by Brian Tomasik
First written: Feb.-Mar. 2016; last update: 22 Mar. 2017


This piece reviews ways in which cattle grazing affects the net primary productivity of pasture land. This is one piece of a larger puzzle regarding how grazing impacts wild-insect populations and hence wild-insect suffering. While it appears that grazing can sometimes increase the productivity of pasture plants, in practice the reverse is often true, especially when we consider desertification and loss of soil fertility, which are especially common with overgrazing.


Cattle near grants passMost insects that are born die before reaching maturity, often in painful ways. From the perspective of reducing insect suffering, it seems good, ceteris paribus, to reduce insect populations. Cattle grazing is one way in which humans affect the world's insect populations on a massive scale.

Total insect abundances may sometimes be limited by food availability. Thus, reducing the amount of plant food eaten by insects should, in some cases, reduce insect populations. In that case, it's important to know how cattle grazing influences the amount of food eaten by insects.

The growth rate of insect biomass can be expressed as

secondary insect productivity = (net primary productivity) * (fraction of primary production eaten by insects) * (digestibility of the primary production) * ECD.

Fraction of primary production eaten by insects

Cattle grazing probably reduces "fraction of primary production eaten by insects", since cattle eat a lot of vegetation that could otherwise feed insects, although the issue becomes more complicated when we consider that insects also eat cattle poop.

This textbook gives a comparison of a leniently grazed pasture vs. a heavily grazed pasture:

The end result was that 42% of the energy captured by ryegrass in the leniently grazed pasture entered the detrital food chain while only 13% was consumed by livestock. By contrast, only 26% of the energy captured by ryegrass in the severely grazed pasture entered the detrital food chain, while 25% entered the grazing food chain.

Insofar as insects are part of the detrital food chaina, it happily seems as though the heavy-grazing condition reduced "fraction of primary production eaten by insects".

Digestibility of the primary production

More mature grass tends to have more indigestible lignin and less protein and accessible energy per gram. Does grazing grass keep grass's fraction of digestible material higher, or is the trend toward rougher, less digestible biomass a generic feature of more mature grasses? If grazing did keep grasses more digestible, this would be bad from an insect-suffering perspective because a given mass of more digestible grass probably supports more insects than the same mass of indigestible grass does? My impression is that lignin is mostly decomposed by bacteria and fungi. Some insects appear to also process lignin, but I assume this is pretty rare?

Net primary productivity

In this piece, I focus on the first factor in the above equation: "net primary productivity". If cattle grazing tends to decrease this factor, then because grazing also plausibly decreases the fraction of primary production eaten by insects, grazing may reduce insect abundances (assuming that plant food is a growth-limiting factor for insect populations). On the other hand, if grazing increases net primary productivity, then the ultimate impact of grazing on insect abundances is less clear.

What if you care about bacteria?

The above equation is for the number of insects supported on a pasture field, on the assumption that most of the suffering in a pasture takes the form of small animals at the insect level of organization. However, one might apportion a large amount of ethical importance to the collection of all bacteria and other unicellular organisms that a pasture supports. If one cares about the suffering of bacteria, one should probably also care about the suffering of individual cells within larger animals (including insects and cattle). So if one thinks that the cell level of organization has the most total suffering, then rather than trying to minimize the total number of insects, one would probably want to minimize something like (gross or net) primary productivity, since primary productivity is what provides the energy for all living cells.

Here I'm ignoring differences between sizes and metabolic rates of different cells, but perhaps moral concern for a cell should scale somewhat linearly in its size and metabolic rate, in which case total primary productivity (= total energy for biological use) would indeed be roughly proportional to total suffering.

If one cares about both insect-level suffering and cell-level suffering (e.g., one cares both about the suffering of an insect and separately about the suffering of all the cells within the insect), then reducing insect numbers while keeping primary productivity constant would still be good, since at least one of the two kinds of suffering would be reduced. But it could be bad to reduce insect suffering while increasing primary productivity (which might happen if, for example, cattle stimulate plant growth but reduce the amount of plant food available to insects).

Does grazing stimulate plant growth?

This page quotes an article saying that cow grazing on grass stimulates new grass growth. Likewise, this page claims that "By grazing, cattle stimulate plant growth and increase annual forage yield."

One might think that cutting grass would inhibit its ability to grow. However, mowing grass actually tends to spur grass growth above ground:

When you mow the lawn, grass responds to the reduced surface area on its blades -- it must produce more growth to maximize photosynthesis processes. As a result, the grass concentrates its energy into blade and runner growth, depending on the species.

However, this isn't a free lunch: "cutting a lawn is stressful for the grass. [...] Cutting stimulates top growth at the expense of root growth." This page says: "Instead of sharing the energy resources between the roots and the foliage, the grass concentrates on blade development only." Excessive mowing can turn a lawn brown due to degradation of roots.

This page explains that

Resources are often stored in leaves and specialized storage organs such as tubers and roots, and studies have shown evidence that these resources are allocated for regrowth following herbivory (Trumble et al. 1993; Tiffin 2000; Erb et al. 2009). However, the importance of this mechanism to tolerance is not well studied and it is unknown how much it contributes to tolerance since stored reserves mostly consist of carbon resources, whereas tissue damage causes a loss of carbon, nitrogen and other nutrients (Tiffin 2000).

Mowing too short can reduce primary productivity: "Kentucky bluegrass and fine-leaved fescues should not be cut lower than 1½ to 2 inches. Shorter mowing reduces leaf surface (the plant’s food manufacturing factory) to such a degree that the plant may have to draw food from its root reserves to initiate new growth." This article adds: "When grass is severely cut back the growth of the roots and the plant comes almost to a complete stop until the leaves can recover. This places a huge amount of stress on the plant which is often visible in a yellow to brownish look to the lawn after mowing."

That said, Elaine Ingham argues that if plants are healthy, then mowing/grazing doesn't cause loss of the root system, since the roots aren't a metabolic drain on the plant.

Cutting the top of grass reduces apical dominance and encourages the plant to grow sideways rather than up:

mowing actually helps make your grass grow thicker because the tip of each blade of grass contains hormones that suppress horizontal growth. So when you cut the lawn, you are removing the tips and the hormones which then allows the grass to spread and grow outward faster.

But it's not clear what effect this would have on primary productivity overall.

This page claims:

Apparently, when a plant is defoliated [by cutting or grazing] – a tremendous shock to its system – it exudes a strong pulse of sugary plant sap into its root zone, thus stimulating activity in the symbiotic community of bacteria and fungi surrounding the roots upon which the nutrition of all plants depend. This gives rise to the phenomenon, which I have often observed, of how a field (or even lawn) which has been cut, recovers amazingly quickly, as evidenced by tender young leaf regrowth of two or three inches in just a few days.

Joel Salatin says:

New evidence even suggests that when the animal tugs at the plant to shear off the grass tillers, it excites the roots into renewed productive activity. Kind of like exercise builds new muscles.

This point is discussed more here.

This page mentions "compensatory photosynthesis and growth stimulation by bovine saliva". The page also says:

With increasing number of mowings per year, which may be best observed in the sometimes extremely often mown urban turf-grass areas, species richnessb of both plants and insects (e.g., planthoppers and true bugs) becomes extremely poor. Moderate cutting or grazing may promote grasshopper populations, possibly through tillering rejuvenation or through changes in the proportions of nutritious grasses. Because the quality of grass shoots as a food resource declines with age, the induction of tillers and side shoots by cutting make nutritious food available later in the season (e.g., for the populations of many grasshoppers and enhanced infestations of stem-boring Diptera).

This paper explains that some plants in some environments show compensatory growth in response to grazing, but others do not. And this page explains that

Studies have found branching after [apical meristem damage] AMD to undercompensate, fully compensate and overcompensate for the damage received (Marquis 1996, Haukioja and Koricheva 2000, Wise and Abrahamson 2008). [...] The wide occurrence of overcompensation after AMD has also brought up a controversial idea that there may be a mutualistic relationship between plants and their herbivores (Belsky 1986; Agrawal 2000; Edwards 2009).

This section explains that leaves often operate below maximal photosynthetic capacity, so that herbivory may stimulate increased photosynthesis. "whether the increase in photosynthetic rate is able to compensate for the damage is still not well studied (Trumble et al. 1993; Stowe et al. 2000)."

This page, citing this study, enumerates several additional explanations why grazing may stimulate plant growth. This source also lists some reasons:

low level herbivory can remove aging roots and leaves, allowing new growth of young roots and shoots. [...] The feces of herbioves enrich the soil, which increases the chances of successful seedling growth. [...] Herbivores also prune plants, which allows for more light passage and makes it easier for seeds to fall from a parent plant.

Steinfeld and Gerber (2010) say: "In properly managed grass-based systems, grazing and mowing contribute to increased ecosystem productivity and biodiversity."

Grazing and primary productivity

Sources suggesting grazing reduces primary productivity

This textbook suggests that "The fundamental ecological dilemma encountered in grazed systems is the inability to simultaneously optimize the interception and conversion of solar energy into primary production and the efficient harvest of primary production by herbivores (Parsons et al. 1983)." The underlying premise is that more grazing reduces primary productivity.

The productive tallgrass prairie of central North America declined in part due to "the confined grazing pattern of European cattle versus bison".

This page explains:

Overgrazing reduces plant leaf areas, which reduces interception of sunlight and plant growth. Plants become weakened and have reduced root length, and the pasture sod weakens. The reduced root length makes the plants more susceptible to death during dry weather.

This page says:

Over millions of acres of North America deserts, bison, elk, javelina, and pronghorn never roamed and never grazed the deserts or the patches of grassland within them. These deserts were and are grazed, but by small mammals like rabbits, mice, reptiles such as desert tortoise, and insects. Grasses that evolved being eaten by tortoises and rabbits are not likely to respond well to being eaten in intense, even if short termed, bouts of grazing by the artificially created cow, or Old World animals such as sheep, goats, or horses.

That said, the same article also qualifies that in "humid fertile, open areas [...] the grasses really do respond favorably to proper levels of grazing."

This study found that grazing "decreased root growth, especially in heavily grazed patches (~30% less than in fenced controls)."

Sources suggesting grazing may increase primary productivity

This book says "Grazing or defoliation can increase net primary production (NPP) in many grassland environments (McNaughton, 1985; Williamson et al., 1989; Holland et al., 1992)."

This page says:

In some cases, overgrazing has led to the creation of new ecosystems which are just as productive, if not more, and as environmentally stable as the original one. The Edwards Plateau of Texas is an example. The original vegetation of this area was suitable mainly for cattle and a few deer, turkey and other wildlife species. It has been transformed into one of the most productive livestock and wildlife areas in Texas. It is now suitable for cattle, sheep and goat rearing. Around 6% of Texas' cattle, 77% of its Angora goats and 73% of its sheep are raised on the Edwards Plateau. The area now has one of the highest white-tail deer populations in the U.S. and the highest Rio Grande wild turkey population. The Edwards Plateau is prized for its watershed and recreational values. It also has suitable habitats for various endangered species. These habitats probably did not exist in the beginning. [...]

Continual undergrazing or no plant defoliation at all can result in poor vegetation health and declining productivity. This is often seen in exclosure studies whereas the vegetation outside the exclosure with proper grazing is better than that within it. Pristine vegetation did not develop in the absences of defoliation either by wildlife or fire or both.

This study says "The elimination of perennial understory vegetation and cryptobiotic crusts is a nearly inevitable consequence of livestock grazing in deserts. This opens these systems to annual grass invasion, subsequent burning, and loss of a major carbon sink". Let's ignore the issue of fires for now; it's discussed more below. Biocrusts "are dominated by cyanobacteria, lichens, mosses, fungi and bacteria" (rather than insects), so it's plausible that removing biocrusts allows more insects to live in the area and hence increases insect suffering? If so, this would be an unfortunate consequence of cattle grazing. On the other hand, biocrusts may preserve soil fertility into the future, so removing them might have its upside:

“Biocrusts often are associated with increased soil nutrients and water retention, but their most important task is to stabilize soil surfaces against wind and water erosion,” Barger said.

This page says:

In some regions of the United States under continuous grazing, overgrazed pastures are predominated by short-grass species such as bluegrass and will be less than 2-3 inches tall in the grazed areas. In other parts of the world, overgrazed pasture is typically taller than sustainably grazed pasture, with grass heights typically over 1 meter and dominated by unpalatable species[. ...] in many circumstances overgrazed pastures have a greater sward cover than sustainably grazed pastures.

Does this mean overgrazed pasture can be more productive? Or does it just mean that unpalatable "weeds" take over, and because they're not eaten, they can grow tall -- which doesn't imply anything about higher primary productivity per se?

Grazing optimization hypothesis

This textbook explains:

Grazing has traditionally been viewed as having a negative impact on the subsequent rate of energy capture and primary production within grazed systems [...]. However, the grazing intensity necessary to induce a decrease in primary production is difficult to establish definitively. The "grazing optimization hypothesis" suggests that an optimal grazing intensity can potentially increase primary production over that of an ungrazed system [...]. A limited amount of evidence exists to support the grazing optimization hypothesis (Dyer and Bokhari 1976, McNaughton 1979, Hart and Balla 1982, Paige and Whitham 1987), but it does not appear to be a significant ecological process operating on a regular basis in grassland systems (Belsky 1986, Heitschmidt 1990). Illustrations of the grazing optimization hypothesis tend to exaggerate the potential increases in primary production resulting from an optimal level of grazing relative to the potential decreases which may occur in response to severe grazing, that is, the potential increase in production is shown to be equivalent to the potential decrease. [...]

It is important to recognize that much of the data collected in support of the grazing optimization hypothesis were derived from grazed systems where herbivore density and movement were not directly regulated by humans. In these systems, primary production and herbivore density fluctuate widely in a series of continuous feedback loops in response to climatic variation (Sinclair 1975, Walker et al. 1987). Conversely, herbivore density and movement are rigidly restricted in managed systems and precautions are taken to minimize deleterious consequences on animal production. Consequently, the grazing intensity in many, if not all, managed systems may frequently exceed the intensity required to consistently stimulate primary production as indicated by this hypothesis (Heitschmidt 1990). This difference likely explains why the hypothesis originated with researchers working in natural rather than managed systems and why the hypothesis receives limited support from natural resource managers.

One theoretical argument that I'm partly making up as to why cutting grass might increase net primary productivity is that as a plant gets bigger, it needs to do more respiration, but it can only do a limited amount of photosynthesis. Hence, the fraction of production devoted to respiration should increase in bigger plants, which means the fraction of energy stored (net primary production) should decrease. Of course, if the plant is cut too short, its leaf area index will be smaller, so less photosynthesis will occur. Thus, it seems like there should be an optimal point for net primary productivity between grass being too small (not much photosynthesis) and too big (not much extra photosynthesis but more respiration). In fact, a similar idea is presented in the context of ecological succession here: "Early seral stages are usually marked by rapid growth and biomass biomass accumulation - grasses, herbs and small shrubs. Gross Primary Productivity is low but Net Primary Productivity tends to be be a large proportion of GPP as with little biomass in the early seral stages respiration is low. As the community develops towards woodland and biomass increases so does productivity. But NPP as a percentage of GPP can fall as respiration rates increase with more biomass."

This study, summarized here, of grazing in the Serengeti that found an inverted-U-shaped curve for aboveground net primary productivity as a function of grazing intensity (Fig. 9, p. 270).c In fenced areas from which large grazers were excluded, annual net primary productivity averaged 357 g per m2 per year, while in regions with temporary fences to allow for grazing, annual net primary productivity averaged 664 g per m2 per year (p. 259). "Part of the stimulation of grassland productivity by grazing was due to maintenance of the vegetation in an immature, rapidly growing state" (p. 259). However, the study apparently didn't measure belowground primary productivity because "Estimates of belowground standing crop were highly variable" (p. 262). So is it possible that higher aboveground productivity came at the cost of lower belowground productivity? Also, even if productivity is higher with grazing, the fraction of productivity eaten by big herbivores is also much higher, so the amount of productivity left over for bugs isn't obvious.

This study of grazing by grasshoppers found that aboveground net primary productivity (ANPP) was not significantly reduced and in 2/5 experiments was increased by grazing.

This page reports that "At low nematode densities, feeding by nematodes stimulates the growth rate of prey populations. That is, [...] plant-feeders stimulate plant growth [...]. At higher densities, nematodes will reduce the population of their prey. This may decrease plant productivity [...]."

This paper created a theoretical model of grazing optimization based on the premise that herbivores might lose less nitrogen when recycling plant matter than detritus recycling would, especially in regions like "the humid savanna of Lamto, Ivory Coast, Africa", where annual fires can remove nitrogen. The authors "conclude that grazing optimization is likely to occur in the Lamto savanna."

Cattle in a field below Garth, Llanbedr - - 578739This paper reviews the debate over the grazing optimization hypothesis and concludes that it may operate in certain contexts but that it's a relatively minor issue compared with degradation of productivity by overgrazing.

This book says it's "problematic" to apply the grazing optimization hypothesis in many real-world settings for various reasons. "Given the inherent variation in grazing distribution on extensively managed grazing lands, it is likely that vegetative growth can be 'optmized' only on a small percentage of area available, even when environmental conditions are favorable for plant growth." In contrast, overestimating how much one can graze "can be devastating to the long-term productivity and stability of a site."

Agnostic views

This paper says it assumes that the NPP "of natural, non-degraded grasslands is not affected by low to moderate grazing pressure, as it is impossible to judge whether disturbance effects (e.g. biomass destruction due to trampling) or over-compensatory plant growth induced by grazing and fertilization from manure droppings dominate. Furthermore, evidence from Sub-Saharan Africa suggests that livestock introduction to natural grasslands with populations of large-herbivores will decrease the carrying capacity of natural herbivores resulting in unaltered grazing pressure." However, the paper adds that this agnosticism "does not hold true for artificial grazing lands or on degraded land."

Cultivated pastures

Sometimes pasture grasses are planted, in a similar way as people plant agricultural crops. The table in this section paints a sketchy picture of the extent of cultivated vs. non-cultivated grazing land in the US, since "pasture" as defined there seems to roughly mean cultivated pasture, while "rangeland" seems to mean mainly uncultivated pasture.

It's plausible that cultivated pastures have higher productivity than native pastures, since increasing productivity is presumably one of the motivations for planting them. (Another motivation might be to grow more cattle-edible plants? I haven't looked into this topic enough, so what I say in this section is only tentative.)

This paper says, "In a previous study, we did consider" that NPP could be increased due to human activity "on fertilized pasture land in countries with intensive pasture management".

The pros and cons of cultivated pastures may be similar to those for crop cultivation. For instance, fertilizers and irrigation may significantly increase productivity (and hence insect suffering). On the other hand, poor management may also degrade long-term productivity:

There are concerns that soils under pastures in certain regions of South Africa are degrading as a result of mismanagement, which include practising continuous tillage, improper grazing management, injudicious application of fertilisers and poor irrigation management.

This paper about grazing in Brazil explains:

Generally, in the first years, and as a consequence of the increased soil fertility from incorporation of the ashes [of burned vegetation], cultivated pastures have a sufficiently high productivity. However, later, especially after five or six years of use, a more or less gradual diminishing of the grasses is seen, especially P. maximum. It is estimated that in a period of less than 20 years of developing cultivated pastures in the region, there already are approximately some 500,000 ha degraded or in an advanced stage of decline [relative to a total of "about 2.5 million ha of cultivated pastures in forested areas".] [... However,] The use of Amazonian forest areas as cultivated pastures for several years results in an improvement of the soil chemical properties [...].

Sometimes crops are planted on pasture land. For example, corn may be grown for grazing, and corn produces ~5 tons per acre.d


I would guess that irrigating dry lands to increase growth of pasture vegetation increases total net primary productivity, even after accounting for possible reductions of riparian and groundwater-dependent vegetation due to water withdrawal. If this is the case, then irrigated pastures sadly probably significantly increase NPP.

This post says "Since much of the hay/alfalfa grown in the arid West requires irrigation, the impoundment of rivers with dams is yet another consequence of livestock production.[xxiv]"

EWG (2011): "In California, feed production for animals, principally alfalfa, corn forage, hay and pasture account for 34 percent of all irrigated water pumped in the state".

Oppenlander (2013), pp. 94 and 98:

In California, over 900,000 acres of land are used to raise alfalfa (from 2000 to 2008, it was over 1 million acres), where 75 percent of this alfalfa is given to dairy cows. It's a terrible misuse of land and water because every one of those 900,000 acres in California is irrigated at an average rate of 1 million gallons of water per growing season of six months. [...]

Worldwide, alfalfa is grown on approximately 79 million acres. The majority of it is irrigated[...].

Beckett and Oltjen (1993) report numbers of cattle and amount of irrigated pasture by region of the USA (Tables 1, 2, and 10). I used this information to create the following table, where the last column gives an idea of the intensity of pasture irrigation per cow by region.

Column 1 Column 2 Column 3 Column 4 Column 5 = 1000 * (Column 4)/(Column 2 + Column 3)
Region Beef cow population on 1 Jan. 1992 Replacement heifer population on 1 Jan. 1992 Total water applied on irrigated pasture for beef production (megaliters) Amount of pasture irrigation per cow in the population (kiloliters/cow)
WA, OR, ID 1,544,000 285,000 2,638,717 1400
MT, ND, SD 3,854,000 725,000 969,008 210
WY, CO, UT 1,838,000 368,000 2,915,965 1300
NM, TX, OK 7,933,000 1,270,000 479,793 52
KS, NE 3,085,000 455,000 264,304 75
MO, IA, MN, WI, IL, MI, IN, OH, NY, PA 5,421,000 881,000 22,379 3.6
CA, AZ, NV 1,426,000 226,000 3,727,088 2300
Other 8,733,000 1,537,100 225,353 22

If we assume that the increase in NPP due to irrigation is roughly proportional to the volume of irrigation water used, then the last column of this table gives a sense of in what regions beef production increases NPP due to irrigation the most per cow. In other words, other things being equal, beef production in the dry US West probably increases NPP (and hence invertebrate suffering) more than beef production in wetter Eastern US states.


This page says:

Lands that have sufficient rain to support grasslands can be and are desertified by [among other things ...] livestock grazing beyond the area’s meager capacity, or by grazing animals that differ from those that naturally (originally) grazed the area.

This report says:

A desertified ecosystem is irreversible regarding natural regeneration. It appears that it takes a very long time for rangelands to become desertified due to overgrazing. This is evidenced by the rangelands in the Near and Middle East which have the longest history of domestic livestock grazing in the world. These lands have been grazed for centuries and heavily overgrazed for the last 50 or more years. Yet, many have not reached the decertified stage [...]

Overgrazing is categorically blamed for worldwide desertification, which is partly true and partly false, depending on the situation. [...]

Stocking rates and managerial systems that result in continual destructive grazing are a major cause of desertification on rangelands. The desertification process is accelerated when these practices are maintained during droughts and certain seasons where plants are highly vulnerable to abuse. While it takes a long time, this will eventually result in decertified rangelands.

Destructive grazing also causes poor livestock performance and many private landowners have learned this lesson the hard way. It is, therefore, not as common on private lands as it is on public or community grazing lands where central control is lacking. [...]

That said, the report also claims that "livestock can be used to arrest and reverse the desertification process".

This study reports that "Grazing pressures are surpassing the recommended stocking rates by a factor 2-6 on most rangeland of northern Mexico."

Savory grazing

Allan Savory claims that his approach to grazing can reverse rather than contribute to desertification. He says poop and urine left by herbivores that move from one grazing location to the next can cover soils and prevent desertification.

Savory's claims have been widely criticized. This page says:

Savory has been around for a very long time preaching the same fallacious grazing gospel, and his name raises curled lips among land management scientists the way Velikovsky's name raises the ire of astronomers. He's merely the latest practitioner of a tradition a couple centuries long of land management mythologies based on wishful thinking that don't turn out to work.

It seems clear that Savory's statement about the potential for proper grazing to reverse climate change is absurd, and combined with Savory's dismissal of challenges to his ideas, it seems clear that even if Savory's approach has some merit, he's not highly intellectually trustworthy.

The best argument on Savory's side seems to be that some practitioners vouch for his method anecdotally, including smart people like Joel Salatin and L. Hunter Lovins. Lovins wrote:

I used holistic management several decades ago on a thousand acres of ground taken out of cattle production for 20 years and let to degrade, erode and be overrun by noxious weeds. Its owners believed that resting land would increase its health. This may be true of intact wilderness. But it is demonstrably not true of most of the planet’s agricultural lands [...]. In our case, we restored cattle to the ground, managed as Savory advised, and within two years watched the water table rise, wetland plants returned and the economic value of the property increase.

Of course, those who attest to the success of Savory grazing might be like a practitioner of unvalidated alternative medicine, who sees an improvement and assumes it resulted from a particular treatment rather than from other causes or regression to the mean. But given how complex the world is, I wouldn't be surprised if Savory grazing sometimes improves productivity in some places.

In any case, whether or not Savory grazing can reverse desertification, for the purposes of this essay, it's sufficient to note that Savory grazing is not used by most of the farms that produce the world's beef. The impact of marginal beef consumption (except beef purchased from atypical farms) should be evaluated relative to the average effects that cattle grazing has on land productivity.

This page says that Savory's method might possibly have some validity in Africa, but Australia and the Americas in prehistoric times didn't have the same huge herds that Africa does, so Savory grazing is less likely to apply there. Since most of the beef that Western consumers eat doesn't come from Africa, this observation is especially relevant to them. However, Sheldon Frith rebuts this claim and asserts that "Allan’s Techniques have been proven to be very effective in Australia and the Americas…."

Effects on soil


This book says: "Cattle treading, or trampling, not only decreases soil surface cover, but can also destroy soil surface structure and increase soil bulk density and surface roughness (Betteridge et al., 1999; Russell et al., 2001), especially of moist soils." This can increase losses of nitrogen and phosphorus by 2-3 times.

This study explains one reason why overgrazing can reduce soil fertility:

Unlike nitrogen deposited in plant litter, N in animal excreta is not distributed uniformly across the pasture but much of it is concentrated around drinking stations and shaded or rest areas where, owing to severe trampling, there is little grass to benefit from the nutrients. The results presented by Cantarutti et al. (2002) suggest that it is due to the fact that at high stocking rates most nitrogen is inefficiently recycled via animal excreta, while at more modest stocking rates (typically 1-3 AU/ha) a large proportion of the nutrients are efficiently recycled via plant litter, and plant productivity can be maintained for many years.

This page says

Through their urine and feces, cattle recycle nitrogen, phosphorus, potassium and other plant nutrients and return them to the soil. [...]

Cattle, through their fecal matter, feed bacteria, fungi, protozoa, beneficial nematodes, and other microorganisms found in the soil. These tiny organisms keep nitrogen and other soil nutrients in a water-soluble form, allowing them to be slowly released into the ecosystem as food.

This study found that "Grazing increased net N mineralization rates [part of nitrogen nutrient cycling] from 87% to 617% compared to watersheds without grazers".

While cattle grazing on pasture fertilize the pasture with their manure, cattle grazing in feedlots produce concentrated waste. This report says "The manure from cattle feedlots is stored on site until it is spread on to nearby farm fields." So manure of concentrated cattle feeding also fertilizes soil, although perhaps if crop farmers had less manure fertilizer, they'd just use more artificial fertilizers, in which case the counterfactual fertilization impact of the manure of concentrated cattle is less than one might think?

Eating cattle may also slowly deplete nutrients on pasture fields and send those nutrients to wherever the urine/feces of beef-eating humans go. I don't know how significant this transfer of nutrients is.

Trampling and compaction

When cows trample grass, they may kill formerly alive plants. This may reduce the food available to bugs that eat live grass, but it increases food available to bugs that eat dead grass. So the short-term impact on bugs isn't obvious. However, trampling might matter if it changes the growth rate of future grass.

This document estimates that cows eat 2.5-3% of their weight as forage each day and trample 0.5% of their weight of forage each day.

This page says of regular, non-Savory grazing:

What the cattle will do, is range over the entire field, taking one bite of the very best grass, then graze it for a second time and so on, until the least palatable grass has been eaten, by which time the field will have been trampled on eight or ten times, over a period of up to a week.

Under such treatment, the recovery of grass is delayed, partly because of the compaction and soil damage caused by repeated trampling over time [...].

This page claims that "The hoof action of cattle is advantageous to certain species of prairie plants."

Extent of damaged land

This report says:

About 20 percent of the world’s pastures and rangelands, with 73 percent of rangelands in dry areas, have been degraded to some extent, mostly through overgrazing, compaction and erosion created by livestock action.

This page roughly agrees: "About 70 percent of all grazing land in dry areas is considered degraded, mostly due to poor grazing practices."

This document reports on a US Bureau of Land Management (BLM) survey of "rangeland health" (which includes consideration of soil, water, air, and plants):

35 percent of all 21,273 BLM grazing allotments had been evaluated by the end of fiscal year 2002. This represents evaluation of more than 36 percent of all BLM land in allotments (BLM 2002). Of these assessed allotments, 76 percent were meeting all standards, 8 percent were not meeting all standards for reasons other than livestock grazing, and current livestock grazing management practices or levels of grazing use were determined to be a significant factor in the failure of the remaining 16 percent of all allotments assessed to achieve the standards and conform to the guidelines.

This page says:

Scientists refer to riparian zones as hot spots of biodiversity[...]. The presence of water, increased productivity, favorable microclimate, and periodic flood events combine to create a disproportionately higher biological diversity than that of the surrounding uplands.[56]

[...] A 1988 report of the GAO [estimated] that 90% of the 5,300 miles of riparian habitat managed by the BLM in Colorado was in unsatisfactory condition, as was 80% of Idaho's riparian zones, concluding that "poorly managed livestock grazing is the major cause of degraded riparian habitat on federal rangelands."[57]


Overgrazing can cause soil erosion, thereby reducing primary productivity (and probably animal populations) in the future. This page explains that "trace minerals, such as, zinc, copper, and manganese" are being depleted by "water erosion of the soil and over cropping of land".

This page says:

In soils that have restrictions to root growth, erosion decreases rooting depth, which decreases the amount of water, air, and nutrients available to plants. Erosion removes surface soil, which often has the highest biological activity and greatest amount of soil organic matter.

This report says:

Soil erosion is one of the most serious problems affecting the sustainability of agriculture in Turkey because as much as a third of the cultivated land and extensive areas of rangelands and mountain pastures have steep slopes. About 16 million hectares, or more than 70 percent of the cultivated and grazed land area in Turkey, are affected by erosion, especially in the upper watershed of the Euphrates River in Eastern Anatolia. Extensive livestock systems are a main culprit. Poor rangeland management has led to extensive soil degradation, limiting the scope for natural forest regeneration, and contributing to greatly increased soil sedimentation.

This page notes that "Manifestations of overgrazing in landscapes composed largely of native species include [...] desertification, loss of native topsoil and increases in surface runoff. [...] significant topsoil loss has a regeneration time scale of tens of millennia."

This page explains:

Overgrazing can increase soil erosion. Reduced soil depth, soil organic matter, and soil fertility hurt the land's future productivity. Soil fertility can be corrected by applying the appropriate lime and fertilizers. However, the loss of soil depth and organic matter takes years to correct. Their loss is critical in determining the soil's water-holding capacity and how well pasture plants do during dry weather.

In contrast to the above, this page claims that grazing can sometimes reduce erosion:

By trampling plants that have grown too coarse and brittle to eat, cattle increase the amount of litter on the ground. This reduces soil and water erosion and helps increase the amount of water that enters the ground and aquifers. [...]

The fecal matter of cattle supports insects and organisms (dung beetles and earthworms) which aid in carbon sequestration and water infiltration. This increases the amount of carbon and water stored in the soil, and helps recharge aquifers.

Impact on aquatic animals

Soil erosion can also destroy aquatic habitats, causing significant short-term suffering but possibly reducing the long-run suffering of aquatic animals?

This book says: "Grazing by livestock has damaged as much as 80% of the streams and riparian ecosystems in arid regions of the western USA (Belsky et al., 1999)."

Erosion may increase and/or decrease productivity in water bodies. Excessive soil erosion can inhibit animal growth in water bodies.

This page explains:

Nutrients removed by erosion are no longer available to support plant growth on-site, and when they accumulate in water, algal blooms, lake eutrophication, and high dissolved oxygen levels can occur.

Extent of erosion in the US

This survey reports on soil erosion due to grazing (among other causes) in the 48 contiguous states of the US. It defines (p. 15) T, the "Soil loss tolerance factor" as "The maximum rate of annual soil loss that will permit crop productivity to be sustained economically and indefinitely on a given soil." Then, in Tables 14-15, it presents the following data for how many millions of acres of non-Federal rural pastureland have erosion within a given amount of the threshold T in 2007. (I rounded the numbers and omitted error bars for simplicity.)

<= T T to 2T 2T to 3T 3T to 4T 4T to 5T > 5T
"Estimated average annual sheet and rill erosion" 113 4 1 0.4 0.2 0.4
"Estimated average annual wind erosion" 118 0.3 0.1 0.1 0.01 0.1

Land-use change

Sometimes forests or wetlands are converted to pasture:

historical grazing, along with other land conversion, in Northern and Central California has reduced native chaparral and forest lands by approximately 70 percent. Ongoing grazing expansion {and land conversion} driven by human population growth in this region threatens the remaining integrity of California chaparral and woodlands habitat in this region.[24] [...]

Much grazing land has resulted from a process of clearance or drainage of other habitats such as woodland or wetland.[25]

Grazing cattle also prevents regrowth of forests or other native ecosystems. In cases where the native ecosystems would be more productive, this seems like a positive impact of cattle pasturing.

This post claims that "the majority of new livestock pasturage is carved from forested landscapes. (Most natural grasslands are already under livestock production and have little space available for increasing animal numbers.)" I'm unsure if this is true in temperate regions like the USA, but even if marginal increases in demand for beef come from grazing on grasslands rather than converting forest to grassland, greater beef demand may increase pressure for overgrazing -- which is, like forest loss, more likely to reduce insect suffering than moderate grazing in general.

As a counterpoint, this article says that grazing helps trees increase in numbers because grazing (1) removes competitor grasses and (2) leaves more barren soil, which helps tree seeds flourish.

Pasture vs. forest in the US

This survey reported the land-use changes in the 48 contiguous states of the US between 1982 and 2007. The following table shows transitions from one land-use type to another, in millions of acres. (I rounded the numbers from the original source and omitted error bars to keep things simple.)

Pastureland in 2007 Rangeland in 2007 Forest land in 2007
Pasturelande in 1982 78 5 18
Rangelandf in 1982 3 392 3
Forest land in 1982 5 2 372

So the decrease in forest to make way for grazing was only 5 + 2 = 7 million acres, while the increase in forest on former grazing land was 18 + 3 = 21 million acres. Despite this, beef production in the US increased between 1982 and 2007.

Of course, these numbers don't prove that marginal increases in beef production don't appropriate former forest land. Presumably beef production has become more efficient in the last few decades, but that doesn't mean that slightly increasing beef production doesn't slightly increase pressure to convert forests to pasture. But at least, the effect may not be huge.

Biodiversity loss

This report explains that

Some 306 of the 825 terrestrial ecoregions identified by the Worldwide Fund for Nature (WWF) – ranged across all biomes and all biogeographical realms, reported livestock as one of the current threats. Conservation International has identified 35 global hotspots for biodiversity, characterized by exceptional levels of plant endemism and serious levels of habitat loss. Of these, 23 are reported to be affected by livestock production. An analysis of the authoritative World Conservation Union (IUCN) Red List of Threatened Species shows that most of the world’s threatened species are suffering habitat loss where livestock are a factor.


The net impact of fires on wild-animal suffering isn't obvious. On the one hand, fires are a non-sentient way to eliminate stored plant energy and thus plausibly reduce wild-animal populations somewhat. On the other hand, the pain of burning alive seems to me to be worse than the pain of most other kinds of death.

This study reported that in the Serengeti, "Fire does not appear to have important effects upon the functional properties of the grasslands except for a weak stimulation of productivity in the wet season immediately following dry season burning." If fire actually (slightly) increases net primary productivity (rather than just taking away resources from plant roots), that's (slightly) bad insofar as it increases the rate of production of food for bugs.

Articles suggesting grazing reduces fire risk

This pro-beef article presents several studies suggesting that grazing can reduce risk of extreme fires. It notes that compared against fires, "grazing reduces fuel load in a more selective fashion (Archer 1999) avoiding the potential sterilizing effect that an extremely intense fire may have on soil." If true, this is a bad effect of grazing, since soil sterilization would probably reduce future insect populations.

This article reports favorably on the use of goats to reduce fire risk. Since companies sometimes choose to bring in goats, they probably believe grazing will be effective at preventing fires.

This article says

management in many parks makes use of grazing to help lower fire hazards by reducing the amount of potential fuel, such as large buildups of forage. When the land is not grazed, dead grasses accumulate. These dead grasses are often a large fire hazard in the summer months.

Articles suggesting grazing increases risk of big fires

This article claims that grazing increases risks of intense fires. One reason is that "grazing removes fine fuels, such as grasses, that historically helped carry light intensity fires that once burned at regular intervals throughout lower elevation forest ecosystems in the West." If this is true, the impact of grazing on the total amount of wildfires (weighted by the size of each fire) isn't obvious.

This page agrees: "overgrazing of native fire-carrying grasses has starved some western forests of fire, making them overly dense and prone to unnaturally severe fires."


  1. Lots of bugs may eat decaying grass at various stages. In summer 2016, I saw a few buckets of grass clippings from a nearby church lawn that had begun getting moldy, and the grass clippings were full of what looked like fungus gnats. This piece notes that "Piles of decaying grass clippings [...] serve as good breeding places for flies." And of course, there are tons more decomposer bugs within the soil.  (back)
  2. Keep in mind that species richness is not the same as abundance of individual organisms. So it's unclear how much we can learn about total populations of insects from this statement.  (back)
  3. I haven't read the whole study, so take my comments here with salt.  (back)
  4. Todo: How does this compare with the productivity of native grasses?

    This document gives sample productivities for other grasses:

    • "bermudagrass and ryegrass" as 5.75 tons per acre ("11,500 lb/ac/yr")
    • "fescue" as 3.75 tons per acre ("7,500 lb/ac/yr").


  5. The report defines "pastureland" as "land managed primarily for the production of introduced forage plants for livestock grazing. Pastureland cover may consist of a single species in a pure stand, a grass mixture, or a grass-legume mixture. Management usually consists of cultural treatments: fertilization, weed control, reseeding or renovation, and control of grazing. For the NRI, includes land that has a vegetative cover of grasses, legumes, and/or forbs, regardless of whether or not it is being grazed by livestock."  (back)
  6. The report defines "rangeland" as land "on which the climax or potential plant cover is composed principally of native grasses, grasslike plants, forbs or shrubs suitable for grazing and browsing, and introduced forage species that are managed like rangeland. This would include areas where introduced hardy and persistent grasses, such as crested wheatgrass, are planted and such practices as deferred grazing, burning, chaining, and rotational grazing are used, with little or no chemicals or fertilizer being applied. Grasslands, savannas, many wetlands, some deserts, and tundra are considered to be rangeland. Certain communities of low forbs and shrubs, such as mesquite, chaparral, mountain shrub, and pinyon-juniper, are also included as rangeland."  (back)