Peak Phosphorus

As highlighted by Hubbert first in 1949 (HUBBERT 1949), production of oil resources will eventually reach a maximum rate or “peak” based on the finite nature of non-renewable resources, after which production will decline. Hubbert and later others contest that the important period is not when 100% of the resource is depleted, but rather when it reaches a production maximum, which occurs when 50% of the resource is still in the ground. After this point, production decreases, placing upward pressure on prices and increasing international tensions (CAMPBELL 1997). While the exact timing may be disputed, it is clear that already the quality of remaining phosphate rock reserves is decreasing and cheap fertilisers will be a thing of the past. Like in the 1970’s, phosphate rock is experiencing its first significant price shock – a 700% increase from USD 50 to 350 USD/tonne in just 14 months (LEWIS 2008)

Yet there are no alternatives to phosphate rock currently on the market that could replace it at any significant scale. While various small-scale trials are being undertaken, commercialisation and implementation on a global scale could take decades to develop (see e.g. struvite precipitation).

Most of the used amount of phosphorus origins from phosphate rock. Whereas in 1945 manure was still the main source of phosphorus fertiliser from this point of time onwards the use of phosphate rock as a source of phosphorus fertiliser increased rapidly and started to be the by far largest source (see figure below).

Sources of phosphorus fertilisers. Most of the used amount origins from phosphate rock. Source: CORDELL et al. (2009).

The largest commercially recoverable reserves of phosphorus are located in just three countries: China, the United States and Morocco/Western Sahara (ROSEMARIN et al. 2009).

Distribution of global phosphate rock reserves. Source: ROSEMARIN et al. (2009).

This analysis of peak phosphorus is based on estimated P in current world phosphate rock reserves (approximately 2358 MT P) base on US Geological Survey date (BUCKINGHMA and JASINSKI 2006; JASINSKI 2007/2008) and European Fertiliser Manufacturers Association (2000). The area under the Hubbert curve must equal the depleted P, rather than P₂O₅(containing 44% P) or phosphate rock (containing 29-34% P₂O₅) as commonly used by industry.

The date for production is fitted using a Gaussian distribution (LAHERRERE 2000), based on the depleted plus current global reserves estimate of 3’212 MT P, and a least squares optimisation which results in a global production at peak of 28 MT P/a and a peak year of 2034.

The peak phosphorus curve, which includes the modelled Hubbert curve, and the actual phosphorus production per year, represented by the black points. Source: CORDELL et al. (2009).

However, the actual timing may vary due to production costs (such as price of raw materials like oil), data reliability and changes in demand and supply. As the following example illustrates, estimations of peak phosphorus can be mislead if they are only oriented on the phosphorus production rate, as production rates can be influenced by many factors (e.g. political factors, other resource prices, etc.).In 2007, a peak phosphorus article (DÉRY and ANDERSON 2007) estimating the U.S phosphate rock reserves peaked almost 20 years ago, around 1989. This same article suggests global reserves peaked around the same year. Whilst there was indeed a production peak in this year, like oil peaks in the 70’s, this observed peak was not a true maximum production peak, and was instead a consequence of political factors such as the collapse of the former Soviet Union and decreased fertiliser demand from Western Europe. According to USGS staff, Moroccan and Western Sahara reserves, which account for a significant proportion of today’s global production, are currently being mined at a relatively constant rate that is less than the maximum production.

Thus, at current rate (as of 2009), the US will deplete its reserves within about 30 years while the global reserves may be depleted within the next 75-100 years (ROSEMARIN et al. 2009).

While it is understood that phosphate rock, like oil and other non-renewable resources will follow a Hubbert production curve, a key difference between peak oil and peak phosphorus, is that oil can be replaced with other forms of energy once it becomes too scarce. Whereas there is no substitute for phosphorus in food production (CORDELL et al. 2009). P cannot be produced or synthesised in a laboratory. Quite simply, without phosphorus, we cannot produce food (see also the nutrient cycle).

A second key difference is that oil is unavailable once it is used. While phosphorus is an element that can be captured after use and recirculated for use within economic and technical limits.

Peak phosphorus is also linked to peak oil. For example, the recent oil price shock and growing concern about climate change has stimulated a dramatic increase in biofuel crop production globally, which in turn increases the demand for phosphate fertilisers, and hence the proximity of the phosphorus peak.

A more detailed comparison of peak oil and peak phosphorus is provided in CORDELL and KERSCHNER (2007).

Peak oil sceptics commonly argue that the market will take care of things: that resource scarcity is “relative” and one scarce resource can simply be replaced by another indefinitely, because as price rises, investment in new technology will always improve efficiency of extraction and use (STEWARD et al., 2005). This is the basis of the market system - neoclassical economic theory - which does not acknowledge the finite nature of non-renewable resources like phosphate rock (or oil). This means that the concepts of Peak Oil and Peak Phosphorus are not supported by the market system.Other sceptics do not deny that peaks will one day occur, rather they dispute the timeline and insist a peak is more in the distant future (CAVENY, 2006).

While the Hubbert peak has been hotly contested since its conception almost 60 years ago, it is increasingly gaining traction as the price of oil shoots well beyond US$100/barrel. In November 2007 the then International Energy Agency Chief Economist stated in a noteworthy interview: “if we do not do anything very quickly, and in a bold manner, our energy system’s wheels may fall off – within the next seven years” (FINANCIAL TIMES 2007).

The current system of phosphorus depletion, usage and disposal is inefficient. Two major reasons for this situation are

1) The time frame considered for planning, both in the market (especially fertiliser market) and governmental institutions, is short-term (5-10 years) which inhibits the adequate consideration of phosphorus as a finite resource (CORDELL et al. 2009). As we are learning from climate change and global water scarcity, a long-term time frame is essential for understanding, managing and adapting our current system in a timely way. The same applies to global food security and phosphorus resources.

2) The inconsistency of international institutional arrangements with the natural phosphorus cycle (see also: the nutrient cycle): This is most evident in the divide between the agricultural sector, where phosphorus is perceived as a fertiliser commodity, and the water and sanitation sector, where phosphorus is perceived as a pollutant in wastewater (CORDELL et al., 2009). This may hinder opportunities to find integrated solutions to the scarcity problem (CORDELL et al. 2009).

While the recent price spike in phosphate rock is likely to trigger further innovations in and adoption of phosphorus recovery and efficiency measures, the market currently does not have enough adaptive capacity to manage phosphorus in a sustainable, equitable and timely manner in the long-term.

Even though there is not enough reliable data today to predict the exact year peak phosphorus will occur, what is clear is that discussion on alternative phosphorus sources and governance models is required now to ensure that the world’s farmers have sufficient access to phosphorus fertilisers in the long-term to feed humanity, without compromising the environment, livelihoods and economies.

Another aspect that has to be considered when discussing possibilities to reduce phosphorus demand is the amount of phosphorus a vegetarian diet demands compared to a meat-based diet. A change from an average western diet to a vegetarian diet could decrease the demand for phosphorus fertiliser by 20-45% (CORDELL et al. 2009). A vegetarian diet requires 0,6 kg P/year for one person, whereas a meat-based diet requires 1,6 kg P/year. These estimations were made by calculating the amount of phosphorus backward from human excreta to the field (for more detail see CORDELL et al. (2009)).

The currently inefficient system of phosphorus depletion, usage and disposal presents many opportunities for both increasing efficiency throughout the system, and for capturing used phosphorus in human and animal excreta and food and crop residues. More efficient extraction of phosphate at source would help to reduce wastage (ROSEMARIN et al. 2009). Also farmer could be encouraged to use phosphorus fertilizer more efficiently, switch to organic fertilisers and composting technologies. This would reduce the demand for additionally mined phosphorus (ROSEMARIN et al. 2009).

Probably the most effective way to reduce phosphorus demand and therefore depletion is the recovery and reuse of phosphorus from organic waste and wastewater streams (ROSEMARIN et al. 2009). Examples of phosphorus recovery and reuse are urine fertilisation small-scale and large-scale, struvite,  use of dehydrated faeces, fertiliser from sludge, use of compost and arborloo. In developing countries, especially in Africa where farmers use limited amount of chemical fertilisers, these recovery and reuse techniques could allow them to become self-sufficient. In developed countries the expansion to more phosphorus recycling would require the retooling of the agricultural infrastructure and the adoption of new farming practices (ROSEMARIN et al. 2009).

See also Linking up Sustainable Sanitation Water Management and Agriculture.