All search results were evaluated using the PRISMA statement (Page et al., 2021). In a systematic review, information was collated to answer the following two research questions: (1) how DPS and STRUBIAS production on in-situ and ex-situ treatment and processing units affect product characteristic, fertilizer replacement value (FEV) and P dynamics in soil, risks of GHG emissions and pathogen and heavy metal pollution? (2) how to combine this information to develop production systems necessary to certify DPS and STRUBIAS products as P-fertilizers in accordance with technical proposals for new fertilizing materials under the Fertilizing Products Regulation? A comprehensive systematic literature search of three online databases was performed, Scopus (www.scopus.com), PubMed (https://pubmed.ncbi.nlm.nih.gov/) and Web of Science (https://apps.webofknowledge.com). Searches were conducted in English on literature from 1983 to (1 July) 2021. All research articles related to DPS and STRUBIAS were identified. In addition, Google Scholar was used to find reports pertaining to some sections of this review. The search terms and keywords used to identify research studies were: dairy processing sludge/waste, fertilizer replacement value, phosphorus, circular economy, treatment, characterization, composition, heavy metals, greenhouse gas, methane, nitrous oxide, E. coli, and PAHs (Polycyclic Aromatic Hydrocarbons). Studies that did not contain an abstract in English were excluded from this study during the screening stage. As there were not enough published studies dealing with DPS and STRUBIAS, no meta-analysis was possible for this paper.

Duplicates were removed manually and abstracts were screened by two screeners against the target research questions. Exclusion criteria, described below, were developed and selected and cross-checking was performed on these excluded articles. If disagreements between the two screeners occurred, a third screener adjudicated. Full-text review was independently conducted by three reviewers and reasons for exclusion were annotated and tracked (e.g., “data pertaining to a different waste other than DPS or STRUBIAS”). The primary reasons for excluding papers were: (i) articles completely un-related to search questions; (ii) general knowledge papers; and (iii) papers that did not follow the basic criteria of scientific research (e.g., experimental design with sufficient replication). Articles clearly meeting the inclusion criteria were obtained for full-text review unless unavailable. These included articles related to the search inquiry, providing that scientific laboratory experiments or field studies had a minimum number of replicas and a negative control. Articles were not considered further when their title and abstract clearly indicated that the study did not meet the inclusion criteria (see Supplementary Figure 1). The studies included in tables consist of those that were reviewed in detail.

Some data needed for the holistic review were not available in the literature and needed to be developed outside of the systematic review within the present study. Greenhouse gas emission from DPS sludge and secondary STRUBIAS products was calculated using a whole systems approach. Herein, methods presented calculate CH₄, N₂O, and ammonia (NH₃) emissions from sludge and secondary STRUBIAS products from production to field application. Ammonia emission is included in the calculation due to risk of N₂O emission from NH₃ deposition to land. In this analysis, various conceptual scenarios (from production to field application) are compared with the aim to find potential scenarios that can minimize or eliminate emissions.

Assumptions for management of sludge and STRUBIAS products for a Danish dairy production site are used in the scenarios as follows: each month 1/12 of 1 ton of annual produced standard sludge is transferred to a store and the temperature in the stored sludge is similar to the monthly average air temperature of Denmark. The tanks with stored sludge are emptied at the start of April and subsequently land applied to fields in April. In the scenarios where sludge is processed to biochar or hydrochar, the same monthly amount is treated and the products are land applied in April. The annual amount of DPS treated is 1 ton.

In the calculations a 100-year global warming potential (GWP100) for CH₄ were set to 34 kg CO₂eqv kg⁻¹[CH₄] and N₂O to 298 kg CO₂eqv kg⁻¹[N₂O] (Myhre et al., 2013).

Methane emissions from stored liquid sludge is calculated with the CH₄ emission model as used in the Integrated Farm Systems Model (IFSM) for manure management in beef and dairy production systems (Chianese et al., 2009). This has been used previously to assess the impact of GHG reduction strategies in agriculture (Rotz and Hafner, 2011; Dutreuil et al., 2014). Equation 1 is used to calculate CH₄ emission from anaerobic stored livestock liquid manure and digestate from biogas plants if concentration volatile solids (VS) and air temperature is known (Baral et al., 2018):

where F_CH4 is CH₄ emission rate (g CH₄ kg⁻¹ VS day⁻¹), A is the pre-exponential factor of 31.2 g CH₄ kg VS⁻¹ h⁻¹ (Petersen et al., 2016), Ea the apparent activation energy set to 81 kJ mol⁻¹ (Elsgaard et al., 2016) giving the temperature response of CH₄ production. R is the gas constant (8.314 J mol⁻¹ K⁻¹) and T the temperature (K). The degradable fraction V_SD is 56% of VS in the sludge (Smith et al., preparation)¹. This equation and the parameters presented here is used to calculate CH₄ emission from stored DPS.

In the Danish GHG emission inventory (Nielsen et al., 2018), it is assumed that 0.5% of total N is emitted in form of N₂O from stored liquid manure and sludge, and 1% of total-N is emitted from slurry and sludge applied to soil [based on the standard emission factors given by IPCC (2019)]. Another assumption is that N₂O emissions from soil N are unaffected by any application of biochar or hydrochar applied to soil. When struvite is applied, it is assumed that 1% of the ammonium (NH4+) is emitted as N₂O as is the case for mineral fertilizer N applied to soil (IPCC, 2019).

During storage of the sludge 34% of the proteins are transformed to total ammonium nitrogen (TAN = NH₃ + NH4+) (Mottet et al., 2010) and may be emitted from the storage. Emission of NH₃ from stores with a cover of PVC roof are set to 2.6% of TAN i.e., similar to emissions from covered stored liquid manure (Hansen et al., 2008) and of sludge injected into black soil to 2% of TAN (Olesen et al., 2020). The negative charge of biochar will contribute to a negligible NH₃ emission from soils to which hydro-biochar is applied (Chu et al., 2019), biochar from pyrolysis do not contain TAN and the NH4+ in struvite is assumed not to volatilize due to the low pH (Sommer et al., 2004). At deposition on land or water, a fraction of NH₃ will be transformed to N₂O and be emitted to the atmosphere. This indirect N₂O emission is estimated using Equation 2 (IPCC, 2019):

where F_N2O is given in kg N₂O, F_NH3 is given in kg NH₃-N emitted, 0.01 is a default factor given by the IPCC for calculation of the climate warming effect of emitted NH₃, and 44/28 is to calculate from concentration given in 2^*N g mol⁻¹ to 2^*N + O g mol⁻¹ (IPCC, 2006).

During storage of untreated sludge, a fraction of C is lost in form of carbon dioxide (CO₂) and CH_4. This emission is calculated assuming that 2.6 kg VS is lost in the form of CO₂ and CH₄ for each kg of CH₄ produced. The sludge is heated in hydrothermal carbonization due to the oxidation of C, and in this process about 30% of C is lost in the form of CO₂, and about 70% of C in treated biomass is retained in the hydrochar (Kambo and Dutta, 2015). In the pyrolysis process, more CO₂ is lost and between 25 and 35% (avg. 30%) of solids (40% C) are retained in the biochar (Kambo and Dutta, 2015).

In the calculations used herein, the C retention of field applied sludge-C was set to 25%, which is between 12% of manure C input in the longer term (avg. 18 years; Maillard and Angers, 2014) and 35% for a 20-year period by scaling results from a study of transformation of C in digestate applied to soil (Thomsen et al., 2013). The longer retention time of C in sludge than in animal slurry was due to sludge organic matter (OM) from a wastewater treatment plant that has been transformed to a more stable form of C. It is calculated that C concentration in VS is 517 g C kg VS⁻¹, and this estimate is used to calculate C concentration in sludge, where OM is measured as VS.

It has been calculated that between 90 and 97% of the C in biochar from pyrolysis of a range of different biomasses will still remain in the soil after 100 years (Lehmann and Joseph, 2015; Wang et al., 2016). No study of the recalcitrance of biochar from pyrolysis of dairy sludge was found in the literature, and a conservative/cautious estimate is that 90% of the C in biochar from pyrolysis (PC) of dairy sludge is recalcitrant.

Hydrochar from hydrothermal carbonization (HTC) is produced at a lower temperature than when producing biochar by pyrolysis, and C component in the hydrochar has been oxidized less than pyrolysis products, i.e., less oxygen is added during the process. Our assumption is that C in hydrochar is less recalcitrant than biochar produced by pyrolysing dried biomass, and that only 50% C in hydrochar is recalcitrant. This assumption is supported by the study of Malghani et al. (2013), who stated that “although both HTC and PC chars were produced from the same feedstock, PC chars had markedly higher potential for carbon sequestration than HTC.”

The dairy processing industry generates a large volume of waste, which is high in OM. Discharge licensing, which is site specific, aims to prevent a reduction in surface water oxygen (O₂) concentrations and the onset of surface water eutrophication (Neal and Heathwaite, 2005). To achieve such discharge thresholds, a chain of treatment before discharge at dairy processing plants is needed (Figure 2). In fact, most DPS applied to land in the EU is first treated, composted or incinerated before being applied to land (Figure 2). Research now focusses on converting DPS into secondary STRUBIAS products (Figure 2).

A schematic of typical dairy wastewater treatment processes and DPS generation is shown in Figure 2. Pre-treatment removes particulate matter and retains fats that could interfere with subsequent treatment. The pre-treatments and physio-chemical treatments most frequently used on dairy processing plants are buffer tanks, neutralization, sieving, flotation and degreasing (Droste and Gehr, 2018). Flocculation (coagulation) is the most simple and economical pre-treatment method, and reduces water turbidity by reducing particulate substances and fats, which could interfere with subsequent treatments (Carvalho et al., 2013). Pre-treatment by flocculation enhanced by the presence of lactic acid bacteria (Lactobacillus plantarum), which ferment the lactose and produce lactic acid. The acidity of this compound precipitates the milk proteins, and thus significantly reduces the chemical oxygen demand (COD). This reduction increases with addition of flocculants such as chitosan or carboxymethyl cellulose (CMC) to this solution. The COD removed varied by 65–78% for CMC and 49–82% for chitosan (Dyrset et al., 1999).

In the secondary treatment, sludge undergoes aerobic treatment that is cost efficient and controllable (Kolev Slavov, 2017). This treatment step can reduce more than 90% of COD (Carta-Escobar et al., 2004; Kushwaha et al., 2011; Carvalho et al., 2013). In aeration ponds, the aerobic pathway uses microorganisms contained in biological reactors, which in the presence of O₂, degrade the OM suspended in the sludge.

The soluble organic components transformed and emitted mainly in the form of CO₂ and NH₃, while the insoluble pollution fraction, including the microorganisms, is recovered by separation and form “activated” sludge (Sustarsic, 2009). A large number of aerobic treatment types are being used e.g., rotating biological contractors (RBCs), sequencing batch reactors (SBRs), or membrane reactors (MBRs) (Goli et al., 2019). In particular, SBRs are effective at varying loading capacities (Kolev Slavov, 2017). In this step, a single tank used for filling, aeration, settlement and effluent withdrawal, and recycling of solids. The sludge produced during primary and secondary treatment may be treated anaerobically in biogas reactors. During this process, OM is transformed by microorganisms to CH₄ and CO₂. This process produces small volumes of sludge (<0.05 kg of dry matter (DM) per kg of COD eliminated) (Omil et al., 2003). The most common anaerobic reactors are up-flow anaerobic sludge blanket (UASB), completely stirred tank reactors (CSTRs) and membrane anaerobic reactor systems (MARSs). Finally, where needed, the N, P, micropollutants or pathogenic microorganisms in the wastewater can be further reduced using a finishing treatment e.g., for P content reduction alternation between aerobic/anaerobic conditions or P coagulation and precipitation by ferric chloride or aluminum sulfate can be deployed (Rivas et al., 2010). After these treatments, the particles settle in a clarification pond. The supernatant is discharged, while the precipitated sludge is dewatered after addition of polymers to promote flocculation and separation between water and suspended matter (dewatering). This is then thickened (thickening) during its passage through dewatering grids. This sludge is stored in tanks or treated further and then used as a bio-based fertilizer (Huygens et al., 2019).

The volume of wastewater produced at processing plants can be high and is product-dependent. For example, 1 liter of processed milk can produce up to 10 L of effluent (Lateef et al., 2013). In the cheese manufacturing industry, the whey is the main pollutant discharged to water and soil. This is mainly due to its high carbohydrate content (4–5%) of which lactose is the main constituent (Kolev Slavov, 2017). Consequently, DPS produced from cheese production, called Cheese Whey Wastewater (CWW), contains high concentrations of organic components contributing to a high COD and biological oxygen demand (BOD) (Ahmad et al., 2019). The COD of CWW is higher (0.79-77.3 g L⁻¹) than COD of milk plant effluent (0.183–10 g L⁻¹) (Carvalho et al., 2013) or milk-treated condensate wastewater (<0.001 g L⁻¹), which is the water obtained during the concentration and evaporation processes of milk and its by-products (Bourbon and Huet, 2018). The CWW contains some milk or milk by-products, oils and greases, and cleaning water containing sterilizing agents, acid and alkaline detergent (Carvalho et al., 2013; Ahmad et al., 2019). As a result, the concentration of inorganic compound may be heterogeneous, with ranges observed for P from 8 to 510 mg L⁻¹ and N from 14 to 1,462 mg L⁻¹ (Demirel et al., 2005).

The amount and composition of DPS is determined by the composition of milk and use of additives, the dairy production line and cleaning and disinfectant products used for cleaning. In addition, wastewater treatment, varies between plants (Figure 2), and depends on the type of dairy processing involved (Rico Gutiérrez et al., 1991; Karadag et al., 2015; Ahmad et al., 2019; Shi et al., 2021a). Recently, physical and chemical characteristics of 63 DPS samples (9 dairy processing plants in Ireland) were reported (Ashekuzzaman et al., 2019a). The main DPS types included in the study were bio-chemically treated activated sludge leading to aluminum-precipitated sludge (Al-DPS) and iron-precipitated sludge (Fe-DPS) depending the dosing of alum or ferric salt to remove P, and to lime-stabilized calcium-precipitated sludge (Ca-DPS) generated after dissolved air floatation (DAF). In some processing plants, mixing of DPS generated from aerobic biological wastewater treatment and DAF processes occurs before land application or further disposal processing. A few of the examined plants have anaerobic digesters (AD) that produced AD sludge. That concentration of components varies between sludge categories and treatments and differences in composition was highest for N, P and K (Ashekuzzaman et al., 2019a). This difference is higher between activated and DAF sludge types than those in the combined sludge (sludge mixed from both activated and DAF process). The addition of chemicals such as Fe-, Al-, or Ca-based coagulants during the removal of P from the effluent may explain these results. In addition, the N concentration was lower in DAF sludge compared to AD sludge in contrast to the increased P concentration (Ashekuzzaman et al., 2019a). The addition of lime during the DAF process explains this result by an increase of the pH, which promotes the volatilization of N in the form of NH₃. On the other hand, anaerobic degradation during the AD process increases the TAN concentration in the sludge due to organic N mineralization.

Tables 2, 3 illustrate the physicochemical characteristics of dairy processing effluents and DPS reported in the literature. These characteristics highlight the variability in dairy effluent compositions related to the type of bio-products processed (i.e., milk, cheese, yogurt, and butter) (Omil et al., 2003; Carvalho et al., 2013; Karadag et al., 2015), and the wastewater treatment processes used (Britz et al., 2006; Ashekuzzaman et al., 2019a).

In France, information and data pertaining to the agronomic benefits and risks of applying organic waste to agricultural soils was collated by Houot et al. (2014) and used to re-evaluate the EU Sewage Sludge Directive (86/278) (CEC, 1986). This work emphasized the need for sludge used in agriculture to have P recycling as a main priority, and that this use must not be a risk to the environment or to human health due to their contents of heavy metals, organic trace compounds, pathogenic microorganisms and pharmaceutical compounds. To avoid some of these concerns, the EU Council Directive 86/278/EEC set limits for the content of heavy metals (Cd, Cu, Hg, Ni, Pb, and Zn) (CEC, 1986), and individual European countries have set limits for synthetic organic compounds and pathogens (Hudcová et al., 2019).

In a comprehensive study across nine Irish dairy plants, the concentration of heavy metals (i.e., Cr, Cu, Ni, Pb, and Zn) was examined in all major DPS types with lowest concentrations found in DAF sludge and highest in AD sludge (Ashekuzzaman et al., 2019a). Overall, the heavy metal concentrations across all tested DPS samples were significantly lower than limits set by the EU for avoiding accumulation in agricultural soil to which sludge is applied (CEC, 2008) and the levels were below those of livestock manure (Sommer et al., 2013), and composts (Bernal et al., 2017). The results of the Irish study are in line with the current knowledge on heavy metals content of DPS (Table 4) and indicates that heavy metal concentrations will not be a limiting factor for legal and safe application rate of DPS to agricultural soils (Ashekuzzaman et al., 2019a; Shi et al., 2021b). The concentration varies between dairies and this is due to the diversity of the milk bio-products and the various possible steps in the treatment of the effluent. It is important to have knowledge pertaining to the heavy metal content of DPS and DPS-derived STRUBIAS products before land application, because farmers and society must be assured that the heavy metal content is lower (in soil and plants) than the limits given for use before making final decisions and rules of use of the waste as a fertilizer (Shi et al., 2021b).

Organic Trace Compounds (OTCs) are chemical products (hydrocarbons and their derivatives, degradation products, solvents, etc.) present in organic waste or derived due to degradation of the organic compounds by the microorganisms in sewage treatment plants or in the soil. They often accumulate by biomagnification and bioaccumulation in biological organisms and cause irreversible damage to biological systems. They are directly or indirectly toxic to humans and animals (such as endocrine disruption and tumor initiation) (Barret et al., 2012). Table 5 presents the European Commission limit values for organic contaminants (CEC, 2009).

In Europe, there are proposals for limits on polycyclic aromatic hydrocarbons (PAHs) contents in municipal sewage sludge applied to land (CEC, 2010). Depending on the country-specific regulations, the type of OTCs and the threshold limits differ. Where organic waste is land applied, the following three PAHs i.e., fluoranthene 5, benzo(b)fluoranthene 2.5, benzo(a)pyrene 2 and 7 PolyChloroBiphenyls (PCBs) (PCB 28, 52, 101, 118, 138, 153, 180) are considered good indicators of compound resistant to biodegradation, and according to French regulation must be below concentration limits (Hudcová et al., 2019). In contrast, German authorities do not regulate PAHs, while the threshold for the sum of nine PAHs is more stringent in Denmark than for the sum of three PAHs in France. Companies that handle wastes must be aware that the concentration of 16 PAHs depends on the type of sludge (Boruszko, 2017). Research has shown that anaerobic fermentation and post-flotation may reduce the content of PAHs up to seven times its original concentration (e.g., 689 μg kg⁻¹ of DM) (Boruszko, 2017), with most of the reduction of hydrocarbons taking place in the final phase of fermentation. The concentrations of PAHs in DPS are low and do not exceed the amount allowed by the European Commission (CEC, 2009; Table 5). Therefore, their use in agriculture will not be limited by PAHs (Pérez et al., 2001; Boruszko, 2017), but it is recommended that these are investigated in DPS-derived STRUBIAS products.

DPS contains living microorganisms originating from the treated wastewater. There are pathogens (viruses, bacteria, fungi and parasites) derived from animal manure (Sobsey et al., 2006), and if dairy cows ingest grass from fields where dairy manure has been applied, then there is a risk for transfer of pathogens in infectious levels to the cows and subsequently to milk and therefore to wastes. The European Commission (CEC, 2000) has in its third draft of the Working Document on Sludge, proposed the following thresholds for a range of bacteria and worms for sludge to be recycled to soil: (1) E. coli <5 × 10⁵ colony forming units (CFU) per gram (wet weight) of conventional treated sludge; (2) for advanced sanitized sludge E. coli must be below 1 × 10³ CFU g⁻¹ wet colony of treated sludge; (3) Salmonella Senftenberg W775 in sludge spiked with this microorganism must be reduced 99.99%; (4) no content of Ascaris ova (5) a sample of 1 g DM of the treated sludge must not contain more than 3 × 10³ spores of Clostridium perfringens, and (6) a sample of 50 g (wet weight) of the treated sludge must not contain Salmonella spp.

There are few studies reported about reduction of pathogens in raw DPS. Laboratory studies have quantified reductions of microbial infectivity (inactivation) in animal organic wastes under controlled temperature conditions and the samples have been stored aerobically or anaerobically (Sobsey et al., 2006). For example, the high initial level of pathogens (Enterobacteriacea, fecal coli forms and E. coli) in dairy slurry was higher than in sludge from two urban wastewater treatment plants where anaerobic digestion was followed by mechanical dehydration (one treatment also received a heat-dried process). The concentrations in treated dairy slurry were 5.1 × 10⁷ CFU g⁻¹ of DM of Enterobacteriacea; 4.4 × 10⁷ CFU g⁻¹ of DM of fecal coli forms, and 4 × 10⁶ CFU g⁻¹ of DM of E. coli and all are below the limits set by the regulation. Concerning the persistence of these pathogens in the soil, then after a 80-day trial across soil/sludge treatments, the populations of fecal coliforms and E. coli decreased considerably or were not detectable (Estrada et al., 2004). The study of Ravva et al. (2006) also found that the pathogenic strain E. coli O157:H7 introduced in water from on farm dairy waste lagoons, failed to establish and proliferate in dairy wastewater microcosms with or without circulating aerators. On the other hand, high concentrations of Listeria are found in manure and sewage sludge and have survived in topsoil between 12 and 182 days (Sobsey et al., 2006). If these are present in DPS, then additional sludge treatments such as anaerobic digestion, hygienization by adding lime, or composting will reduce the concentration of pathogens to allowable values.

Dairy processing wastes (even after treatment at source) may contain harmful substances, which need testing and quantification across all the DPS and STRUBIAS types. The substances in focus should be antimicrobial drugs, hormones, pesticides, emerging contaminants, pathogens, disinfectants, persistent organic pollutant residues, microplastics, and nanoparticles in DPS or DPS-derived STRUBIAS (Shi et al., 2021a).

Another strategy that is being deployed to manage DPS is to further process these raw products into other more usable and stable forms. Struvite (magnesium ammonium phosphate hexahydrate, MgNH₄PO₄, 6H₂O) is widely used in agriculture due to its N and P content, which is in a form that efficiently can be used by plants (Adam et al., 2009). Phosphorus can exist as particulate and dissolved species in both organic and inorganic forms. The inorganic P species is mainly in orthophosphate form, which is plant available and important for soil fertility but can be readily lost to the environment (Frossard et al., 1996). However, the chemical composition of struvite obtained from DPS is not always consistent with pure struvite equivalents (Hall et al., 2020). Metal impurities such as Al, Fe, Ca, and small amounts of heavy metals can precipitate along with the struvite and could pose problems later for crops and soil when land applied.

The term “char-based materials” is used here to replace “biochar” in the STRUBIAS acronym as they have different terms depending on the technology. Char-based materials, obtained from the thermochemical conversion of biomass in an oxygen-depleted atmosphere, are porous and carbonaceous, and are more stable and C-rich and less toxic than the feedstock (Kambo and Dutta, 2015; Atallah et al., 2020). The significance of thermochemical treatment lies in overcoming the structural inferiority of biomass, which enhances the chances of energy and resource recovery from waste (Kambo and Dutta, 2015). There are many functions of char-based materials including, but not limited to, energy production, agriculture, C sequestration, wastewater treatment, and bio-refinery (Kambo and Dutta, 2015). The utility of a specific char-based material for any particular application depends on its inherent properties. Feedstock, pre-treatment method, and temperature are all important (Amoah-Antwi et al., 2020). However, thermochemical treatments increase the risk of producing chars with other highly toxic compounds produced from high-temperature reactions such as PAHs, PCBs, dioxins, furans, and PCDD/Fs (Kambo and Dutta, 2015; Amoah-Antwi et al., 2020). Heavy metals present in the feedstock are most likely to remain and concentrate in the chars (Shackley et al., 2010).

Ashes are characterized as fly ash or bottom ash, or a combination formed through the incineration of biowastes by oxidation (Huygens et al., 2019). Ash normally contains valuable plant macronutrients such as K, P, S, Ca, and Mg (Haraldsen et al., 2011; Knapp and Insam, 2011; Brod et al., 2012). In addition, they contain large amounts of P (13.7–25.7% P₂O₅), which are comparable to commercial superphosphate (Xu et al., 2012). Obstacles to the use of ash as a fertilizer or soil amendment could be their heavy metal contents (Franz, 2008; Herzel et al., 2016).

DPS and DPS-derived STRUBIAS products are used or research is underway to ascertain their potential as bio-based fertilizers in agriculture (Shi et al., 2021a). DPS is typically stored off site until applied to land in spring, whereas STRUBIAS products can be processed and stored until needed. Many knowledge gaps still exist pertaining to their respective effects on P dynamics once spread onto soils and their FEV. It should be noted that FEV is used herein (Shi et al., 2021a), but can be often known in the literature as Mineral Fertilizer Equivalence (MFE; Delin, 2012) or Mineral Fertilizer Replacement Value (MFRV; Schröder et al., 2007).

In cropped agricultural systems, P applied to soil with fertilizers can be utilized by crops, absorbed by soil minerals (González Jiménez et al., 2019), or lost along surface (runoff) or subsurface (leaching and loss along natural or artificial lateral transport or deeper recharge to groundwater) pathways to surface waters (Murnane et al., 2016). Soil P transformation passes through several interconnected pools. These are the soluble P pool, which is considered to be immediately available for plants; labile or weakly adsorbed P, insoluble P chemically bound with Ca ions in calcareous and alkaline soils or occluded by Fe and Al oxides in acidic soils, P strongly adsorbed by hydrous oxides of Fe and Al, and insoluble organic P within soil organic matter (Stevenson and Cole, 1999; Bennett and Carpenter, 2002). Figure 3 presents a simplified diagram reflecting P soil cycling and interactions between these pools. Briefly, the soil P cycle consists of the following processes: weathering and precipitation, mineralization and immobilization, adsorption and desorption, and P losses through surface or near surface runoff and subsurface leaching with eventual recharge to groundwater (the proportions of which are dependent on soil/subsoil/bedrock permeability and chemistry).

Mineralization and immobilization of P are part of the organic P cycle. Mineralization is a process of transformation of organic P to soluble H₂PO4- or HPO42-. Mineralization of organic P slowly releases soluble P, which is crucial during the growing season as it provides a continuous supply of P to crops. Mineralization of organic P in soil occurs through breakdown of organic bonds, which is driven by the release of enzymes produced by plants and soil microflora. Organic P mineralization is driven by phosphatase enzyme activity in soil, which mainly occurs during the growing season when soil temperature ranges between 18 and 40°C (Prasad et al., 2016). Phosphatases synthesis are believed to be driven by P availability and enhances under P limiting conditions (Luo et al., 2017). However, in some cases application of mineral P (Paredes et al., 2011), and organic amendments of soil (Parham et al., 2002) can increase phosphatases activity. Some other factors which have an impact on phosphatases production are P availability and availability of other soil nutrients (Marklein and Houlton, 2012), soil moisture, pH, and availability of other soil nutrients, and energy supply (Acosta-Martínez and Waldrip, 2014; Prasad et al., 2016).

Precipitation and dissolution and desorption and absorption are part of inorganic P cycle. The direction of P transfer between inorganic P soil pools though precipitation and dissolution can be either reversible or irreversible, and can be impacted by a number of factors, including geochemical soil composition and soil pH. For instance, in acidic soils P precipitation occurs in the presence of Fe, Al, and Mg, and soluble P in such soil can be limited, while in alkaline soils precipitation primarily occurs through reactions involving Ca²⁺ compounds (Prasad et al., 2014). Adsorption, or fixation, binds soluble P compounds to soil particles, whereas desorption releases P which is bound with soil minerals to soil solution, thereby increasing the soluble P pool. Unlike precipitation, this process is reversible, and P does not involve permanent change in chemical and structural changes in P-containing compounds and soil minerals.

The consideration of the aforementioned P fluxes in agricultural soil and recycled DPS composition is essential for developing guidelines of alternative P fertilizers. Specifically products derived from chemically treated dairy effluents treated with lime, ferric sulfate or aluminum chloride may contain elements, which can limit P release into available P pool such as Ca, Fe, and Al (Ashekuzzaman et al., 2019a). An inherent soil pH range optimal for P fertilizers to remain in soluble pool is between 6 and 7.5. Decreasing soil pH can lead to soluble P fixation by Fe and Al oxides. While fixation can be a limiting factor for soil P availability for crops (Daly et al., 2015; Prasad et al., 2016), fixation of P by minerals present in the soil is also a limiting factor. To ensure sustainable use of the P source and avoid P losses into the environment, such best practice should be followed (Science Communication Unit University of the West of England Bristol, 2013; Arenas-Montaño et al., 2021).

The FEV defined as the equivalent application rate of an inorganic fertilizer achieved by an organic waste to achieve the same crop yield or nutrient uptake (Brod et al., 2012). The efficiency of most bio-based organic fertilizers is lower than inorganic fertilizers because of their slow nutrient release rates (Chen, 2006). The FEV of an organic fertilizer can both provide a quantitative estimate of the amount of efficient nutrients in bio-based fertilizer and estimate of the actual value when compared with a chemical equivalent. This information, which is currently lacking, would give growers accurate information to help with nutrient management planning on farms.

Two methods used to assess the FEV of bio-based fertilizers such as DPS or DPS-derived STRUBIAS products, are pot or field-scale studies, which include different fertilizer rates, crops, and soils. The most common method is to compare yields or nutrient uptake results from DPS or DPS-derived STRUBIAS treatments with uptake from commercial mineral fertilizers as used with other organic fertilizers e.g., Lalor et al. (2011) examined the FEV of dairy cattle slurry. Typically, data fitted to linear, quadratic, or cubic polynomial regressions, creates a relationship equation. For example, Figure 4 illustrates a fitted polynomial function, describing crop yield or nutrient uptake corresponding to different mineral fertilizer application rates. This is the method used to determine the corresponding mineral fertilizer rate (×1) to any crop yield or nutrient uptake by a bio-based application. The mineral fertilizer rate, ×1, expressed as a percentage of total nutrient applied from that bio-based treatment and estimates the FEV. Alternatively, calculation of FEV by the apparent nutrient recovery method without the need of a response curve is used. There is, however, a difference between apparent N or P recovery (ANR or APR) and N-P FEV. The first is the N or P fraction taken up by the test crop of total applied nutrients and the second is the ratio of the apparent N and P recovery of bio-based fertilizer and that of mineral fertilizer at the same rate (Cavalli et al., 2016; Sigurnjak et al., 2019). They are determined as follows using Equations 3–6:

The most comprehensive grassland study on the FEV of DPS, conducted by Ashekuzzaman et al. (2021a,b), examined two main types of DPS. The first is aluminum or iron-precipitated activated sludge (Al- or Fe-DPS) and the second is a lime-stabilized calcium-precipitated sludge (Ca-DPS). At field scale, an assessment of N and P availability for crop yield and uptake in comparison to reference mineral fertilizers over one seasonal year was undertaken. Ashekuzzaman et al. (2021b) found N-FEV of 22–25, 54, and 8%, respectively, for Ca-DPS, Fe-DPS, and Al-DPS. They indicated that N-FEV varied between activated and lime treated DPS types, as affected by wastewater and sludge treatment processes and storage. The different treatments affect the proportion of mineral and organic N in the DPS, and thus the available N pool in amended soil. With regards to P availability, the results of Ashekuzzaman et al. (2021a) show that first-year cumulative P availability (over the four harvests) differs significantly between Al- and Ca-DPS where Al-DPS P-FEV was 109% compared to mineral P (applied at 40 kg P ha⁻¹) and Ca-DPS P-FEV was only 31%. Their findings show that mineral P fertilizer was a better starter fertilizer that at application provided more readily available P for plant uptake than either Al-DPS or Ca-DPS, as they observed 50 and 16% P-FEV for the two DPS, respectively, in the first harvest. Although the Al concentration (1,122 mmol kg⁻¹) in Al-DPS did not limit first-year P bioavailability, the initial nature of P fractions, and their biological and bio-chemical mineralisation processes, might be the reason of lower P availability for immediate uptake by plant. For Ca-DPS, high Ca content (Ca/P molar ratio 1.86) and alkaline pH in Ca-DPS was likely to be associated with formation of low soluble Ca-P compounds and low P availability. Future studies on the aspect of P composition and mineralisation process in DPS would help to realize and correlate P uptake efficiency.

Other literature pertaining to the FEV of DPS, and especially DPS-derived STRUBIAS products is still limited. Many factors affect the calculation of FEV such as the treatment processes used to produce a DPS or DPS STUBIAS type (crop type, fertilizer application rate, duration of experiment and scale of experiment (pot vs. field) (Brod et al., 2012; Cerný et al., 2012).

The dosing of Al or Fe salts used to capture P in sewage wastewater treatment plant affect P-FEV, because high concentrations (more than 2,800 mmol kg⁻¹) of either or both may significantly reduce P bioavailability (Khiari et al., 2020). The P fertilizer effects of 14 different bio-based fertilizers had been tested in pot experiments with ryegrass (Delin, 2016). At the first cut, the P-FEV of Fe- and Al-precipitated sewage sludge were 37 and 33%, respectively (Delin, 2016). Falk Øgaard and Brod (2016) found that P-FEV of 11 sewage sludges treated with Al and/or Fe salts varied significantly between sludges, but was low for all sludges in a pot experiment with ryegrass. It was lowest at first cut, where it ranged from 2 to 24% (Falk Øgaard and Brod, 2016). Both studies indicated that sludge derived from a treatment with Fe had a higher P-FEV than when coagulation with Al salts occurred. This is due to a higher solubility of Fe phosphate compared to Al phosphate. The amount of Fe used is also important and sludge with a Fe:P ratio at 1:6 contains more plant available P than sludge with a higher ratio (e.g., Fe:P ratio of 9:8) (Kahiluoto et al., 2015). Calcium is another element that has an effect on the P availability. High dosing of Ca in the wastewater with a Ca:P ratio of 2:1 reduce the P-FEV due to formation of Ca-P compounds such as hydroxylapatite, which has a low solubility, an effect shown when using the sludge to produce compost and biochar (Nest et al., 2021). As mentioned above, solubility of the P crystals are affected by pH and liming increases the plant-available P in sludge produced from the wastewater treated by Al or Fe salts (Krogstad et al., 2005; Montgomery et al., 2005; Bøen and Haraldsen, 2013).

When processing DPS to a STRUBIAS product, the FEV will change. This is influenced by the untreated DPS physiochemical characteristics and the processing methods and parameters used. For example, the P-FEV in ash produced from incineration is low when wood is used (P-FEV, 8–16%), but gets higher using chicken manure (P-FEV = 13–39%) (Yusiharni et al., 2007), and highest when incineration of biogas residue is used (MFE = 76–99%; Kuligowski et al., 2010). The plant availability of P in thermochemical products such as ash and biochar depends on the temperature during combustion/pyrolysis, and is halved by increasing the incineration temperature from 400 to 700°C, which is due to hydroxyapatite formation (Thygesen et al., 2011). To increase the amount of P in STRUBIAS products, the use of flocculants or biological processes to increase P availability in raw DPS whilst avoiding high temperatures during the production of STRUBIAS could be implemented.

For N-FEV, the proportion of ammonium N (NH4+-N) to the total N content and the C/N ratio of the DPS or DPS-derived STRUBIAS products are the most important factors for the FEV (Sommer et al., 2013; Webb et al., 2013). Ammonium is immediately available for the crop and is often a growth-limiting factor (Brod et al., 2012; Gómez-Muñoz et al., 2017). When fertilizer rates increased, the N-FEV of meat and bone meal (MBM) and composted fish sludge (CFS) decreased from 76–65% to 67–53%, respectively (Brod et al., 2012; Gómez-Muñoz et al., 2017). This is consistent with crop response trials, where increasing amounts of N are applied (e.g., Brod et al., 2012, see Figure 4). In that example, two industrial composts (i.e., neutral and acid Dynea composts) had only N-FEV values ranging from 7 to 30%, as they contained low amounts of NH4+ and the N-mineralization rate was low. Acidification of the compost increased the ANR and N-FEV of Dynea composts compared to untreated compost (Brod et al., 2012). This is due to a reduced NH₃ emission during composting, as is seen when acidifying stored pig and cattle slurry. In long-term studies adding human sewage sludge (i.e., biosolids) to silage maize, the FEV was 55% at low application rates and 64% at high application rates (Cerný et al., 2012). This result, compared with the finding by Brod et al. (2012), implied that in short-term fertilizer application, doubling the application rate might not have a higher ANR and FEV, but in the long-term application, an increase of the rate increase ANR and FEV. The reason can be that higher application rate in short-term studies leads to emission of easily available N (NH₃ emission, denitrification) and that reduces FEV, while in long-term studies there still is this immediate loss but organic N increases in the soil and this will lead to higher amounts of N mineralised with time. Gómez-Muñoz et al. (2017) found in the long-term experiment that continuous application of agricultural and urban wastes improved soil quality, and long-term N availability correlates with the accumulation of N and C in soil. That study reported the ANR and FEV in the final year (2013) had generally increased compared to those in the first year of the study (2003), except for composted household waste and cattle deep litter. The effect of C:N ratio was documented for biochar produced by the pyrolysis of eucalyptus wood, as ANR values increased from 28–40% with increasing C:N ratios (2–4.9). Therefore, as new DPS-derived STRUBIAS products are emerging, there needs to be a test phase before their use in agriculture. This should involve short to long-term pot and field trials across crop and soil types to investigate P dynamics in soil and their N-P FEV values.

During the storage and land application of DPS, there may be the risk of nutrient loss or emissions to waters (surface and subsurface) and/or the atmosphere, respectively.

Soil microorganisms play a pivotal role in nutrient cycling in agricultural ecosystems. Their activities enhance the availability of essential nutrients for crop growth and contributes to an improvement of soil properties such as OM content and water retention capacity. The application of organic residues like DPS modifies soil microbial communities significantly. These modifications are highly variable and depend strongly on the composition of the bio-based residues applied.