We outline a path to sustainable development that would give future generations the chance to be as well-off as their predecessors without running out of natural resources, especially metals. To this end, we have to consider three key resources: (1) the geosphere or primary resources, (2) the technosphere or secondary resources, which can be recycled and (3) human ingenuity and creativity. We have two resource extremes: natural resources which are completely consumed (fossil fuels) versus natural resources (metals) which are wholly recyclable and can be used again. Metals survive use and are merely transferred from the geosphere to the technosphere. There will, however, always be a need for contributions from the geosphere to offset inevitable metal losses in the technosphere. But we do have a choice. We do not need raw materials as such, only the intrinsic property of a material that enables it to fulfil a function. At the time when consumption starts to level off, chances improve of obtaining most of the material for our industrial requirements from the technosphere. Then a favorable supply equilibrium can emerge. Essential conditions for taking advantage of this opportunity: affordable energy and ingenuity to find new solutions for functions, to optimize processes and to minimize losses in the technosphere.

Sustainable development with regard to non-renewable resources can best be defined in terms of the inter-generational challenge of the Brundtland commission and the intra-generational challenge worked out in Agenda 21 of the 1992 Rio de Janeiro conference of United Nations Conference on Environment and Development (UNCED). In meeting these challenges, the trilemma of security of supply under conditions of economic viability and environmental sustainability also needs to be addressed in order to achieve sustainable development. To fulfil the natural resources needs of future generations we have three resources at our disposal: (1) the geosphere or primary resources; (2) the technosphere or secondary resources and (3) human ingenuity and creativity driving innovation. Man does not need natural resources as such, only the intrinsic property of a material that enables the fulfilment of a function is required. Any material that can perform the same function more efficiently or cheaply can replace any other material. In our constant drive to secure the supply of efficient raw materials, the feedback control cycle plays an indispensable role by virtue of it reacting to price signals on both the supply and demand sides. The feedback cycle of course goes hand in hand with a continuous learning process. On the supply side, the learning effects are in technology development around primary resources and the increased use of secondary resources; on the demand side with thriftier use of raw materials.

Several recent papers deal with concerns about the longevity of the supply of the mineral phosphorus. The paper by Edixhoven et al. (2014), for instance, expresses doubts about whether the upward estimate of reserves by the IFDC (2006, 2010) and the USGS (2010) provides an accurate, reliable, and comparable picture, as it is based on reports that do not clearly differentiate between phosphate ore and phosphate products (i.e., marketable phosphate rock concentrate). Further, the indistinct use of the terms reserves and resources is criticized. Edixhoven et al. (2014) call for a differentiated inventory of world phosphate reserves including "guidelines which determine the appropriate drill hole distances and a detailed granularity". The claim that "humanity is on the safe side" with respect to future phosphate supply is doubted, as the validity of the IFDC's upgrading of the Moroccan data to 50 Gt phosphate is questioned. The main achievement of Edixhoven et al. (2014) is to elaborate that in the literature frequently used data on phosphate rock ore and phosphate concentrate are not properly distinguished, resulting in incorrect summary figures. In addition, it is commendable to raise the question how transparency concerning reserve and resources data and information on the geopotential of phosphate can be achieved because phosphorus is a special element. As fertilizer, it cannot be substituted and there are no unlimited resources as for the other main nutrients potassium in sea water and nitrogen in the air. However, the paper by Edixhoven et al. (2014) contains in the opinion of the authors some incorrect statements. Our comment elaborates first that several statements, such as that the upgrading of the Moroccan data is "solely based" on one scientific paper, are incorrect. Secondly, the paper comments on and illuminates a set of, in our opinion, misleading statements. These include the fact that the dynamic nature of reserves (which depend on price, technology, innovation for exploiting low-grade deposits, etc.) is acknowledged, but the right conclusions are not drawn, including the mixing of finiteness and staticness, and the way in which the critique of the USGS upgrading of the Moroccan reserves has been linked to Peak P. In particular, we clarify that reserves are primarily company data that serve mining companies for their strategic planning and may, by no means, be used as proxy data for providing global Peak P estimates. Likewise, we elaborate that drilling plans for assessing reserves have to be adjusted to site characteristics, in particular, in the case of four plateaus in Morocco and the Western Sahara comprising an area greater than 10 000 km². We reconstruct the IFDC and USGS estimates and conclude that there is no evidence for considering the somewhat surprising increase to 50 Gt phosphate concentrate to be an unreasonable estimate for Moroccan reserves. However, the partial mixing of different units (e.g., phosphate ore and phosphate concentrate or marketable product) in the USGS data may be avoided by improving the database and using proper conversion factors. When applying these factors and assessing all reserves of marketable Gt of phosphate rock (PR-M), which is a common scale for measuring annual consumption, the magnitude of the 2014 USGS estimates of 67 Gt PR reserves does not change essentially but decreases from 64 (IFDC assessment) to 57.5 Gt PR-M (a worst-case calculation). We agree that a better harmonization of the (national) classification systems is meaningful. The discussion includes several ideas and thoughts that go beyond the paper by Edixhoven et al. (2014). We suggest that the discrepancies in the resource estimates are often caused by missing system understandings, different conceptions of sciences, and diverging world views. Finally, we suggest the establishment of a solidly funded, international standing committee that regularly analyzes global geopotential for assuring long-term supply security.