Changes in organic fractions, cations, and stabilization from feces to fecal sludge: implications for dewatering performance and management solutions

Reliable dewatering performance remains a key challenge in fecal sludge management, and the controlling factors or mechanisms are not well understood. There remain limited studies on constituents in feces and fecal sludge and how they affect the dewaterability of fecal sludge. This study aimed at evaluating a range of constituents in feces, and to gain empirical knowledge toward a mechanistic understanding of how they in ﬂ uence dewaterability. In this study, cellulose reduced capillary suction time, decreased supernatant turbidity, and increased cake solids. While hemicellulose decreased supernatant turbidity, lignin increased supernatant turbidity, capillary suction time, and cake solids. Extracellular polymeric substances (EPS) increased both capillary suction time and supernatant turbidity and decreased cake solids, whereas lipids increased turbidity. Cations had no signi ﬁ cant effect on dewatering properties. Overall, fecal sludge stored in contain-ments had better dewatering performance than ‘ fresh ’ fecal sludge, which was attributed to stabilization. Field fecal sludge had a higher relative abundance of Pseudomonas , which is associated with better aggregation, and fewer small particles ( , 10 μ m) that clog ﬁ lters to reduce dewatering performance. Further understanding of stabilization and developing an agreed-upon metrics of stabilization are essential for predicting fecal sludge dewatering performance, and developing smaller footprint dewatering treatment technologies.


INTRODUCTION
Globally, 31% of sanitation needs in urban areas are met through non-sewered sanitation (WHO 2021).Fecal sludge (FS) is defined as what accumulates during storage in onsite containments, and can include feces, urine, flush water, greywater (e.g., kitchen and bathing), food waste, and rubbish (Strande et al. 2014).In urban areas of low-income countries, at least 60% of FS is not safely managed, placing a huge burden on public and environmental health (Peal et al. 2020).Typically, FS consists of more than 95% water (Gold et al. 2017), making it expensive and difficult to transport.Dewatering has proven to be inconsistent and unpredictable, making effective dewatering one of the greatest knowledge gaps for the sustainable management of FS (Ward et al. 2019), and a barrier for implementing low footprint technologies (Mercer et al. 2021a).Unlike wastewater sludges, dewaterability of fresh feces, FS, nor FS stored in onsite containments is well understood.FS is different from waste-activated or anaerobically digested sludges, and has variable solids, organic, and nutrient concentrations, which can be up to two orders of magnitude greater than municipal wastewater (Ward et al. 2019).We have started to develop an understanding of the role of properties like extracellular polymeric substances (EPS) (Sam et al. 2022), particle size distribution (PSD) (Ward et al. 2023), and cations (Ward et al. 2019) in dewaterability of FS, but on their own they do not yet explain observed dewatering performances.
Feces are a component of FS, but there remains a general lack of knowledge of constituents and the role they play in the dewatering behavior of FS.Examples of existing knowledge include mass balance reports on the chemical composition of feces, mainly conducted by NASA during the early stages of a space program (Goldblith & Wick 1961), and studies on gut microbiome that started around the turn of the 21st century (Hopkins et al. 2002).Other examples include soluble material and fiber content in feces, in relation to their effect on fecal weight and transit time in the intestines (Stephen & Cummings 1980), and proteins and lipids in relation to inflammatory bowel disease and steatorrhea (Wenzl et al. 1995;Hidaka et al. 2000).Research initiated by 'reinvent the toilet challenge (RTTC)', a program initiated by the Bill and Melinda Gates Foundation for 'back end' treatment at the source of production (no transport) with mainly thermal or chemical treatment at source (Bhagwan et al. 2019), has also investigated the rheological properties, energy recovery, microwave treatment technologies, and hydrothermal liquefaction of human feces (Watson et al. 2017;Mercer et al. 2021a).A review by Rose et al. (2015) summarizes the reported ranges of microbial biomass, unabsorbed macromolecules (proteins, polysaccharides, and lipids), and trace constituents such as secretions and inorganic fractions.Gold et al. (2017) and Krueger et al. (2021) also analyzed the fiber content of feces in relation to fecal bioconversion by black soldier fly larvae and thermal decomposition.Penn et al. (2018) attempted to develop synthetic feces and FS recipes for research purposes.However, the properties related to dewaterability could not be replicated and the synthetic FS recipe had 60% reduced dewaterability in comparison to FS. Diet is often considered to have an influence on the total fecal output or generation rate (Rose et al. 2015); however, the role of fecal composition in terms of organic fractions and microbial communities, and if that translates into FS characteristics or dewatering properties, is not clear.
Research in centralized wastewater treatment spans over 100 years (Stensel & Makinia 2014), and based on that we know that EPS (proteins, polysaccharides), lipids, fibers, cations, and microorganisms play an important role in dewatering performance.EPS are highly charged and incorporate up to 99% water in a cross-linked hydrogel through electrostatic interactions, hydrogen bonding, or Van de Waal forces (Pfaff et al. 2021), and promote floc formation by bridging particles together, which enhances settling performance (Christensen et al. 2015).Lipids adsorb onto bacterial biomass and decrease the specific gravity and settling of sludge particles (Chipasa & Medrzycka 2006).Fibers (cellulose, hemicellulose, and lignin) increase water-holding capacity due to peripheral hydroxyl groups of glucose that interact with water through hydrogen bonding.Divalent cations form strong and compact flocs in activated sludge through their interaction with extracellular polymers, whereas monovalent cations have a detrimental effect on dewatering (Christensen et al. 2015).Due to the importance of these constituents in governing the dewatering performance of wastewater sludges, they were selected for investigation of the role they play in governing dewaterability of fresh FS and field FS.Fresh FS is defined here as FS containing feces, urine, flushwater, and cleansing water, that has not yet been stored in a containment.
The objectives of this study were to (1) evaluate the concentrations of constituents in feces that have been identified as governing dewaterability in wastewater sludges (i.e., EPS, proteins, polysaccharides, lipids, fibers, cations); (2) gain empirical knowledge toward a mechanistic understanding of how these constituents affect dewatering properties of FS; (3) compare these constituents and dewaterability of fresh FS to FS collected from onsite containments in eight different countries; and (4) to evaluate the role of stabilization from feces to FS in dewaterability.In this study, dewaterability is defined as the process of separating solid matter and liquid fractions, employing well-established metrics for filtration quantified by capillary suction time (CST), settling measured by supernatant turbidity following centrifugation, and water-holding capacity as measured by the dewatered cake solids.

MATERIALS AND METHODS
Source of feces, fresh FS, and field FS Feces were collected from 20 anonymous volunteers (from 11 different countries) at Eawag in Dübendorf, Switzerland, using a double-blind systemso no information could be traced back to individuals.Participants completed a questionnaire including diet and use of medications for a 24-h period, prior to sample collection.The questionnaire is provided in Supplementary material and the responses are in the published data package.Feces were stored at À20 °C and brought to room temperature before analysis was performed.Ten samples of FS from containments (field FS) were obtained from eight countries: Ghana, Senegal, Uganda, Kenya, Lebanon, India, Canada, and Guatemala.Samples from Ghana and Senegal were septic tank sludges obtained from vacuum tracks during discharge at FS treatment plants, and all other samples were obtained by direct sampling from onsite containments, as described in Shaw et al. (2022).Previous studies by Ward et al. (2019Ward et al. ( , 2021) ) have established that the time since last emptied, for FS in containments, is not a predictor of dewaterability; hence, it was not considered in sample collection.Characteristics of 20 feces, 10 field FS, and one fresh FS sample used in the study are provided in Supplementary material.Samples were stored immediately in cooling boxes with ice packs before being airfreighted to Switzerland under refrigerated storage.For more consistent FS characteristics, fresh FS was prepared by mixing freshly collected feces and urine from a urine separating dry toilet at Eawag to obtain a feces-to-urine ratio of 1:2.5 by wet weight and diluted with water to 2% TS, according to Sam et al. (2022).The ratio of feces-to-urine is according to the daily per capita production of feces and urine, with an average of 400 g of feces and 1 L of urine (1:2.5)(Colón et al. 2015) and it was diluted to a TS of 2%, which is within the range of reported TS concentrations in FS (Velkushanova et al. 2021).Although the fresh FS recipe used in this study established a baseline for understanding dewatering performance, the results should be further validated with a range of fresh feces.Field FS and fresh FS were stored at 4 °C prior to experiments.

Analytical methods
Analysis of physical-chemical characteristics, including total and volatile solids (TS and VS), was performed gravimetrically according to methods for FS analysis (Velkushanova et al. 2021).pH and electrical conductivity (EC) were quantified with a WTW pH/conductivity of 3,320 meters (Xylem Analytics Germany GmbH).Total and soluble chemical oxygen demand (COD and sCOD) (Method 5220), total nitrogen (TN), and total phosphorus (TP) (Method 4500-NC and 4500-PE) were quantified with HACH Lange test kits, as described in Velkushanova et al. (2021).Analysis of feces samples involved diluting 0.1 g of feces in 50 ml nanopure water.
Total organic carbon (TOC) was analyzed using the Shimadzu TOC-VCPN TOC Analyzer according to Velkushanova et al. (2021).Mono-and divalent cations Ca 2þ , Na þ , Mg 2þ , and K þ were quantified with Inductively Coupled Plasma Optical Emission spectroscopy (ICP-OES)(Perkin Elmer, Waltham, MA, USA) on the supernatant after centrifuging samples at 3,000 g for 20 min.The supernatant was filtered through a 0.45-μm filter and acidified with 65% nitric acid in a 1:100 dilution (Park et al. 2006).Cations were reported in mg/L because they were also a part of the liquid fraction of feces and FS.
The PSD of FS samples was analyzed using static light scattering measurement, according to AHPA standard method 2560D (Velkushanova et al. 2021), using a Beckman Coulter LS 13 320-Laser Diffraction Particle Size Analyzer.
Total and intact cells were measured for both feces and field FS by flow cytometry using CytoFLEX (Beckman Coulter, Brea, California, USA) at a flow rate of 60 μL/min for 60 s.SYBR Green I stain was used to determine the total cells, while intact cells were determined with the aid of Sybr Green I/Propidium Iodide (SG/PI) stain.Total and intact cells helped to quantify dead and live cells (Van Nevel et al. 2017).
Sludge dewatering properties filterability, settling measured by supernatant turbidity (after centrifugation), and water-holding capacity were quantified with well-established metrics that are accepted and widely applied in wastewater and fecal sludge research, including capillary suction time (CST), supernatant turbidity after centrifugation, and dewatered cake solids (Christensen et al. 2015;Ward et al. 2019;Velkushanova et al. 2021).CST, supernatant turbidity, and dewatered cake solids (per gram wet weight) were determined according to the methods for FS analysis (Velkushanova et al. 2021).
The extraction and quantification of soluble and loosely bound EPS in both feces and FS were carried out by sonication and size exclusion chromatography (SEC), respectively, according to Ward et al. (2019), with dilution of the feces samples prior to analysis.Proteins and polysaccharides were determined on diluted feces and FS samples using colorimetric methods; the BCA protein assay with bovine serum albumin (BSA) as the standard, and the Anthrone assay with glucose as the standard, respectively (Raunkjaer et al. 1994).Lipids were extracted with the soxhlet extraction method using petroleum ether (Lenaerts et al. 2018) in a Gerhardt classic soxhlet apparatus at 40-65 °C (Gerhardt, Königswinter, Germany).The extraction was carried out for 7-8 h (about 24 cycles).The fiber content of feces and FS was analyzed according to Van Soest fiber analysis methods (Van Soest et al. 1991) using the Fibretherm FT 12 (Gerhardt, Königswinter, Germany; AOAC index no.973.18).

Effect of representative constituents on dewatering properties
To gain an understanding of the effect of constituents that govern dewaterability (as described in the introduction) specifically on fresh FS, materials identified as representatives for each of the constituents were used in a series of jar tests as described in FS analysis methods (Velkushanova et al. 2021).The representative materials identified from the literature for cellulose, hemicellulose, lignin, and lipids were cellulose fiber (CAS Number: 9004-34-6), xylan (CAS 9014-63-5), lignin alkali (CAS 8068-05-1), and sunflower oil (Migros, Switzerland), which is a standard material to represent food lipids (Krueger et al. 2021).NaCl, KCl, CaCl 2 , and MgCl 2 salts were used as sources to provide Na þ , K þ , Ca 2þ , and Mg 2þ in solution, and freeze-dried EPS extracted from activated sludge with the Na 2 CO 3 method (Schambeck et al. 2020) was used for EPS.Jar tests were conducted in quadruplicate with one control using 400 mL fresh FS.A higher dose (10X) of each constituent based on baseline concentration (1Â) (Supplementary material) was added to fresh FS and the system was stirred at a uniform speed of 60 rpm for 2 min (Shaw et al. 2022), after which, metrics of dewaterability were quantified.

Microbial community analysis
DNA of feces and field FS was extracted following a modified method by Griffiths et al. (2000) described in Sam et al. (2022).16S rRNA gene amplicon sequencing was carried out by Novogene on an illumina MiSeq platform based on bacterial and archaeal V3-V4 regions.Raw sequences were analyzed within the QIIME2 environment with the MIDAS database (Nierychlo et al. 2020).One FS sample, FS10 from Lebanon failed the sequencing test and is, therefore, not included in microbial community analysis.

Data analysis and statistics
R software and R Studio version 4.1.1(R Studio Inc., Boston, MA, USA) were used for the analysis of data.The significance of differences in mean organic and inorganic constituents of different food groups was tested with the analysis of variance (ANOVA) single test.Note: the data for this study are available at https://polybox.ethz.ch/index.php/s/j7a0E1PYqjU4iTQfor reviewers (password: FS*1234) and will be openly available in Eawag Research Data institutional Collection (ERIC) at https://doi.org/10.25678/0007MP.

Characteristics of feces
Feces collected from 20 study volunteers had an average TS concentration of 0.23 + 0.04 (g/g wet wt) and VS of 0.87 + 0.02 (g/gTS), and were within the range of 0.14-0.37(g/gTS) and 0.84-0.93(g/gTS), respectively, reported by Rose et al. (2015).Feces contained a TP concentration of 0.02 + 0.01 (g/gTS), which is the same as 0.02 (g/g) reported by Vinnerås et al. (2006).The TN of 0.07 + 0.03 (g/gTS) was comparable to 0.05 + 0.02 (g/gTS) in feces reported by Stephen & Cummings (1980), and is thought to come from undigested proteins or the protein content of microbial biomass.The total cells in feces were between 6.7 Â 10 11 and 16.1 Â 10 11 cells/g dry weight, which is in the range of 5 Â 10 11 + 4 Â 10 11 bacterial cells per gram dry weight of feces reported by Suau et al. (1999).The total cells in feces accounted for 8-19% of the TS, which is less than 25-54% of dry solids reported in the review by Rose et al. (2015).However, only 5% of the total cells were viable (Supplementary material) in contrast to 40-57% reported by Ben-Amor et al. (2005), which could be due to the freezing of samples.
The results of constituents in feces that are likely to govern the dewaterability of FS are illustrated in Figure 1.Overall, it was observed that the fiber content was predominantly lignin, as also observed by Krueger et al. (2021).Lignin fractions remain in feces due to low digestibility and bioavailability, resulting from limited enzymatic hydrolysis (Pérez et al. 2002).The concentration of cellulose was lower than 0.17 (g/g TS) reported by Gold et al. (2017).In this study, soluble and loosely bound EPS accounted for about 3% of the TS, which is higher than that previously observed in FS from septic tanks (0.7-1.0%), but comparable to EPS from pit latrine FS (2-3%) (Sam et al. 2022).The concentrations of proteins, polysaccharides, and lipids in feces were comparable to each other, and proteins and polysaccharides were within the range of 0.11-0.56 and 0.14-0.85(g/gTS) reported by Rose et al. (2015), although lipids were slightly higher than the reported 0.087-0.16(g/gTS).
Although soluble and loosely bound EPS contains proteins and polysaccharides, the proteins and polysaccharides quantified by colorimetric methods are greater, as they are the total amount contained in feces and can include undigested food matter, biomass, and EPS (Rose et al. 2015).Cations in feces reported by Rose et al. (2015) were within 0.08-0.72%,which is lower than in this study.Analyzing all these constituents for the same sample set allows for a direct comparison of the individual constituents, which was previously difficult.A comparison of the characteristics of feces and fecal sludge samples is provided in Supplementary material.

Effect of diet on feces characteristics
Feces were compared for differences in organic fractions reported in Supplementary material, Figure S2, by reported consumption of dairy, fruits, proteins, and vegetables, 24 h prior to the collection of feces (Kolodziejczyk et al. 2012).Five people reported consuming fruits and vegetables, and two did not, whereas six people reported consuming proteins and dairy, and six did not.None of the participants reported taking antibiotics or probiotics, and grains were not included as they were reported as consumed by all volunteers.Overall, there were no distinct differences in the organic constituents of feces based on diet, in this study (based on the ANOVA single test), although it is, in general, difficult to draw conclusions on the effect of diet on feces.One factor is that transit time in intestines is quite variable, reported from 24-48 h ( de Vries et al. 2016) to 40-60 h (Degen & Phillips 1996), which is thought to also vary by gender, physical activity, age, and body mass index (Procházková et al. 2023).Another is the role of diet on total mass, where it is worth noting that even in controlled diet studies where participants are given the same food, the characteristics and total production of feces are variable (Attebery et al. 1972).Rose et al. (2015) conclude greater feces production by vegetarians; however, the cited studies actually observed the same total production (Silvester et al. 1997) or methods of data collection that are not culturally relevant, based on purely observational data (e.g., 'White' versus 'Indian' women in London) (Reddy et al. 1998).Additionally, differences in the microbial communities in feces (section 3.4.2) did not vary by diet (Supplementary material, Figures S3-S6).This is in agreement with Ferrocino et al. (2015), who observed that environmental factors due to geographic location are stronger factors than diet.In conclusion, although the effect of diet or differences between individual stools may be of interest for technologies that are treating feces at the individual level (Mercer et al. 2021b), it is most likely not of relevance for community-or semicentralized management following storage of FS in containments.

Dewaterability of fresh and field FS and the role of representative constituents
In order to gain an understanding of constituents that are governing the dewaterability of FS, the influence of individual constituents on dewatering properties was evaluated with fresh FS, and compared to the range of constituents observed in field FS.Using the concentrations in fresh FS as the baseline, dewatering performance was evaluated following the addition of 10x aliquots of proxy constituents for cellulose, hemicellulose, lignin, lipids, EPS, and mono-and divalent cations.

Role of representative constituents on filtration time (CST)
As illustrated in Figure 2(a), the addition of cellulose fiber decreased CST for fresh FS (red circle) over the baseline case of fresh FS alone (green squares), and the addition of lignin alkali and EPS increased CST over the baseline.All constituents in field FS were similar to those in fresh FS, however, the CST was always lower.The effect of xylan/hemicellulose, oil/lipids, and mono-and divalent cations was not substantial (Figures 2(a) and Supplementary material, Figure S6).Cellulose has been observed to improve filtration in wastewater sludges by providing surfaces for the attachment of small particles and microorganisms, thus preventing clogging of filter media or reducing resistance to filtration (Zhang et al. 2020).The role of free and bound water plays a role in the dewatering of wastewater sludges, with bound water more difficult to remove (Chen et al. 2002).Alternatively, filtration is hindered by cross-linked polymeric networks of cellular and particulate components that can form an impermeable layer (Skinner et al. 2015).This appears to be the case for FS, where EPS behaves like colloidal and suspended organic substances that clog pores of filter media, and is more relevant than bound and free water (Ward et al. 2019).

Role of representative constituents on supernatant turbidity (after centrifugation)
The addition of cellulose fiber and xylan/hemicellulose decreased the supernatant turbidity of fresh FS (Figure 2(b)), elevated lipid/oils, EPS, and Ca 2þ increased supernatant turbidity, and monovalent cations did not have an effect (Figures 2(b) and Supplementary material, Figure S6).The turbidity of field FS was always lower than fresh FS, indicating that field FS contained less fine, unbound particles.Fibers are reported to reduce supernatant turbidity in wastewater sludges by reducing suspended residual colloids or small particles (Huang et al. 2019).In contrast, Höfgen et al. (2019) report that fibers reduce gel point TS by inhibiting the compressional effects of gravity, making fibers detrimental to settling performance through increased turbidity.Mercer et al. (2021aMercer et al. ( , 2021b) also predict a better settling performance in fresh feces than field FS, but it is difficult to compare the results as the feces were macerated and turbidity was not measured in their study.In this study, compression yield stress was also not measured.In contrast to wastewater sludges where Ca 2þ is reported to decrease supernatant turbidity, based on the cation bridging theory (Higgins & Novak 1997), it does not appear to hold true for FS, which was also observed by Ward et al. (2019), and is interesting, as it is a marked difference to wastewater sludges.Based on these results, the cation bridging theory proposed for wastewater sludges does not apply in FS (Ward et al. 2019).One potential explanation for this is that the theory of divalent cation bridging is based on the premise that divalent cations drive bioflocculation by bridging EPS to negatively charged sites on cell surfaces or by interlinking EPS molecules.However, the interactions between divalent cations and EPS have been observed to be significant for floc formation at high EPS concentrations of the sludge (activated sludges) (.100 mg/gTSS).

Role of representative constituents on water-holding capacity (dewatered cake solids)
The addition of cellulose fiber and lignin alkali increased dewatered cake solids but EPS, lipids, and cations did not have a substantial effect.Elevated cake solids with higher fiber concentration are thought to be due to the chelating properties of fibers (Zhang et al. 2020).Potentially, EPS from feces or fresh FS have different properties and water-holding capacities than EPS in activated sludge which is built up during growth in an aerobic environment, as it did not affect the water-holding capacity in FS (Guo et al. 2020).

Interdependence of dewatering metrics
In this study, as a first step, we evaluated one by one whether organic constituents and cations have an effect on dewaterability.More extensive research is needed in order to fully understand the mechanisms underlining dewatering and the complex interactions of fecal sludge components, in order to be able to reliable model or predict dewatering behavior.In addition, the reported metrics of dewaterability are interrelated.High filtration rates correlate with low supernatant turbidity because the small particles responsible for clogging pores are also responsible for turbidity.Greater sludge cake solids means more particles are captured, which also translates to lower supernatant turbidity.Our results largely corroborate this interrelatedness.However, there were some discrepancies, for example, an increase in supernatant turbidity due to the addition of oil was expected to increase CST and decrease cake solids, but this was not observed.Methods based on the rheological properties of FS and the use of filtration models could potentially provide further insight into these observations (Skinner et al. 2015).

The role of stabilization on dewaterability
Field FS had better dewaterability than fresh FS for all the dewatering metrics, which is most likely due to differences in the level of stabilization.In this study, field FS had a VS/TS of 0.51-0.77and a dark gray to black color, appearing to be more stabilized than fresh FS, which had a VS/TS of 0.85 with a light brown color (Ward et al. 2021).Additionally, the VS concentration of fresh FS (86%) was greater than field FS (66%), which also suggests that fresh FS has a high water-holding capacity, if VS follows similar relationships to wastewater sludges (Skinner et al. 2015).During stabilization, the breakdown of organic substrates results in changes in the physical properties of the sludge, for example, the breakdown of alginate-like exopolysaccharides (ALE) forms of EPS, which are gel-forming agents with higher water-holding abilities (Lin et al. 2010).Insoluble fibers that contribute to water-holding capacity and gel formation, and are poorly digested by gut microbiota, are more likely to be present in fresh FS (Wenzl et al. 1995).As observed in this study, soluble and loosely bound EPS in fresh feces and fresh FS were higher than in field FS.In general, during the first week of anaerobic storage, fresh FS has been observed to have a reduction in readily biodegradable organic matter and an improvement in the dewatering performance, which then levels off after one week (Sam et al. 2022;Ward et al. 2023).Lignin, hemicellulose, and cellulose were higher in field FS than fresh FS (32, 12, and 10.7%), which could also be due to a relative increase as other constituents are degraded, in addition to fiber inputs from toilet paper and food waste (i.e., fruits, vegetables).
The distribution of organisms in feces and field FS is quite different (Figure 3), which should be expected as the environment in the intestines is quite different from that within onsite storage.The presence of common microorganisms such as Firmicutes indicates the persistence of some fecal-associated microbiota that are able to survive outside the human body.The microbial community in feces had commonly reported gut microbiota such as Blautia (17 + 8%) and Subdoligranulum (13 + 7%), while field FS had a high relative abundance of Pseudomonas (21 + 22%) and Clostridium (14 + 16%) (Cai et al. 2014).This was also observed in pit latrines in Tanzania, where a shift from gut microbiota to environmental and wastewaterrelated communities was seen with depth (Ijaz et al. 2022).Differences in microbial communities also help to explain the differences in stabilization and dewaterability in fresh FS and field FS.For example, Proteobacteria are known versatile consumers of organic substrates and are largely responsible for organic matter removal in wastewater systems (Belila et al. 2013), and Pseudomonas has been associated with larger aggregate formation and faster filtration in FS (Ward et al. 2019(Ward et al. , 2023)).Additionally, methane-producing Methanosaeta (Euyarchaeota) was observed in a higher abundance in fecal sludge (11%) compared to feces (0.03%), indicating biomass degradation (stabilization) in field fecal sludge samples.

Implications
This study clearly demonstrates that fresh feces and fresh FS that have not been stored in containments are different from field FS, and the observed differences are accredited to the overall changes that take place during stabilization.A further understanding of the role of stabilization during storage in containments will lead to a better understanding and control over treatment performance.The results of this study shed light on governing mechanisms of FS dewaterability due to fibers, EPS, lipids, and cations; however, our findings suggest that conventional characterization parameters (e.g., VS and COD) or individual organic contents (e.g., cellulose, lipids, and lignin) do not adequately capture important differences in characteristics or properties that mediate dewatering performance.Furthermore, the relationship between dewatering and stabilization and how to best predict dewatering performance require more study.Currently, there are not yet any agreed-upon metrics for quantifying the stabilization of FS, or how metrics of stabilization relate to dewaterability.It is commonly considered that the time since last emptied is equivalent to overall storage, and is a predictor of stabilization (Mercer et al. 2021a(Mercer et al. , 2021b)).However, the time since last emptied does not equal the total storage time, as there are continual fresh inputs to the containments, and anaerobic storage is not the same as anaerobic digestion, which is process-controlled for optimum degradation (Shaw & Dorea 2021).Furthermore, the time since last emptied has no statistical relation to stabilization (Ward et al. 2023), and we have confirmed with our laboratory research that following the initial first week of storage, additional storage time in containments does not result in a continual degradation or stabilization (Sam et al. 2022;Ward et al. 2023).Hence, even though field fecal sludge that has been stored in containments is, in general, more stabilized than fresh fecal sludge prior to storage, and has better dewatering properties, the time since last emptied does not predict the level of stabilization.Based on 1,206 data points from 13 countries, a VS/TS of 0.49 (R 2 ¼ 0.88) has been observed for field FS (Andriessen et al. 2023), which is lower than 0.6-0.8 for primary wastewater sludge (Tchobanoglous 2014).However, VS/TS relationships of stabilization observed in municipal wastewater are not directly transferable to FS because organic fractions of VS in FS are comprised of less readily biodegradable organics (Krueger et al. 2021).In addition, there is a greater concentration of inert TS of FS from soil intrusion and non-biodegradable garbage, whereas sewerage typically undergoes grit removal.It is also clear that other common indicators such as mono-/divalent cations and dewatering performance in wastewater treatment processes are not directly transferable to FS. Metrics of bioavailability that appear more promising for FS include color (Ward et al. 2021) and metrics of biological activities such as biochemical oxygen demand (BOD) and biomethane potential (BMP).
Knowledge of stabilization and its relation to treatment performance will be useful for the design of non-sewered treatment solutions that treat excreta closer to the source of production, reducing reliance on trucking large volumes of water through cities to treatment.To improve the dewatering of feces or fresh FS, they should first undergo some level of stabilization.For instance, fresh FS from no-flush container-based sanitation (CBS) solutions could employ anaerobic digestion for the conversion of readily biodegradable fractions in feces-to-biogas before dewatering on a drying bed.The variation between individual feces is not relevant for community-scale to semi-centralized treatment of FS that has undergone storage and transport to treatment, as the complexity of individual variation is averaged out.However, differences in individual feces may be relevant for the development of emerging RTTC technologies that combine the front end (user interface) with the back end (containment) for simultaneous onsite containment and treatment.

CONCLUSIONS
The specific conclusions from this research include: • Feces, fresh FS, and field FS have distinct characteristics.Although FS includes feces, there are many additional inputs, differences in microbial communities, levels of stabilization, and particle size distribution, all of which contribute to the properties of dewaterability.
• Fresh FS has worse dewatering performance than field FS that has been stored in containments, which is likely due to the role of stabilization, although stabilization processes during storage are still not well understood.
• Fibers, EPS, and lipids appear to be key factors that control the dewaterability of FS.
o Filtration was reduced by EPS and lignin, and improved by cellulose.o Supernatant turbidity was reduced by cellulose and hemicellulose, and decreased by lipids.o Cellulose and lignin increased cake solids and decreased EPS.o Mono-and divalent cations did not have an effect on dewatering performance in FS.

AUTHORS CONTRIBUTIONS
S.S.B. was responsible for the experimental design, collection, and analysis of samples, and took the lead in writing the manuscript.E.M. provided critical feedback and helped to shape the manuscript structure.S.L. conceived the project idea, contributed to writing, obtained funding, and supervised the project.

Figure 1 |
Figure 1 | Boxplots showing (a) concentrations of fibers (cellulose, hemicellulose, and lignin), EPS, and the organic content (proteins, polysaccharides, and lipids) per gram TS of feces and (b) mono-and divalent cation concentrations in mg/L in feces.

Figure 2 |
Figure 2 | Dewaterability of fresh and field FS showing the effect of representative constituents on dewatering properties: (a) change in filtration time (CST) in fresh FS and CST of field FS; (b) supernatant turbidity (after centrifugation; and (c) water-holding capacity (dewatered cake solids).Green squares represent the baseline dewaterability of fresh FS, red circles represent the addition of aliquots to fresh FS, and blue triangles represent field FS.Trends in field FS are also provided in Supplementary material at different scales for clarity.Please refer to the online version of this paper to see this figure in colour: http://dx.doi.org/10.2166/washdev.2023.086.

Figure 3 |
Figure3| Relative abundance of the 10 most abundant phyla and genera in feces (F1-F20) and field FS (FS1-FS9) in this study.'Others' are organisms identified, but not part of the 10 most abundant.