Understanding and quantification of phosphorus (P) fluxes are key requirements for predictions of future forest ecosystems changes as well as for transferring lessons learned from natural ecosystems to croplands and plantations. This review summarizes and evaluates the recent knowledge on mechanisms, magnitude, and relevance by which dissolved and colloidal inorganic and organic P forms can be translocated within or exported from forest ecosystems. Attention is paid to hydrological pathways of P losses at the soil profile and landscape scales, and the subsequent influence of P on aquatic ecosystems. New (unpublished) data from the German Priority Program 1685 “Ecosystem Nutrition: Forest Strategies for limited Phosphorus Resources” were added to provide up‐to‐date flux‐based information.Nitrogen (N) additions increase the release of water‐transportable P forms. Most P found in percolates and pore waters belongs to the so‐called dissolved organic P (DOP) fractions, rich in orthophosphate‐monoesters and also containing some orthophosphate‐diesters. Total solution P concentrations range from ca. 1 to 400 µg P L⁻¹, with large variations among forest stands. Recent sophisticated analyses revealed that large portions of the DOP in forest stream water can comprise natural nanoparticles and fine colloids which under extreme conditions may account for 40–100% of the P losses. Their translocation within preferential flow passes may be rapid, mediated by storm events. The potential total P loss through leaching into subsoils and with streams was found to be less than 50 mg P m⁻² a⁻¹, suggesting effects on ecosystems at centennial to millennium scale. All current data are based on selected snapshots only. Quantitative measurements of P fluxes in temperate forest systems are nearly absent in the literature, probably due to main research focus on the C and N cycles. Therefore, we lack complete ecosystem‐based assessments of dissolved and colloidal P fluxes within and from temperate forest systems.

Clay minerals and pedogenic metal (oxyhydr)oxides are the most reactive soil mineral constituents controlling the long-term persistence of organic carbon (OC) in terrestrial ecosystems. However, their co-occurrence in most soils complicates direct assessment of their individual contribution to OC persistence. Making use of unique mineralogical combinations in soils located in the East Usambara Mountains of Tanzania, we disentangled the contribution of clay-sized aluminous minerals (kaolinite, gibbsite) and pedogenic Fe (oxyhydr)oxides (predominant goethite and hematite) on OC storage and stabilization under natural forests and croplands. Topsoil samples, varying in contents but not types of aluminous clays and pedogenic Fe (oxyhydr)oxides, were identified by selective extractions, X-ray diffraction, and Mössbauer spectroscopy. Associated abundance of particulate and mineral-associated organic matter (OM) was quantified by density fractionation and their changes during land-use conversion were determined as a measure of OC persistence. Additionally, we assessed the resistance of OC to chemical oxidation as well as microbial decomposition in a 50-day laboratory incubation. We found that the ratio of pedogenic Fe to aluminous clay is more consequential for OC storage and stabilization than their individual contents, despite the fact that Fe (oxyhydr)oxides generally exert a stronger impact on OC than aluminous clays. Conjunction of large amounts of Fe (oxyhydr)oxides with low aluminous clay contents caused the strongest accumulation of mineral-associated OC, a low soil respiration, high OC stability against chemical oxidation, and high OC persistence during land-use change. Our study suggests that certain mineralogical combinations in the humid tropics alleviate OM losses during land conversion because of the strong and selective mineral control on OC stabilization, particular if the weight ratio of pedogenic Fe to aluminous clay exceeds the threshold range of 0.44‒0.56.