Can Foraging for Earthworms Significantly Reduce Global Famine in a Catastrophe?

Abstract

Earthworms are a resilient group of species thriving in varied habitats through feeding on decaying organic matter, and are therefore predicted to survive an abrupt sunlight reduction scenario, e.g., a nuclear winter. In this study, the feasibility and cost-effectiveness of foraging earthworms to reduce global famine in such a scenario with or without global catastrophic infrastructure loss was considered. Previously reported earthworm extraction methods (digging and sorting, vermifuge application, worm grunting, and electroshocking) were analysed, along with scalability, climate-related barriers to foraging, and pre-consumption processing requirements. Estimations of the global wild earthworm resource suggest it could provide three years of the protein needs of the current world human population, at a median cost of USD 353·kg⁻¹ dry carbohydrate equivalent or a mean cost of USD 1200 (90% confidence interval: 32–8500)·kg⁻¹ dry carbohydrate equivalent. At this price, foraging would cost a median of USD 185 to meet one person’s daily caloric requirement, or USD 32 if targeted to high-earthworm-biomass and low-labour-cost regions; both are more expensive than most existing resilient food solutions. While short-term targeted foraging could still be beneficial in select areas given its quick ramp-up, earthworms may bioaccumulate heavy metals, radioactive material, and other contaminants, presenting a significant health risk. Overall, earthworm foraging cannot be recommended as a scalable resilient food solution unless further research addresses uncertainties regarding cost-effectiveness and food safety.

1. Introduction

Earthworms feed on decaying plant matter in and on the soil; it is therefore reasonable to think they will continue to survive in the immediate aftermath of an abrupt sunlight reduction scenario (ASRS). The co-occurrence of an ASRS and global catastrophic infrastructure loss (GCIL) would result in extreme levels of food insecurity without prior preparation. This scenario is described fully by Denkenberger et al. [1]. In brief, a large-scale nuclear exchange could trigger a nuclear winter, blocking out sunlight, and multiple high-altitude electromagnetic pulses, destroying electricity grids over large areas and collapsing industry. Another combination scenario is the blocking of the sun in the future if our energy system is very dependent on the sun (relying on technologies such as solar and wind power). Alternative paths to this scenario exist, such as asteroid impact, comet impact, or supervolcanic eruption for blocking out sunlight; and extreme solar storms, coordinated cyber attack, or extreme pandemic causing people to be unwilling or unable to come to work in critical industries for industry collapse. These latter scenarios, those requiring two independent events, appear less likely than nuclear war [1]. However, a possible mechanism for the dependence of events could be that in times of famine, pandemics are more likely [2]. This is due to people’s impaired immune systems, but could also be caused by people eating more wild animals, exposing them to zoonotic pathogens [3].

Preventing global food insecurity requires planning resilient food solutions; defined by Pham et al. [4] as “foods, food production methods or interventions that would allow for significant food availability in the face of a global catastrophic food shock. These solutions should be well-suited for contributing to an adequate food supply for the greatest number of people even in the worst scenarios”. Numerous promising resilient food solutions have been proposed, such as ramping up seaweed production [5], crop relocation, extracting edible calories from killed leaves, growing mushrooms on dead trees, fishing [1], repurposing paper mills to produce lignocellulosic sugar [6], and building methane single-cell protein factories [7]. Resilient foods also increase the chance that trade is maintained, which would dramatically increase the number of people fed [8]. However, many of these are reliant on industrial processes and have significant ramp-up times. While wood gasification could possibly provide an energy source for tractors, it too is associated with a significant ramp-up time. One advantage of earthworm foraging would be its immediate accessibility, providing a potential short-term food source for humans while alternatives are developed.

Across the globe, earthworms have provided a source of nutrition for people in many societies [9,10,11,12] as well as a safe feed ingredient for livestock [13,14,15]. There are also reported instances of people earning a living from collecting and selling wild earthworms [16]. Earthworms can be collected using basic equipment and technology. This raises the question of whether recovering societies could source a significant fraction of their nutritional needs through foraging earthworms following an ASRS with or without GCIL. The objective of this study is to use pre-existing data to estimate the nutritional potential of wild earthworm populations, then review the reported methods for earthworm foraging and evaluate their utility in a global catastrophic scenario based on the predicted financial costs, food safety criteria, and the accessibility of necessary equipment. To the best of the authors’ knowledge, there are no previously published studies addressing this question available in English (to which the literature review was limited). The scopes of previous studies, as reflected in the literature cited in this article, are limited to the nutritional content of specific species of earthworms, proposed methods for farming and processing earthworms to produce animal feed, descriptions of the techniques available for collecting wild earthworms for ecological surveillance, or research into alternative resilient foods for use in an ASRS with or without GCIL. To fulfil this objective, it is necessary to address the following points:

• How many earthworms are there in accessible land on Earth?
• What is the potential nutritional value of the earthworm population?
• What methods of collecting earthworms are available, and how suitable might they be for the above-mentioned scenario?
• What would be the time and labour costs for producing nutrition from earthworms, and how do they compare to other resilient foods?
• How confident can we be in these calculations, and what limitations contribute to our uncertainty?

Academic sources were prioritised when writing this article, and it has been made clear where uncertainties were introduced by lower-quality sources. A mathematical model estimating the availability of earthworms and costs of foraging for them was produced using Guesstimate, a probabilistic modelling tool. This software performs Monte Carlo analysis to propagate uncertainty of parameters in calculations [17]. A range of sample parameters is generated each time the model is accessed and run, see Figure S2; therefore, the precise values in the model may differ slightly from the representative summary values stated below. They remain well within an order of magnitude and therefore do not change the general conclusions of this article.

2. Materials and Methods

2.1. Overview of Methods

The methodology for this study can be broken down into three main interlinked sections: work on biomass, nutrition, and safety information; screening foraging methods; and the Guesstimate model (Figure 1). Firstly, in order to know the total earthworm biomass, what nutrition it can provide, and whether large amounts are safe to eat, a literature review was required to gather published data, and basic calculations were necessary to answer the research questions (points 1 and 2 from the introduction). Secondly, we identified foraging methods by reading the literature, media reports, and web pages before screening them for suitability for large-scale foraging, narrowing down four identified methods to one (point 3 from the introduction). This screening involved logical considerations and Fermi estimates to gauge feasibility. Lastly, a Guesstimate model was generated to probabilistically analyse the cost-effectiveness of foraging, taking in findings from the previous two sections and additional estimates of labour and electroshocking efficiency (point 4 from the introduction). All of the results from these three sections, along with analysis of general practicality and safety considerations, fed into the final output of profiling the feasibility of earthworm foraging (point 5 from the introduction). Full calculation details are provided in the Supplementary Materials (Table S1) for all work except the Guesstimate model, which is publicly available online [18].

2.2. Estimating the Global Biomass of Wild Earthworms

Estimates of the total live biomass of the global wild earthworm population were calculated based on data extracted from the literature on earthworm ecology and land usage.

A recent study [19] suggests the abundance of earthworms varies significantly across different environments, with the majority of sites having a wet matter abundance between 1 and 150 g·m⁻². The predicted mean for the territories earthworms inhabit was 51 g·m⁻². Unless specified otherwise, all reported masses of earthworms are wet weight.

The total amount of land available for earthworm foraging can be estimated using data on land usage from the Food and Agriculture Organization of the United Nations [20] and the analysis available from Our World in Data [21]. Excluding uninhabitable or barren land; land covered by ice sheets, fresh water, sand or bare rock; and urban land, where infrastructure could prevent access to earthworms; there are 10.3 billion hectares of accessible land. This is composed of arable land, pasture, shrub covered land, and forests [21]. The feasibility of accessing land in each subtype and earthworm foraging will vary, as will the abundance of earthworms, but for the purposes of this article, it is assumed that all the land area is available for foraging with the mean abundance of earthworms stated previously. The product of these two numbers gives a total earthworm biomass of 5.3 billion tonnes.

2.3. Estimating the Nutritional Potential of the Global Wild Earthworm Population

Estimates of the nutritional potential of the global wild earthworm population were derived from the aforementioned calculated global earthworm biomass and data on the nutritional content of multiple earthworm species extracted during literature review.

Estimated total earthworm biomass can be combined with dry matter content and nutritional data to quantify the calories, protein, and fat available from worms. The dry matter content of earthworms varies with the environmental conditions [22] and species [23], with values of 15.2% [23], 15.7% [24], and 20% [23] present in the literature; each one represents a species of a different ecological category cf. [19,25,26]: Lumbricus terrestris (anecic) cf. [25,26], Eisenia fetida (epigeic) cf. [26,27], and Allolobophora chlorotica (endogeic) cf. [25,26], respectively. Since the categories signify earthworm living and feeding behaviour and thus aboveground or belowground presence [26], taking the distinction into account for evaluating and implementing certain foraging endeavours may be important. Nevertheless, such incorporation remains challenging: Many species belong to multiple categories [25], while proposed models for category abundances feature significant uncertainties [28] or limitations [29], with biotic interactions between categories [28,30] and the type of habitat [19,28,29], among others, contributing to the earthworm diversity. As a result, the estimate remains agnostic in terms of detailed distribution of earthworm dry matter content and nutritional value per each category, generalising the findings through the use of arithmetic mean. Similarly, habitat is generalised to earthworm-habitable land, as per above. Considering the three dry matter content values, a mean of 17% suggests there are 900 million tonnes of earthworm dry matter available globally.

Given the estimate for total earthworm dry matter of 900 million tonnes, macronutrient data can be calculated by multiplying this figure separately by the relative amounts of protein, energy, and fat reported in the literature. The nutritional value of earthworms is described for earthworm meal with dry matter crude protein content in the range of 53.8% to 72.9% [11,12,24,31] depending on the species and the method of preparing earthworms for consumption. The energy content of earthworms is less widely reported, but a mean estimate of 4.068 kcal·g⁻¹ of dry matter can be derived from Sun et al.’s value for partially dried E. fetida worm meal [24] and Kavle et al.’s [11] value for Eisenia andrei. Lastly, the epigeic earthworms have a mean dry matter fat content of 13.7% as derived from the literature [11,24].

2.4. Methods for Extracting Living Earthworms from the Soil and Efficiency Thereof

2.4.1. Literature Review and Initial Checks

A literature review was conducted to identify different collection methods, considering techniques used in scientific studies, by commercial foragers, and by recreational foragers such as fisherfolk. Numerous foraging methods have been reviewed comprehensively by Rhea-Fournier and González [32], but a select few are discussed below after consideration of their suitability for foraging edible earthworms following an ASRS with or without GCIL. In general terms, extraction techniques rely on disruptive excavation of soil with physical separation of earthworms from the substrate, or induction of surfacing behaviour in earthworms so they can be collected from the ground. Each method was investigated until it was eliminated as a viable option or until a full Guesstimate cost model was generated (i.e., for electroshocking [18]). The predicted time and costs incurred for different methods have been expressed in terms of extracting the global earthworm resource estimated above but, more practically, as a cost to provide the nutritional requirements for each person. Three quantitative factors must be known for an estimate of the usefulness of each method:

• The percentage of the earthworm population in the soil that can be harvested (extraction percentage);
• The area of land foraged per application of a method (effective area);
• The time taken per application of a method (cycle time).

Focusing on land areas with higher than average earthworm biomass would increase the efficiency of foraging. Therefore, where selective collection methods may be applied, the availability of earthworm biomass is adjusted upwards.

2.4.2. Guesstimate Model of Electroshocking

As far as the authors are aware, no studies have been conducted with the specific purpose of maximising the biomass of earthworms harvested using electroshocking. Therefore, estimates of the potential yield, labour, and energy costs of this technique were produced by extrapolating from a number of reports (

Table 1. Input parameter ranges used in a model describing sourcing food through earthworm foraging by electroshocking.

Parameter	90th Percentile Range	Reasoning
Extraction percentage	10–88%	Satchell [34] did not give an extraction percentage for the total earthworm population, but the data show electroshocking to yield 10% of the number of worms extracted using potassium permanganate. A patent filed by Thiellemann [35] for an electrode octet earthworm extractor reports 87.7% of earthworms were extracted with the equipment.
Effective area	0.22–2.6 m2	Weyers et al. [36] built an electroshocking device that covered an area of at least 0.22 m2. Satchell [34] refers to earthworms surfacing up to 3 feet from the electrode. A circle of this radius has an area of 2.6 m2.
Cycle time	20–40 min	Thiellemann [35] reports a cycle taking approximately 20 min. Satchell [34] gives figures for earthworms collected after 40 min.
Electrical supply voltage	30–480 V	Thiellemann [35] states a minimum voltage required of 30 V. Weyers et al. [36] applied an effective maximum of 480 V.
Electrical supply current	0.2–4 A	Rhea Fourier and González [32] state 0.2 A as the lower effective current for extraction. Satchell [34] maintained an upper limit current of 4 A.
Number of electrodes managed in series	3–10	Estimate based on unpublished reports of earthworm electroshocking in the media [37].
Hourly wage for labour	USD 1.8–13.8 per hour	The OECD.Stat data show the range of hourly minimum wages in member nations was USD 1.8 to 13.8 per hour in 2022 when expressed with purchasing power parity [38].
Length of a working day	6–12 h	Daily rest requirements of 12 h were assumed, as informed by International Labour Organization data [39,40].

). Using the parameter ranges stated below and estimates of earthworm abundance as described previously, we created a mathematical model for the costs of earthworm foraging using electroshocking. The model is publicly available through Guesstimate [18]. This tool applies random variation and Monte Carlo simulations to parameters in order to generate a predicted range of outcomes. The estimated range of values for earthworm biomass density in this model is based on Phillips et al. [19]. The upper limit of 150 g·m⁻² was derived from the authors’ own description of the data, “biomass typically ranged (97.3% of pixels) between 1 g and 150 g per m²”, with the median value of 6.16 g·m⁻² used as the lower limit. It was assumed that pilot sampling of land using digging and hand sorting/wash sieving could increase the productivity of foraging through selection of the 50% of sites with the highest earthworm biomass density. Examination of the raw dataset indicated that a log-normal distribution within this range gave the closest fit [33].

Whereas the input parameters in the model should give an estimate of global earthworm foraging cost-effectiveness, it is also valuable to have an estimate of more targeted foraging strategies. As such, the model was also run with earthworm biomass density set to one standard deviation higher than the mean, and labour costs set to one standard deviation lower than the mean. This is reproducible within the model by viewing scientific mode, which displays mean and standard deviation data, and updating the values for biomass density and labour costs. This provides a biomass density of 111.9 g·m⁻², and labour costs of USD 1.85 h⁻¹. We feel this is a sensible line of reasoning, given many of the areas with the highest reported biomass densities [19] fall within countries with minimum wages close to USD 1.85 h⁻¹: China (USD ~1.83–3.72 h⁻¹) [41], Russia (USD ~1.21 h⁻¹) [42], Kazakhstan (USD ~1.12 h⁻¹) [43], and Ukraine (USD ~1.08 h⁻¹) [44].

3. Results and Discussion

3.1. Global Nutritional Potential of Wild Earthworms

Through combining published data on earthworm distribution, land area, earthworm dry matter content, and earthworm nutrition, approximate values were obtained for the total global nutritional potential of wild earthworms. The global earthworm nutritional protein resource was estimated at 540 million tonnes, the total nutritional energy resource was estimated at 3.65 × 10¹⁵ kcal equivalent (910 million tonnes of dry carbohydrate), and the total dietary fat resource was estimated at 122 million tonnes.

When compared with global nutritional requirements, these estimates indicate that the total wild earthworm population could provide a significant supply of energy and macronutrients for humans (

Table 2. Estimations of the nutritional potential of the global earthworm resource in relation to the nutritional demand of the current world human population. * Value was calculated in this study; see Table S1 for further details.

Metrics for Calculating Global Requirements			Value
World human population			8.07 billion [48,49]
Daily protein requirement per person			51 g * [50,51]
Daily energy requirement per person			2100 kcal [52]
Daily fat requirement per person			315 kcal * [45,52]
	Daily global requirement	Potential earthworm resource	Potential days of nutrient from earthworms
Protein	4.15 × 1011 g *	5.44 × 1014 g *	1310 *
Energy	1.69 × 1013 kcal *	3.65 × 1015 kcal *	220 *
Fat	2.54 × 1012 kcal *	1.10 × 1015 kcal *	440 *

). The most abundant macronutrient, relative to demand, is protein, but the fat supplied could be of particular value following an ASRS with GCIL [45]. Based on World Health Organisation requirements for daily fat intake to comprise at least 15% of daily caloric intake [46,47], earthworms could represent a significant nutritional resource. Each person would require 500 g of fresh earthworms to meet their daily protein requirement harvested from an average area of 10 m². It is necessary to establish whether this resource can be accessed by people and at what cost.

3.2. Feasibility and Cost-Effectiveness of Extracting Earthworms from the Ground

3.2.1. Mechanical Sorting of Soil

The simplest methods of extracting earthworms from the soil are to locate and collect worms by manually or mechanically sieving excavated earth. Digging followed by hand sorting or washing-sieving [53] is the reference standard method for measuring the total population size of earthworms, with up to 100% of macroscopic earthworms collected. However, it is incredibly labour intensive [32], so mechanical tools have been developed to accelerate the process [54]. The depth of soil excavated varies from 0.2 to 0.5 m. Even if 100% of earthworms could be collected by removing the top 0.35 m of soil from any given area, the total volume of excavated soil to harvest the entire earthworm population would still be excessively high at 3.61 × 10¹³ m³. To meet the daily protein requirement of a single person would require sorting an average of 3.41 m³ (5120 kg) of soil. Lin et al. recently published results on a novel specialised mechanical worm separator capable of retrieving 83.8% of worms from 21.2 kg of substrate per minute [54]. At this rate, it would require 5 h to collect a person’s daily protein intake of earthworms without consideration of the time taken to excavate and transport the soil to the sorting machine. Field trials of excavation and soil sorting are necessary to improve the predictions of the precise costs of this method; however, the preliminary modelling suggests sorting of soil would not be a practical method of harvesting wild earthworms to feed a large population. Scaling up the method would require production of a large number, at least one per five people, of mechanical earth sorters and the removal of enormous quantities of soil. In a scenario with GCIL, it would not be feasible.

3.2.2. Worm Charming or Grunting

Worm charming, or grunting, is another physical technique for collecting earthworms. Rubbing a metal ‘rooping’ iron, such as an automobile leaf spring, across the top of the stake creates vibrations inducing worms to surface within a 12 m radius of the stake [16]. There are numerous reports of worm collecting competitions. Generally, participants agitate the ground and generate vibrations with music, garden tools, or probes that cause earthworms to come to the surface [55]. It has been suggested that earthworms surface to avoid moles, with the vibrations mimicking that of the predator [16], although this has only been reported as a causal relationship for one species of earthworm.

The literature on worm grunting is limited, but a few authors have quantified the number of worms surfaced from such techniques. Catania [16] describes how two professional worm grunters, Gary and Audrey Revell, earn a living in Florida, USA by selling earthworms for fishing bait. The distribution of surfaced worms charmed by the Revells is analysed, but time and extraction efficiency are not listed. Mitra et al. [55] conducted a similar experiment in Apalachicola National Forest, Florida. They did record time, but not extraction efficiency. While these studies show promising numbers of surfaced worms, the results cannot be extrapolated for global analysis without extraction efficiency data. At this time, there is insufficient evidence to support grunting as a viable option for scalable, cost-effective foraging. However, the equipment required is inexpensive, could be scavenged in large quantities from readily available sources, and requires no electricity or mechanised power. This technique may also allow large areas of land to be foraged quickly. If further research showed that untrained individuals were able to reach high levels of extraction efficiency, grunting could be revisited as a foraging technique.

3.2.3. Chemical Earthworm Expellents

There are a wide range of chemicals, known as vermifuges, that cause earthworms to surface when applied as a solution onto soil [32]. These include household detergent, formaldehyde, potassium permanganate, mustard solution, onion solution, and allyl isothiocyanate (AITC) [32,56]. Household detergents are reported to yield an order of magnitude fewer earthworms compared with the other solutions [56]. As a known carcinogen and hazardous solution [57], formaldehyde is not suitable for treating food. Potassium permanganate is recognised as safe for the treatment of food. However, it is reported that worms collected from soil treated with potassium permanganate solution disintegrate unless fixed in formalin [58], again limiting its applications for foraging. Mustard and onion solutions are both non-toxic, but using them requires the diversion of agricultural products away from direct food consumption. While it is reported that the application of onion solution to soil yields more earthworms than application of formalin or AITC [59], the mass of onion required is unlikely to be sustainable. Steffan et al. reported optimal extraction efficiency using 700 g of fresh onions in 4 L of aqueous solution to treat an area of 0.196 m², yielding 16 g·m⁻² of earthworms [60] equating to the cost of approximately 223 g of onion per gram of fresh earthworms (Table S1). Even if the efficiency of this process were increased tenfold, it is unlikely to be appropriate following an ASRS with GCIL. AITC is a synthetic version of the active compound in mustard solution, with production dependent on industrial chemical processing, making it unsuitable for the scenario described above.

Another practical consideration against the use of vermifuges is the large volume of solutions that would be required to treat soil. Taking a rate for vermifuge application of 20 L·m⁻², a conservative lower estimate based on a range of chemical extraction studies [59,60,61,62,63], and a 50% extraction efficiency suggests 400 L of solution would be required to harvest the daily nutritional protein for each person, a potentially prohibitive mass of fluid to transport and apply repeatedly.

Chemical expellents do not appear to be a promising solution for the large-scale foraging of earthworms.

3.2.4. Electroshocking

In the academic literature, there is a long history of a technique known as electroshocking: Applying an electrical current to soil for the purpose of collecting earthworms. Academic overviews are provided by Rhea-Fourier and González [32] and Singh et al. [56]. The basic technique requires insertion of electrodes approximately 45 cm into the soil followed by connection to a power supply so that current flows through the surrounding soil. The current is applied for a set period of time, during which earthworms escape to the surface in a radius around the electrodes. Here, they can be collected alive from the surface once the electrode is switched off or while wearing electrically insulating gloves. A unipolar approach is simplest, with current applied through a single electrode directly to the soil [34]. The use of both positive and negative electrodes to create a defined electric field through the soil is also possible. An electrical octet, arrangement of four pairs of electrodes applying voltage across the soil in multiple directions, is reported to give the highest earthworm extraction efficiency [32,35,36,64]. Alternating current (AC) is preferred over direct current (DC), necessitating either a generator for the electricity power supply or a DC to AC inverter if batteries are used [36].

A probable benefit of electroshocking as a scalable foraging technique is the very low requirement for operator skill, though safe product design is crucial to minimise risks of electric shocks [65]. It can also be performed without physical or chemical disruption of the ecosystem [32], and so may be compatible with other land usage such as growing crops.

The results of modelling electroshocking cost-effectiveness suggest that large-scale foraging is prohibitively expensive, but there may be a small subset of locations where it could be practical. The Guesstimate model predicted mean labour costs (90% CI) of USD 0.31 (0.0081–2.1) kcal⁻¹ of nutritional energy, USD 2.3 (0.053–14.0) g⁻¹ of protein, and USD 1.3 (0.029–8.6) kcal⁻¹ of fat. This indicates a mean dry carbohydrate equivalent labour cost of USD 1200 (32–8500) kg⁻¹, with 17 h of foraging required to meet one’s daily protein requirement, or a median of USD 353 kg⁻¹, with 6 h of foraging needed to do so. Although this rules out electroshock-foraging as a feasible supply of protein across the globe, limiting foraging to areas with low labour costs and high earthworm biomass density may be able to provide some nutrition to such regions, e.g., parts of China, Russia, Kazakhstan, and Ukraine. This strategy of targeted foraging would reduce labour costs to USD 60 kg⁻¹, equivalent to meeting a person’s daily caloric needs for USD 31.50. Compared with other resilient foods (

Table 3. Comparison of earthworm foraging against other resilient foods. Affordability: Retail cost equivalent to the amount of each food required to fulfil the caloric requirement for an average person (2100). Adapted from Martínez et al. [45] (lignocellulosic sugar, synthetic fat, methane single cell protein, seaweed, greenhouses, and acetic acid), and Denkenberger et al. [66] (farmed mealworms, white button mushrooms, artificial light algae, and artificial light vegetables).

Resilient Food	Affordability (USD/Person/Day)
Lignocellulosic sugar (repurposed paper factory) [6]	0.50
Methane single cell protein [7]	1.60
Lignocellulosic sugar (new construction) [6]	1.70
Synthetic fat from petroleum [45]	1.70
Seaweed in tropical areas (Southeast Asia prices) [67]	2.20
Low-tech greenhouses in tropical areas (crops) [68]	3.00
Acetic acid from microbial electrosynthesis [69]	5.40
Farmed mealworms [66]	9.45
Earthworm foraging (targeted)	31.50
White button mushrooms [66]	35.18
Artificial light algae [66]	57.75
Earthworm foraging (median)	185.33
Artificial light vegetables [66]	315.00

), this is still 1–2 orders of magnitude more expensive than the cheapest options. However, the least expensive resilient foods require ramp-up times and significant industrial processes, whereas earthworm foraging could be performed as quickly as basic equipment could be gathered. Earthworms are also more rich in terms of micronutrients and macronutrients compared with some industrial resilient foods, such as lignocellulosic sugar and synthetic fat. Provided safety concerns are addressed, this suggests foraging may have some utility in reducing famine in a small subset of targeted locations, but in most locations, alternative resilient food solutions are likely to be more successful.

Real-world examples of earthworm foraging suggest these targeted cost estimates are somewhat close to reality, but not perfect. In Vietnam, electroshock-foraging can reportedly yield a daily average of 10 kg per person [70], though there are also claims of capturing over 100 kg in a single night [71]. These are sold fresh for VND 35,000–80,000 kg⁻¹ according to various media sources [37,70,71], equivalent in dry weight to USD 8–19 kg⁻¹. Similarly, news reports from China suggest electroshock-foraged earthworms are sold for RMB 20 kg⁻¹, equivalent in dry weight to USD 16 kg⁻¹. These figures are lower than the USD 60 kg⁻¹ presented from the Guesstimate model for the targeted case, but roughly within the same order of magnitude. This does suggest that some of the data used in the model may be pessimistic. Of note, the similar pricing between Vietnam and China is in conflict with the global earthworm biomass dataset, which suggests high biomass density values in parts of China, but low values in Vietnam. Targeted foraging could therefore potentially be applicable to a larger range of countries such as Vietnam, though additional information would be needed to expand this list.

Brief consideration can be paid to the electrical energy needs for scalable electroshocking of earthworms. The Guesstimate model predicted a mean use of 2.3 MJ of electricity per g of dietary protein, equating to approximately 340% of global electricity production in 2022, to meet the global annual requirement for protein from earthworm foraging. The median prediction was 0.4 MJ of electricity per g of dietary protein, or approximately 59% of global electricity production in 2022. For comparison, production of methane single cell protein, a resilient food, requires 0.11–0.15 MJ of energy per g of dietary protein [7]. However, the majority of this energy is from natural gas, with a smaller fraction from electricity. As electricity supplies may have to be prioritised for the most productive activities following an ASRS with GCIL, and there may be additional difficulties bringing large amounts of electricity into the field, this method does not appear viable for combination catastrophes. However, given the median electricity requirement is comparable to that of methane single cell protein, electricity does not appear to be the bottleneck for targeted electroshock foraging in an ASRS without GCIL.

3.3. Climate-Related Barriers to Earthworm Foraging

Wild earthworms represent part of a large natural ecosystem that is liable to change substantially following an ASRS with likely ramifications for the feasibility of foraging for them as food. The most pertinent are changes in soil temperature and precipitation levels.

3.3.1. Reduced Temperature May Prevent Earthworm Foraging

Following an ASRS, global surface temperatures are predicted to fall substantially. The degree of global cooling caused by nuclear conflict is hotly debated, with predictions of up to an 8 °C [72] reduction. This uncertainty precludes precise predictions of how earthworm foraging will be affected, yet there are some general considerations.

As ectotherms, the biological activity of earthworms is affected by the environmental temperature [73]. Singh et al. [74] provide a thorough review of the effects climate change may have on earthworm populations, with many studies observing a reduction in reproductive rate and physical activity of earthworms at cooler environmental temperatures [75,76,77]. Moreover, earthworm adaptive behaviours include burrowing deeper into soil as air temperature falls [78]. These findings suggest a reduced efficiency of foraging following global cooling.

Overlaying 150 Tg nuclear winter temperature predictions [72] onto earthworm biomass density distributions [33] suggests the long-term availability of worms may be considerably lower in a severe ASRS (Figure 2). Many of the areas with the highest earthworm biomass density, including Canada and Central and East Asia, would suffer freezing conditions which could substantially reduce the accessible wild earthworm resource. The ground could become too hard to break through, preventing humans from inserting electroshock probes into the ground or digging for worms. Additionally, earthworms may not be able to penetrate the topsoil to surface. Moreover, the previously described risks of reduced earthworm activity would extend to this area and beyond, including land that reaches cold but not freezing temperatures (Figure S1). These effects add a time-sensitive aspect to earthworm foraging, though collecting worms in the early stages of a catastrophe before temperatures drop significantly may minimise the difficulties of freezing zones.

3.3.2. Reduced Precipitation May Affect Earthworm Abundance

Climate modelling of an ASRS predicts a reduction in precipitation [72] alongside lower temperatures. Earthworm activity and reproduction are affected by the moisture level in soil, a factor that will be influenced by changes in precipitation and temperature [74]. This factor warrants consideration in future work modelling the availability of earthworms as a food source in this scenario, though it is less important for the immediate harvest available.

3.4. Processing Earthworms for Consumption

Once harvested, it is recommended to remove the soil content from the gastrointestinal tract of earthworms before consumption. Soaking earthworms in water for 3 h [14] or manually expelling the gut content [12] are among suggested solutions. Several methods of earthworm slaughter are described, including immersion in acetone [14], lyophilisation (freeze-drying), osmotic shocking (salting or brining) and blanching [15] in 60 °C to boiling water, and eating them fresh [12,79]. Blanching in water is technologically the simplest processing method, maintains the highest nutritional value of earthworms, and reduces taint from coelomic fluid lysenin protein [15]. Through cold water soaking followed by blanching, earthworms can be prepared for eating without any significant technological, labour-intensive, or costly processes. For microbiological safety of human consumers, creation of sterilised defatted earthworm meal is recommended; this is a more technologically involved process and may reduce the nutritional value of the end-product, see Tedesco et al. [80] for full details.

3.5. Significant Limitations and Uncertainties

In reviewing the material available for this study, it was apparent that there are several areas of uncertainty which could affect predictions of how suitable earthworm foraging would be as a food source in an ASRS with GCIL. Specific examples are highlighted below with reference to whether they suggest greater or lesser feasibility of humans using wild earthworms as a food source.

3.5.1. Dataset for Global Earthworm Abundance Is Limited and May Underestimate the Resource

The estimates of global earthworm abundance in this article are based on a published study by Phillips et al. [19] and the corrected dataset associated with it. However, the model described in the paper has been criticised for excluding earthworm survey data collected prior to the year 2000 [81,82]. Moreover, it has also been criticised for extrapolating from geographical point data, which may have led to underestimation of earthworm abundance [82]. Earthworm populations in the tropics are underrepresented in the dataset, and its authors admit surprise that their model predicts such a low biomass density in tropical areas. An illustration of the academic uncertainty is a report by Blakemore, which concluded that the global earthworm dry biomass amounts to 4.5 gigatonnes [83], fivefold higher than the estimate calculated above. Blakemore’s figure is based on a series of earthworm surveys predating 1985 [84] and assumes the much larger area of earthworm habitable land of 2.4 × 10¹⁴ m², which may explain the discrepancy in the findings. It is possible that future work will recognise a much higher biomass of earthworms in the tropics and possibly a higher average biomass globally, both of which could increase the size and accessibility of wild earthworms as a food source.

An additional area of uncertainty is the exact nutritional value of the global earthworm population due to the heterogeneity of species, differences in reported methods for measuring the nutritional value of earthworm samples, and inconsistencies in the reported metrics of nutritional content. An exhaustive literature review was outside the scope of this article, but, in the authors’ opinion, differences in the nutritional content would not be large enough to alter the conclusions of this study.

3.5.2. Worm Grunting, or Charming, Is Promising, but Reliable Data Are Lacking

The academic literature on quantifying worm grunting and charming is scarce, but these techniques have quite a significant presence in public media. Unreviewed reports indicate that potential earthworm yields may be very high with relatively low labour and equipment costs. The Guinness World Record [85] for worm charming is 567 earthworms collected in 30 min from 3 m² of land, and worm grunters reportedly can collect thousands of worms in a few hours [55]. With scientific studies validating these findings and providing information on the extraction percentage for the techniques, it would be possible to predict, and possibly recommend, the use of such promising methods for foraging for earthworms in a disaster scenario.

3.5.3. Reports of High Yields from Electroshocking Are Unconfirmed by Scientific Literature

Recent media attention suggests earthworm electroshocking equipment is cheap and readily available in China and Vietnam. Furthermore, many people claim to collect over 100 kg of earthworms per day for slaughter and sale [37,71,86]. This yield is orders of magnitude greater than that predicted by models based on the scientific literature. Reliable descriptions of the equipment and context that permits such large quantities of worm biomass to be foraged could improve predictions of the scalability of this method. If such a technique were to be recommended, it would be necessary to establish quantitative estimates of the availability of resources such as AC generators.

3.5.4. Foraged Earthworms May Be Harmful for Humans

Earthworms cannot be recommended as a resilient food, intended to supply a significant proportion of a person’s macronutrient requirements, unless proven to be safe when consumed as a substantial fraction of the human diet. Annual consumption rates of up to 2 kg of earthworms per person per year are reported in the literature [12]. However, if foraged earthworms were adopted as a scalable resilient food source, annual consumption per capita could be much higher, increasing the risk of consumers being exposed to any foodborne hazards. Food hazards are commonly categorised as chemical (e.g., heavy metals or pesticides), biological (e.g., microbial pathogens), or physical (e.g., extraneous material) [87]. The risk of food derived from foraged earthworms presenting a risk to human health due to these hazard groups is addressed in turn below.

Earthworms bioaccumulate toxic substances from the soil substrate in their body tissues [88,89]. This process is pertinent to food safety because human industrial activities have released a number of toxic chemicals into the soil which pose a risk to human health should they enter the food chain [90]. Removing chemical hazards, such as heavy metals, from food requires expensive and technologically advanced techniques [91], preventing a simple mitigation strategy in an ASRS with GCIL. The level of chemical residues in earthworms is dependent on the level of chemicals in the soil [88,89] or substrate, suggesting the risk of this hazard to human health will vary across different environments. It has been demonstrated that wild earthworms can contain measurable quantities of heavy metals [88], including several which have clear maximum tolerable concentration limits [92], e.g., mercury and cadmium. It has been recognised that even farmed earthworms can accumulate toxic substances, such as cadmium, above safe levels for animal feed [13]. The authors are not aware of any studies of the human health impacts of consuming a diet rich in foraged earthworms. However, in the authors’ opinion, there is reasonable evidence that such a diet could be harmful and so should not be recommended unless starvation is the alternative.

As reported, without specific processing (i.e., creating sterilised defatted earthworm meal), farmed earthworms do not meet certain microbiological contamination standards for safe food [80]. Moreover, we, the authors, are not aware of any studies explicitly demonstrating that wild earthworms can meet modern standards of freedom from bacterial, fungal, parasitic, prion, or viral disease-causing agents. Heat treatment is a common method for killing pathogens, and so the processing methods discussed previously should remove the majority of biological threats [87]. Similarly, as sterilisation, in general, is used for near complete inactivation of microorganisms [93], sterilising methods other than the aforementioned one might be effective. Further experimental evidence in these areas is desirable.

Modern animal husbandry and meat processing techniques include provisions to prevent physical hazards such as bone shards, hypodermic needles, and hardware from damaged machinery entering food or feed [87]. Small-scale facilities processing foraged earthworms for human consumption are reported in the media. Based on previous work on the processing of farmed earthworms into animal feed [14,15], it should be achievable to engineer a method for converting foraged earthworms into food that is free from physical contaminants.

A large-scale nuclear war is the most likely single event to cause an ASRS with or without GCIL [1] and could also have widespread ecological effects that make earthworm foraging unsuitable. Nuclear detonations can disperse radioactive material [87], known as fallout, across large tracts of land as well as starting firestorms that may induce toxic rain to fall in the surrounding area [94]. Foraging for earthworms in a contaminated environment may be unsafe for workers due to the risk of direct radiation exposure. There may also be a risk for consumers of earthworms as, while earthworms are relatively resistant to the lethal effects of radiation, Krivolutzkii et al. reported the presence of only one species of earthworm in heavily contaminated forest plots for two years following the Chernobyl nuclear disaster in 1986 [95], suggesting heavy contamination with radioactive materials may reduce earthworm biodiversity. Rybak et al. supported the hypothesis that earthworm populations are affected by environmental contamination with radioactive material [96]. Moreover, earthworms can take up radioactive material from their environment [97,98,99]. Fujiwara et al. reported that E. fetida earthworms metabolise radioactive caesium compounds rapidly, as their internal concentrations of radionuclides fall soon after removing the specimen from radioactive substrates [100]. Thus, a purification stage of processing earthworms from contaminated soils could reduce the risk of harm from ingesting radioactive caesium in foraged earthworms. Previous studies have concluded that the public health risks from radiation following nuclear war are expected to be dominated by external direct doses rather than the one ingested from irradiated food [101]. Clearly, this is a field with considerable uncertainty; therefore, further evidence on the hazard of radioactivity in earthworms for food would be informative.

Ingestion of bioaccumulated chemical hazards is expected to present the greatest public health risk from foraging earthworms for food. Further research is required to address concerns of biological, physical, and radioactive hazards, but these appear to be more negligible issues.

3.5.5. Earthworm Foraging Could Reduce the Future Ecological and Agricultural Value of Land

The activities of earthworms recycle nutrients through a food web and increase soil fertility. They are recognised to play a vital role in maintaining current agricultural production [102], and their relation to soil quality differs across the ecological categories [26]. An ASRS would cause severe disruption to all ecosystems on Earth, but society would likely return to agricultural practices at some point in the future. A sustainable food solution for people should avoid impeding future production, and the possible effect of removing earthworms from large areas of land may warrant consideration. The sustainability needs might be met by earthworm farming, as nutritionally promising species such as E. andrei and E. fetida are extensively studied and utilised in vermicomposting and vermiculture settings [103]. The cocoon lifestage of some earthworm species is reportedly freeze-tolerant [78,104], suggesting farmed populations may be scaled up rapidly following an ASRS, using frozen stockpiles of viable organisms. Researching a food-production-optimised process for earthworm population growth (including species selection) and evaluating the feasibility and cost-effectiveness of such a process is out of this study’s scope, but it is worth mentioning that nutritional values in adults of the same species might differ between wild and cultured ones cf. [24,105].

3.5.6. Earthworms May Suffer during Capture, Processing, and Slaughter

There are considerable uncertainties regarding the capacity of whether invertebrates can suffer, and so whether it is beneficial to alleviate pain at the time of killing [102]. It is apparent that feeding a substantial fraction of the human population with earthworms would entail the collection and slaughter of an enormous number of earthworms; thus, even a small degree of suffering per earthworm could be deemed significant. Some authors suggest there is evidence of nociceptors and endogenous nociception-regulating chemicals in earthworms [106]. However, nociception may be necessary for pain, but it is not sufficient according to the definition of pain given by the International Association for the Study of Pain: “An unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage”. There is a dearth of evidence regarding earthworm sentience [107], and a full review of the literature is outside the remit of this paper, so it will suffice to conclude that future works may want to consider the net effect on the welfare of all species if earthworms were to be harvested for food.

4. Conclusions

Based on the current scientific evidence, the estimated global population of earthworms could be a significant source of nutrition, especially protein, for the current world human population in an abrupt sunlight reduction scenario. However, uncertainties around cost-effectiveness and food safety remain a barrier. Many techniques for harvesting wild earthworms are available, and some have been described at length in the scientific literature, while others are less well characterised. Extrapolating data from academic studies on earthworm populations, mechanical soil sorting, chemical expellents, and electroshocking do not appear to be cost-effective methods for foraging enough earthworms to meet dietary needs at scale. Given its quick ramp-up, targeted foraging using electroshocking may be feasible in select regions with high earthworm biomass density and low labour costs, though food safety concerns warrant caution. The methods used to forage earthworms for commercial purposes, worm grunting, and electroshocking in East Asia may be more cost-effective than implied by the results shown here, but further research is required to describe the efficiency of these techniques. If new data on earthworm biomass or foraging methods suggested much greater cost-effectiveness, an in-depth analysis of species-specific considerations would be warranted. Beyond the specific methods for earthworm foraging, the environmental changes associated with an ASRS with GCIL, global cooling, and radioactive pollution may reduce their effectiveness as a food source. As bioaccumulators, wild earthworms may contain heavy metals and other contaminants beyond safe levels for human consumption. These uncertainties should be considered in greater detail before global foraging of earthworms can be recommended as a resilient food source.