al. (2021)). This shows that academic groups should perform more of what is known as “knowledge translation and transfer,” “getting research into practice,” and “extension research” (Grip workshop, 2023).

Conceptual reviews on ERW have largely told a story that ERW is best suited for deployment in warmer regions, at times termed the “Global South” (Institute for Carbon Removal Law and Policy, 2023), in view of higher weathering rates and more acidic soils (Taylor et al., 2016). This has led some studies to omit the deployment of ERW in the vast farmland regions of wealthier nations located in colder or dried climates, such as Canada and the Northern Plains of the US (Hicks et al., 2022). Considering mineral sources and financing resources, ERW should be pursued in most regions of the world as a means to accelerate the development of logistics, guidelines, regulations, and MRV methodologies, and to identify risks and limits before it is deployed at large scales in regions where impacts may be less likely to be studied.

Historically, certain misunderstandings and disagreements have either slowed the progress on ERW until additional studies could move the science forward, or created conditions that could discourage large-scale ERW adoption. Examples include the following:

(i) In the early days of EW research, it was posed that silicic acid accumulation in natural waters would hinder the weathering reaction progress based on the principles of reaction equilibrium (Köhler et al., 2011), but a debate with Schuiling et al. (2011) showed that silicon removal via the formation of solid phases and biological sinks is the more plausible mechanism, and one that remains well-grounded by experimental observations and is used in reactive ERW models.

A couple of ways to overcome roadblocks 3.3.1 and 3.3.2 have been: 1) to use faster-weathering minerals (such as wollastonite, slags, or concrete residues) (Renforth et al., 2009) and 2) to perform experiments in more amiable climates (e.g., Brazil and Malaysia) (Silva et al., 2021; Larkin et al., 2022). Such strategies have worked and should remain a key means of accelerating ERW understanding, deployment, and risk and impact analysis. However, shortcuts can also create traffic jams, themselves becoming roadblocks (e.g., exclusive focus on 2), so other pathways should be explored alongside.

Farmers have used rock dust and other mineral amendments for decades and, in some cases, have used the same rocks and minerals being tested for ERW. An evident shortcut to generating more robust multi-year data, previewing long-term impacts, and looking for evidence of past weathering is to visit these locations to collect soil and water samples for study. Few publications have touched on this strategy (Haque et al., 2020; Taylor et al., 2021), possibly due to the challenge of forensic MRV versus forward-looking MRV, so opportunities remain.

A closing thought on the complexities of MRV is whether there is the absolute need for such complexities and to what extent. It is inherently difficult to quantify the ERW rate and extent under field conditions, considering the workload involved in collecting representative soils, waters (and even gases), and the uncertainties involved in their analyses and data interpretation. Sticking to basics is a reminder that at a very fundamental level, ERW should work on most croplands, to an extent, as it consists of exposing silicate minerals to conditions in which they are likely to weather much faster than if left to natural geological processes. The following questions are raised:

• Should farmers be rewarded for participating in ERW based on mineral application (respecting agronomic limits that farmers should self-impose), or based on carbon drawdown? One of these is bound to be more equitable and the other more profitable.

It is important to answer these questions from the point of view of the global and multi-decadal potential that ERW has and not only what scientists and early startups need to confirm the science. Agriculture has been historically based on trust in science to a great extent, so it remains to be seen if ERW will find its place as a mainstream agricultural technology in addition to being a geoscience technology.