Biodiversity, Healthy Soils, and their Combination in Regenerative Agriculture can Reduce but not Replace Fertilizer

“The collective simplification of agroecosystems has led to a loss of biodiversity and to reductions in the supply of key ecosystem services to and from agriculture. Without these ecosystem services, monocultures become dependent on off-farm inputs.” (Kremen and Miles, 2012)

“Dependent on off-farm inputs…” Can it be otherwise? People inside and outside of science like to point to how nature needs no inputs, yet still produces. Then, as above, they wonder why agriculture cannot do the same. If we could just mimic nature on the farm, work with nature, then we would not need these costly inputs with undesired environmental impacts. If we could just restore natural soil fertility, we could give up “hazardous chemical fertilizers.” This theme of “natural processes can replace external inputs” is central to regenerative and organic agriculture, and agroecology, but is it possible? Here I will explore how far biodiversity, soil health, and soil biology can take us in replacing synthetic fertilizers.

“No one is fertilizing the rainforest.” – David C. Johnson, regenerative ag speaker

Quantity of N and P

First, let’s narrow the topic. I will focus on nitrogen and phosphorus, N and P, as they are needed in the greatest quantities for crop growth (Drinkwater et al., 2017). In providing N and P to crops, the most important issues are the quantity of the nutrients and the rate at which they are supplied in a form that plants can use.

We can address the quantity issue quickly. For relying on natural sources of N, we are talking about biological nitrogen fixation, getting N from the air. The air is 78% N, so the source quantity is huge, plenty of N for our needs.

Phosphorus is found in soil minerals and the bedrock which produced them. This quantity of phosphorus in many soils is also large. Outside of old and highly weathered soils in the tropics and subtropics, most soils contain sufficient P in their minerals for many centuries of crop production.

In the very long-term, given a finite quantity of a substance, what is removed must be replaced or the quantity will decrease. Agriculture exports a large quantity of food every year including nutrients taken from the soil. If we do not replace the exported P and other nutrients that come from soil minerals, we will have turned agriculture into a soil mining operation. However, this is not the fundamental problem. Let’s look at the rates of nutrient supply to crops.

Rate of N and P Supplied

Here we must compare the rates of nutrients required for modern crop production to what natural processes give us. The rate is important because many soils can supply enough N and P given enough time. This is how shifting slash and burn agriculture worked; by abandoning the land for several years, the nutrient levels would build up to levels that would produce crops for several years. We no longer have that luxury and so will look for rates of nutrient supply sufficient for a continuous annual harvest of crops.

Tropical rainforest
Tropical rainforest canopy near Cairns, QLD.

“It is often surprising for students of agriculture to learn that some of the most productive terrestrial ecosystems on Earth—the tropical rainforests—often grow from the most highly weathered and thus least fertile soils. How is this possible? Simply put, an array of nutrient cycling mechanisms prevent nutrient losses from exceeding what are often very low levels of nutrient inputs, thus allowing for high levels of photosynthesis to take place.” Crews 2005. Photo: CSIRO, Wikimedia Commons

Nitrogen from Organic Matter

The amount of nitrogen in soil organic matter (~5%) can be large, but not large enough that losses can be ignored. Therefore, N (and P) released from decomposing organic matter must be replaced or soil health declines. We should recognize the N and P release from soil organic matter in management decisions, but this will not replace fertilizers in the long-term.

Nitrogen from Legumes

Biological nitrogen fixation is found in two forms: symbiotic N-fixation in legumes and non-symbiotic N-fixation by free-living bacteria in the soil. We already get a lot of N from legumes through production of soybeans, alfalfa, dry edible beans, etc. But could legume-N replace synthetic-N fertilizers? Crews and Peoples (2004) studied the issue and concluded, in agreement with two other studies, that our population is beyond that which could rely on legume-N to feed us. We cannot simultaneously use more land for legume N-fixation, increase food production for an increasing population, and limit expansion of farmland to limit climate change effects.

Nitrogen from Soil Biology

Free-living bacteria can fix nitrogen from the air without being in a root nodule. Estimated annual rates of N provided from these bacteria range from <1 to ~19 lb. N per acre (Cleveland et al., 1999; Ladha et al., 2016). These rates are much lower than needed in modern agriculture. Unless we can increase their fixations rates, free-living bacteria will not replace N fertilizer.

[Illustration] Legumes, bacterial fixing N; Fertilizer-introduced P. Additional P present from biological weathering and ???
Soil nitrogen and phosphorus sources other than fertilizer.
Some suggest that both legume-associated and free-living N fixation be used to supplement, not replace, N fertilizers. The problem here is that use of even moderate amounts of N fertilizer shuts down biological N fixation in soils (Smercina et al., 2019) and in legume roots. If the N is in the soil from fertilizer, legumes and bacteria use that N and save the high energy cost of biological N fixation.

Phosphorus from soil biology

The ratio of N to P in plants is about 10:1, so crops need P in much lower quantities than N. However, soil microbes cannot add P to the soil. Rather, mycorrhizal fungi and other microbes more effectively tap into what is already there. The often-mentioned mycorrhizal fungi are a fungal fracking system, helping plants extend their range and extract more P and N.

Newman (1995) estimates the maximum annual rate provided through biological weathering of soil minerals (mycorrhizae and other microbes) at 5 lb. P per acre in a soil with high P concentration and high weathering rate. More typical rates are around 0.3 lb. P per acre per year. These rates will not sustain the high yields of modern agriculture. And as with N fertilizer, using P fertilizer to bring P supply levels up to crop needs often stops these biological mechanisms from operating in soils.

Mycorrhizae and other microbes can help with nutrient cycling, potentially improving a crop’s nutrient use efficiency, but again, this does nothing to reduce the need to replace the P taken off in harvest each year.

Biofertilizers are not Fertilizers

Is there potential for biofertilizers to replace fertilizers? First, despite the name, biofertilizers are not fertilizers: “biofertilizers are living microbes that enhance plant nutrition by either mobilizing or increasing nutrient availability in soils.” (Mitter et al. 2021). As concluded above, these microbes can help improve nutrient cycling and efficiencies, but will not replace fertilizer in the long-term. With few exceptions, the rates of nutrient supply provided by microbes from the mineral portion of soil are low, and increasing these rates only depletes the soil faster.

Residual phosphorus

Using P fertilizer is notoriously inefficient: much of the applied P ends up in forms not available to crops (“slowly exchangeable,” see van der Bom et al., 2019). However, recent research (Menezes-Blackburn et al., 2018; Wade et al. 2019) finds that crops can use much of this “residual P”, and soil biology is probably a major factor in helping crops access it. Menezes-Blackburn et al. (2018) speculate that with management, this pool of soil P might supply crops for 100+ years on soils with a history of high P fertilizer application. Wade et al. (2019) found that residual P could maintain corn and soybean yields for at least 11 years without P inputs. While drawing down this residual P could improve profits and reduce the environmental impacts of P, this pool is nowhere near as large as the soil mineral P pool in many soils. Whether residual P lasts 11, 25, or 100 years, agriculture remains dependent on P fertilizer in the long-term and often the short-term.

Results of Long-term Unfertilized Experiments

Several long-term experiments show that unfertilized crops can produce for a long time, but at much reduced yields. The hay plots at Rothamsted, England, have been in production since 1856. These unfertilized plots have reached a stable, but low production level of 1.2 tons per acre per year (Jenkinson et al., 1994). Glover et al. (2010) found similar results with Kansas prairie hay over 75 years. Corn yields from the unfertilized Murrow plots in Illinois averaged 36 bu. per acre for over 100 years (Aref and Wander, 1997). In a study of British grain production and field P balances, Newman (1997) found that P fertilizer inputs and P exports in grain exceeded natural inputs and losses by at least a factor of 10. Again, the problem when relying on natural soil fertility is not the nutrient quantity available but the rate of nutrient supply, which is not fast enough to maintain current yields. In some situations this may pay the bills but it will not feed us all.

In an insightful study, Crews (2005) investigates if a perennial cropping system can rely solely on nutrients from weathering of minerals and nitrogen fixation. He concludes that such a system might be possible under specific conditions: young soils with high nutrient release rates, improved synchrony between crop demand and nutrient availability, and optimized management of legume density, harvest, and N-fixation in mixtures. And, I might add, development of high-yielding perennial grains (Vico et al., 2016). The author did not make a case for doing this with annual grain cropping systems.

3-D visualization of a root tuber
Root tuber colonized by an arbuscular mycorrhizal fungus. Scivit, Wikimedia Commons

Agriculture is not Nature

In my review of research, I find little support that healthy soils, biodiversity, or the combination in regenerative agriculture, can replace fertilizer inputs in modern agriculture. Farmers can sometimes increase profits by cutting inputs, and cut inputs by reducing cropping intensity, but what is good for the individual farm may not be good for society in general. Balancing farm profit with global food production involves policy and a many other issues that are not my focus here.

Let’s get back to the original assumption, that because nature sustains itself without fertilizer, a farm should as well. This compares the inputs of nature and agriculture; let’s compare their outputs. Nature provides food for few people, farms feed billions. Unmanaged nature produces mostly low-N, inedible biomass from native perennials. Managed agriculture produces highly edible, high N (protein) food from improved annual crops. Nature recycles nutrients in place, exporting little. Agriculture loses most of its nutrients every year in the harvest of food. Why are we comparing these two systems if they have different parts and outcomes?

Using nature as a model for agriculture is like using an airplane as a model for a car because we like the way the plane flies. Planes and cars share some basic features – wheels and engines – but they have fundamentally different objectives, and therefore designs.

Even in unmanaged ecosystems where biological processes are presumably humming along nicely, research finds they are often limited by N or P availability (Vitousek et al. 2002; Hou et al., 2020). If natural ecosystems that export little or no food are nutrient deficient, then why would we expect agriculture with its high level of food export to be self-sufficient in nutrients?

“What can we change in agriculture to make it more like nature?” is the wrong question. Agriculture resulted from asking, “What can we change about nature to get more food?” It should not be surprising that most common result of making agriculture more like nature is less food. Therefore, we need an agriculture different from nature, one that produces lots of food. Such an agriculture will run differently from nature, it must. While some aspects of nature may be instructive, many are not, because nature is not agriculture, agriculture is not nature.


  • Aref, S., and M.M. Wander. 1997. Long-Term Trends of Corn Yield and Soil Organic Matter in Different Crop Sequences and Soil Fertility Treatments on the Morrow Plots. In: Sparks, D.L., editor, Advances in Agronomy. Academic Press. p. 153–197
  • van der Bom, F.J.T., T.I. McLaren, A.L. Doolette, J. Magid, E. Frossard, et al. 2019. Influence of long-term phosphorus fertilisation history on the availability and chemical nature of soil phosphorus. Geoderma 355: 113909. doi: 10.1016/j.geoderma.2019.113909.
  • Cleveland, C., A.R. Townsend, Schimel, D, Fisher, Hank, Howarth, Robert W., et al. 1999. Global patterns of terrestrial biological nitrogen (N2) fixation in natural ecosystems. Global Biogeochemical Cycles 13(2): 623–645. doi: 10.1029/1999GB900014.
  • Crews, T.E. 2005. Perennial crops and endogenous nutrient supplies. Renewable Agriculture and Food Systems 20(1): 25–37. doi: 10.1079/RAF200497.
  • Crews, T.E., and M.B. Peoples. 2004. Legume versus fertilizer sources of nitrogen: ecological tradeoffs and human needs. Agriculture, Ecosystems & Environment 102(3): 279–297. doi: 10.1016/j.agee.2003.09.018.
  • Drinkwater, L.E., M. Schipanski, S. Snapp, and L.E. Jackson. 2017. Chapter 7 – Ecologically Based Nutrient Management. In: Snapp, S. and Pound, B., editors, Agricultural Systems (Second Edition). Academic Press, San Diego. p. 203–257
  • Glover, J.D., S.W. Culman, S.T. DuPont, W. Broussard, L. Young, et al. 2010. Harvested perennial grasslands provide ecological benchmarks for agricultural sustainability. Agriculture, Ecosystems & Environment 137(1): 3–12. doi: 10.1016/j.agee.2009.11.001.
  • Hou, E., Y. Luo, Y. Kuang, C. Chen, X. Lu, et al. 2020. Global meta-analysis shows pervasive phosphorus limitation of aboveground plant production in natural terrestrial ecosystems. Nature Communications 11(1): 637. doi: 10.1038/s41467-020-14492-w.
  • Jenkinson, D.S., J.M. Potts, J.N. Perry, V. Barnett, K. Coleman, et al. 1994. Trends in herbage yields over the last century on the Rothamsted Long-term Continuous Hay Experiment. The Journal of Agricultural Science 122(3): 365–374. doi: 10.1017/S0021859600067290.
  • Kremen, C., and A. Miles. 2012. Ecosystem Services in Biologically Diversified versus Conventional Farming Systems: Benefits, Externalities, and Trade-Offs. Ecology and Society 17(4). doi: 10.5751/ES-05035-170440.
  • Ladha, J.K., A. Tirol-Padre, C.K. Reddy, K.G. Cassman, S. Verma, et al. 2016. Global nitrogen budgets in cereals: A 50-year assessment for maize, rice and wheat production systems. Scientific Reports 6(1): 19355. doi: 10.1038/srep19355.
  • Menezes-Blackburn, D., C. Giles, T. Darch, T.S. George, M. Blackwell, et al. 2018. Opportunities for mobilizing recalcitrant phosphorus from agricultural soils: a review. Plant Soil 427(1): 5–16. doi: 10.1007/s11104-017-3362-2.
  • Mitter, E.K., M. Tosi, D. Obregón, K.E. Dunfield, and J.J. Germida. 2021. Rethinking Crop Nutrition in Times of Modern Microbiology: Innovative Biofertilizer Technologies. Front. Sustain. Food Syst. 5. doi: 10.3389/fsufs.2021.606815.
  • Newman, E.I. 1995. Phosphorus inputs to terrestrial ecosystems. Journal of Ecology 83. doi: 10.2307/2261638.
  • Newman, E.I. 1997. Phosphorus Balance of Contrasting Farming Systems, Past and Present. Can Food Production be Sustainable? Journal of Applied Ecology 34(6): 1334–1347. doi: 10.2307/2405251.
  • Smercina, D.N., S.E. Evans, M.L. Friesen, and L.K. Tiemann. 2019. To Fix or Not To Fix: Controls on Free-Living Nitrogen Fixation in the Rhizosphere. Appl. Environ. Microbiol. 85(6). doi: 10.1128/AEM.02546-18.
  • Vico, G., S. Manzoni, L. Nkurunziza, K. Murphy, and M. Weih. 2016. Trade-offs between seed output and life span – a quantitative comparison of traits between annual and perennial congeneric species. New Phytologist 209(1): 104–114. doi:
  • Vitousek, P.M., K. Cassman, C. Cleveland, T. Crews, C.B. Field, et al. 2002. Towards an ecological understanding of biological nitrogen fixation. The Nitrogen Cycle at Regional to Global Scales. Springer, Dordrecht. p. 1–45
  • Wade, J., S.W. Culman, S. Sharma, M. Mann, M.S. Demyan, et al. 2019. How Does Phosphorus Restriction Impact Soil Health Parameters in Midwestern Corn–Soybean Systems? Agronomy Journal 111(4): 1682–1692. doi: 10.2134/agronj2018.11.0739.