by Brian Tomasik
First published: 24 Sep. 2017; last update: 13 Dec. 2017


Food and other organic wastes thrown out in the trash often end up in landfills, where they feed various kinds of organisms. This page summarizes some information about insects created by garbage. Because some decomposition of organic waste in landfills is anaerobic, landfills probably give rise to fewer total invertebrates (and thus to less invertebrate suffering) than non-high-temperature aerobic composting of the organic waste would. Landfills also prevent some invertebrates from being born by temporarily covering land and preventing plant growth on it, but a rough calculation suggests that this consideration is much less important than the invertebrates created by the organic waste in landfills.

Invertebrates in trash cans

Tossing raw food into a trash bin tends to breed fruit flies and other invertebrates in your garbage. When you throw new materials into the trash bin, you may injure some of these bugs. Those bugs that you don't injure may be crushed in a garbage truck or at the landfill.

Goulson et al. (1999) examined flies emerging from waste that had been collected from neighborhoods at either one-week intervals or two-week intervals (p. 495). The waste was collected from a transfer station in England and then put into garbage cans to watch for the emergence of flies (p. 495). "Adult fly emergence was monitored using an emergence trap attached to the lid of each container[...]. Fly emergence was monitored daily for 28 days" (p. 495). The results were as follows (p. 496):

In total 243 kg of waste was collected, 114 kg of one-week-old [i.e., collected at one-week intervals] and 129 kg of two-week-old. This equates to 0.43 flies emerging per kilo from one-week-old waste and 0.08 from two-week-old waste (mean 0.26).

My housemate throws out food bags or containers whose sides contain remnant bits of food, and this gives rise to fruit flies, especially in warmer months. I would estimate that each medium-sized garbage bag contains at least 5-20 fruit flies, and sometimes as many as ~50. In contrast, I'm careful to wash food particles from bags and containers down the drain and then dry the bags and containers before throwing them out; this way, the bags and containers I throw out contain neither food nor water and so don't support fruit-fly growth. I also seal my garbage bags with a twist tie to keep bugs from getting in.

Sealing food scraps before throwing them out

It's best to prevent invertebrates from being born in the first place by sealing food scraps in airtight bags or plastic containers, so that invertebrates can't get in. This should also reduce infestations by ants, cockroaches, mice, etc.

Bug-free sealed food waste will instead decompose via fungi, bacteria, etc. This video shows the insides of sealed plastic containers of mine where food scraps are decomposing. There are no visible invertebrates, although I can't verify for sure that there are no microscopic invertebrates like rotifers and nematodes.

Once or twice over the past few years I've had fruit-fly larvae in a food-scrap container, but this was probably either because the food scraps I was disposing of had fruit-fly eggs already in them or because fruit flies snuck in when I opened the lid to put new food scraps inside.

When a food-scrap container did get a fruit-fly infestation, I put the whole container in the freezer for a long time (at least many months) to kill the fruit flies in a way that's thought to be relatively humane compared with alternatives. Leaving the fruit flies to multiply would have led to more total fruit-fly births and deaths.

If these plastic containers are enclosed in (possibly a few layers of) garbage bags, hopefully they'll remain sealed when they get to the landfill, though I don't know how often food containers are busted open during transport to and compaction at the landfill.

If your sealed food-scrap containers do get ripped open before or at a landfill, bugs can get in to the food waste. One idea for reducing this problem is to let the containers of food scraps sit in your house for a while (ideally for several years), so that more of the food energy will have been extracted before it gets exposed to the open air at a landfill. However, maybe this approach of letting the food decompose more before throwing it out would actually make the organic matter more digestible to bugs at the landfill? (Also, unless you live alone or with understanding housemates, keeping sealed containers of decaying food waste around your house might not be an option.)

Another possibility I've considered is digging a deep hole near my house, emptying the decayed food-scrap contents into it, and then covering it back up. The hope is that if the hole is deep enough, it might be anaerobic and inaccessible to earthworms and other bugs. I don't know enough about soil and soil animals to say whether this idea would work or whether aerobic animals would still find their way to the deeply buried pile of decaying organic matter.

Invertebrates in landfills

My impression is that most organic waste thrown out ends up in landfills.

US EPA (2015) says "When [municipal solid waste] MSW is first deposited in a landfill, it undergoes an aerobic (with oxygen) decomposition stage when little methane is generated. Then, typically within less than 1 year, anaerobic conditions are established and methane-producing bacteria begin to decompose the waste and generate methane."

Connecticut Department of Public Health (1997) says "The household and commercial wastes brought to landfills decompose over time largely through the action of bacteria." (That is, most decomposition is not done by invertebrate animals.)

Spelch (2016): "garbage in a landfill does decompose, albeit slowly plus in a sealed, oxygen-free environment. Because of the lack of oxygen, bacteria in the waste produce methane gas".

Diggelman and Ham (2003): "Food waste is anaerobically decomposed as it is processed through [... a] landfill" (p. 503).

Wikipedia ("Landfill gas utilization") reports: "MSW landfill gas typically has roughly 45 to 60% methane and 40 to 60% carbon dioxide, depending on the amount of air introduced to the site through active gas extraction." So it sounds like landfills engaged in landfill-gas extraction introduce small amounts of air, allowing for small amounts of aerobic decomposition?

What's the population of flies and other invertebrates in landfills? What fraction of waste at landfills decomposes via invertebrates and what fraction of decomposition is done by microbes? Landfills are sadly not bug-free, but I would guess they have (considerably?) lower total invertebrate populations than compost bins per unit of food decomposed.

Robinson (2005) says (p. 10):

Compacted or loose garbage at the landfill is usually covered to reduce odor and the attraction it has to various pests. Soil is commonly used for cover, and the thickness of the layer is important to fly control. [...] When soil is unavailble or the costs for it are high, other materials, such as paper pulp, fragmented plastic, sand, woven geotextiles, and plastic sheets may be used. In direct sunlight plastic sheets create in the underlying refuse a microclimate with temperatures high enough to prevent fly development. [...]

The house fly and local species of blow flies are the most common insects at urban landfills around the world. At landfills, these flies may breed continuously through the year, but with decreased numbers in the cold months. Crickets and cockroaches, including the German cockroach, can become established at landfills, depending on local conditions.

Enlighten Me (n.d.) adds:

At the landfill, garbage is unloaded onto the tipping face and compacted by bulldozers and other machinery. Each day’s garbage is referred to as a cell, and when a cell is filled, soil, foam spray, wood chips, and/or temporary blankets are used to cover the area. This covering keeps rain and wind from dispersing the garbage and controls insects, birds, and rodents.

This page says: "In [uncontrolled] landfills there is a permanent presence of both insects and rodents, which makes them a source of infectious diseases."

Goulson et al. (1999)

Goulson et al. (1999) used yellow sticky traps to measure fly numbers at several locations (p. 494):

  • "The active landfill cell of Paulsgrove landfill, near Portsmouth"
  • "50 m from this active cell"
  • "Paulsgrove compost site"
  • "household waste recycling centre, Paulsgrove"
  • "Port Solent (marina, restaurant and shop complex 500 m from the Paulsgrove landfill)"
  • "Gosport marina (a control site with no landfill nearby)"a
  • "Bushy Warren compost site, near Basingstoke".

The following figure (p. 496) shows the results. I added text labels to the curves that had the highest values to make reading the graph a bit easier.

Because these are measurements of flies caught by traps, we can't determine absolute numbers of flies per square meter of area, only relative numbers across locations.

Preventing plant growth?

While decaying organic matter in landfills can feed flies and other invertebrates, landfills also prevent some invertebrates from being born by precluding plant growth on their location for some time, until the landfill is covered over with grass or other vegetation. CalRecycle (2003): "When a landfill is closed it may look like a park, a golf course or a range of tree-covered hills. [...] Erosion of the surface of the closed landfill is prevented by planting trees, grasses, and other ground cover".

Landfills vs. incinerators

Landfills use more land than incinerators, but they also allow more invertebrates to eat the organic matter put into them than incinerators do. Here's a rough sketch of a Fermi calculation to compare these effects.

Suppose a wet mass of M kilograms of organic matter is thrown out.

If it's incinerated, assume that it feeds no invertebrates and that it covers no additional land area. (Both of these assumptions are slightly untrue but probably not that far from being correct.)

Now suppose the organic matter is landfilled. Let its density (kg per m3) be D, which means it occupies a volume of M/D. Suppose that materials in the landfill occupy an average height of H meters, with everything packed in tightly. Then the additional area required to landfill this organic matter is (M/D)/H m2. Suppose that this amount of land area is covered for Y years, after which the landfill is covered over with grass, which we can assume has the same net primary productivity as whatever native vegetation the landfill is replacing. Let NPP be the net primary productivity (in kg wet mass per m2 per year) that would have occurred on the land if it weren't covered by the landfill. Then throwing out the organic matter prevents NPP * Y * M/(D * H) kg of wet matter from being created. Suppose that a fraction f of that wet matter would have fed invertebrates. So covering over the land prevents f * NPP * Y * M/(D * H) of invertebrate food.

Meanwhile, the landfilled organic matter creates some invertebrates who directly eat that food, especially when it's exposed to air for flies and other bugs to access it. Let fw be the fraction of the original M kg of waste organic matter that feeds bugs in this way.

Then the question is whether this quantity is bigger than the productivity prevented by the extra landfill space used, i.e., whether M * fw > f * NPP * Y * M/(D * H). Dividing out M on both sides gives the question of whether fw > f * NPP * Y / (D * H).

Absent empirical values for these parameters, we can make some simplifying assumptions. Suppose that f is, say, ~10 times fw, given that landfill decomposition tends to be more anaerobic and thus less amenable to animal life than ordinary decomposition of grass on a field. Also, let's assume that Y and H are both on order of ~100 to ~101 such that Y / H is roughly ~1.b And let's assume that D is comparable to the density of water, which is 1000 kg / m3. Then our inequality becomes 1 > 10 * NPP / 1000, i.e., NPP < 100 kg per m2 per year = 105 metric tons per km2 per year. Since terrestrial NPP values are typically around ~103 metric tons per km2 per year, it's very likely that NPP is in fact less than 105 metric tons per km2 per year. In other words, native-vegetation NPP would not have been high enough that preventing that primary productivity on the landfill area outweighs the increase in invertebrates due to decomposition of the organic matter in the landfill.

Intuitively, this is because a hypothetical landfill composed of only organic matter would have massive piles of decomposing waste, far more than a grass field could produce in a few years.c This conclusion is also consistent with the intuition that a landfill supports many more flies per unit area than grassland would (although maybe one could argue that landfills are particularly fly-dense, while grasslands are dense with non-fly invertebrates).

Thus, this tentative calculation seems to favor incineration over landfilling, ignoring lots of other complicating factors.

One further consideration is inspired by this fact from Diggelman and Ham (2003): "It was assumed that 95% of food waste solids are decomposable [...] of which 84% is decomposed in the landfill during the design life" (p. 509). The materials in food waste may escape in gaseous form as methane and CO2 or in liquid form as leachate. This makes me wonder if food waste, unlike other kinds of trash, doesn't really contribute much to landfill volume at all in the long run? If so, that would strengthen the conclusion of this section that the land-covering effects of landfilling food waste are relatively small.

Mechanical biological treatment

Some communities use mechanical biological treatment (MBT) prior to sending materials to the landfill. I don't know what fraction of municipal solid waste in developed countries undergoes MBT, but my guess is the fraction isn't large.

MBT treats organic material in mixed household waste via anaerobic digestion, composting, or biodrying (Wikipedia "Mechanical ..."). In cases where MBT is used, the invertebrate-suffering impacts of throwing food waste in the trash look more like those of anaerobic digestion, composting, etc. My guess is that anaerobic digestion involves almost no invertebrates. I think large-scale, high-temperature composting often involves some invertebrates but probably fewer than most home composting systems per unit of food decomposed. Refuse-derived fuel produced by biodrying has its stored energy released via combustion, which is a non-sentient (and thus non-suffering) process.

Examples of MBT facilities:

  • Menart Belgium (2016) describes an MBT process that uses windrow composting of organic materials.
  • Associated Engineers Ltd (2017) shows an MBT facility that uses anaerobic digestion.
  • Control Process S.A. (2017) describes an MBT facility in Poland constructed in 2015 that uses "horizontal fermentation" (see 0m44s in the video), a "dry anaerobic digestion process", to produce biogas (1m05s). There's also "a system of oxygen stabilization of biodegradable municipal waste separated from the stream of mixed municipal waste" (2m15s). Does "oxygen stabilization" just mean "composting"?? The facility serves 210,000 citizens (3m07s).
  • misskosdaq (2012) shows an MBT plant in South Korea. Organic wastes are fermented for two days in a rotary drum reactor at 60°C to 70°C (1m20s). Presumably this painfully kills any invertebrates already present in the garbage, but subsequent organic-waste decomposition is (I assume) free of invertebrates, at least until the compost exits the MBT facility. (You can see the generated compost at 2m58s.)


  1. While Goulson et al. (1999) don't say so explicitly, I assume this site is also not near composting facilities, given that it's a "control site". Goulson et al. (1999) do say (p. 497) that "Gosport and Port Solent[...] were not immediately adjacent to waste disposal facilities."  (back)
  2. Diggelman and Ham (2003) report regarding the Dane County Landfill in Wisconsin, USA that "The design life of the landfill is 15 years" (p. 509). Does this mean the landfill prevents plant growth for 15 years (i.e., that Y = 15)? Or do parts of it get covered with greenery before the whole landfill is retired?

    Following is a Google Maps overhead view of the Dane County Sanitary Landfill in Madison, Wisconsin. My guess is that most of the picture other than the southwest corner by the golf course is part of the landfill's total area? I can't tell very well from the picture, but it looks to my untrained eye like non-active parts of the landfill are becoming green again, while the active part in the middle where vegetation is precluded is only a fraction of the total? Since it looks like landfilled waste doesn't prevent plant growth for the full lifetime of the landfill(?), Y is probably somewhat less than 15.

    Diggelman and Ham (2003) also report (p. 509): "The landfill will receive an average of 293 metric tons per calendar day over its design life". So in total, over the lifetime of 15 years, the landfill will hold 293 * 365 * 15 ≈ 1,600,000 metric tons. A single 100-kg deposit of food scraps would constitute (100 kg) / (1.6 * 109 kg) = 6.3 * 10-8 of the total material that will ever be added to the landfill. (Diggelman and Ham (2003) report this as 6.2 * 10-8, perhaps because they have access to unrounded numbers for doing this calculation?)

    Diggelman and Ham (2003) say that every 100 kg (wet weight) of food waste sent to the landfill is responsible for 0.019 m2 of land use (p. 509). So the total area required for the landfill should be (0.019 m2) / (6.2 * 10-8) = 310,000 m2 = 31 ha.

    Diggelman and Ham (2003) explain (p. 509): "The landfill land area is 12 ha; the fenced area is 28 ha[.... ]There are two buildings required for the landfill with a total area of 1300 m2". We can ignore the two buildings because they cover so little land compared with the landfill itself. I infer that the "landfill land area" is contained within the "fenced area" because we need the total land area to add up to roughly 31 ha, as we saw in the previous paragraph, and if the "landfill land area" was separate from the "fenced area", the total would add up to 40 ha.

    I'm not sure if the area that holds municipal waste is just the 12 ha of "landfill land area" or if most of the 28 ha of "fenced area" will eventually hold municipal waste. To make the following calculation simple, let's assume that the total area that will eventually be covered by municipal waste is the average of those two numbers: 20 ha = 200,000 m2. Much of this landfill waste isn't food scraps, so the density of this material may not be extremely close to the density of water, but I'll assume the density isn't vastly different from that of water—viz. 1 kg / L = 1000 kg / m3. Then 1,600,000 metric tons of total waste is roughly 1,600,000 m3 of volume. Dividing by 200,000 m2 of area implies an average landfill height of H = 8 m. Since I said above that "Y is probably somewhat less than 15", it seems about right to assume that Y / H is not too far from ~1.  (back)

  3. Of course, actual landfills are mostly not made of organic matter, but the relevant question here is how much throwing out a given piece of organic matter increases the size of the landfill, so non-organic wastes should be ignored. Of course, throwing out non-organic wastes does increase the size of the landfill without feeding invertebrate populations.  (back)