Effects of different growth media on water quality and plant yield in a catfish-pumpkin aquaponics system

Metadata

Cite key: oladimejiEffectsDifferentGrowth2020
Item type: Journal Article
Authors: A.S. Oladimeji, S.O. Olufeagba, V.O. Ayuba, S.G. Sololmon, V.T. Okomoda
Affiliation: Agricultural Department, National Biotechnology Development Agency (NABDA), Abuja, Nigeria (Oladimeji); Department of Fisheries and Aquaculture, University of Agriculture, Makurdi, Nigeria (Olufeagba, Ayuba, Sololmon, Okomoda)
Journal: Journal of King Saud University – Science 32 (2020) 60–66
Date: 01/2020 (print issue); accepted 4 February 2018, published online ahead-of-print in 2018 — see Extraction notes on the year discrepancy
Date added: 2026-08-07
DOI: 10.1016/j.jksus.2018.02.001
Funding: [not reported]
URL: https://doi.org/10.1016/j.jksus.2018.02.001
PDF: Oladimeji et al. - 2018 - Effects of different growth media on water quality.pdf

Opinion

A short, straightforward media-comparison experiment (4 grow-bed substrates, no hydroponic arm at all) whose main value is the palm-kernel-shell/periwinkle-shell substrate result — both outperformed gravel and were statistically indistinguishable from each other, a genuinely useful “waste-to-wealth” finding for the sub-Saharan African context the authors are writing for. The reporting is thin by this vault’s usual standard: fish growth is given as one pooled figure never broken out by tank/media despite each media treatment running its own independent fish tank, all plant growth/yield data (vine length, leaf count, leaf area, branch number, and the headline “plant yield (kg)” figure) exists only as line/bar charts with no supporting table or text values, and several basic design parameters (stocking density in kg/m3, FCR, SGR, feeding frequency, exact trial duration in days) are never stated in extractable form. Water-quality data is comparatively solid (a full compartment breakdown: fish effluent / inlet / outlet-by-media, Table 2), though the paper never clarifies whether its “NH3/NO2/NO3 (ppm)” figures are N-equivalent or whole-ion values, a common ambiguity worth flagging for anyone pooling this with papers that do specify NO3-N. No hydroponic comparison of any kind, so this paper cannot speak to the aquaponics-vs-hydroponics question — only to substrate choice within aquaponics.

Abstract

Aquaponics production of fish and plant is one of the environmentally sustainable farming methods of the twenty-first century. In this study, two agricultural wastes (Palm kernel shells aka PKS and Periwinkle shells aka PWS) were evaluated for their suitability as growth beds in the propagation of pumpkin Telfairia occidentalis. In addition, conventionally used gravel and its mixture with PKS and PWS (33.33% each) were also tested. The resultant effects of these treatments on water quality of the system were reported. The result obtained reveals better performance of pumpkin in the PKS and PWS medias in terms of vine length, leave area, leave number, branch number and bi-weekly plant yield. The least performance was observed in the plants propagated in the gravel substrate. Water quality and percentage nitrogenous compound reduction (NH3, NO2 and NO3) across the system and in the different grow beds suggests that mixing all substrate resulted in better water qualities for fish and plant growth. Aside the superior growth observed in this study, alternative uses of these agricultural by-products in aquaponics system is highly recommended because of the possibility of converting waste to wealth.

Summary

The authors ran a four-treatment substrate comparison inside a small-scale, media-bed aquaponics facility at NABDA, Abuja, Nigeria: gravel (the conventional control medium), palm kernel shells (PKS), periwinkle shells (PWS), and a 33.33%-each mix of all three, each with its own independent 200 L fish tank (50 African catfish, Clarias gariepinus, juveniles per tank) and its own complete recirculating loop feeding 8 planting troughs (32 troughs total). Fluted pumpkin (Telfairia occidentalis) was direct-seeded into the troughs and monitored biweekly from week 4 to week 16 for vine length, leaf number, leaf area, branch number and yield; water quality (temperature, pH, DO, alkalinity, hardness, NH3, NO2, NO3) was tracked weekly at the fish tank, the shared trough inlet, and each treatment’s trough outlet. Pumpkin grown in PKS and PWS significantly outgrew and out-yielded plants in gravel and in the mixed substrate on every growth measure; PKS and PWS did not differ significantly from each other. All four media showed broadly similar water chemistry except for NH3, NO2 and NO3, which differed significantly by compartment and, to a lesser degree, by media — the mixed substrate achieved the highest ammonia removal. Fish grew well overall (mean gain of roughly 656 g per fish over the four-month trial from a ~10 g starting weight), but this figure is reported only once, pooled, and is never broken down by media/tank. The authors conclude that PKS and PWS are viable, low-cost alternatives to gravel for aquaponic grow beds and recommend their wider adoption as a waste-to-wealth strategy in the region.

Experiment data

Location: National Biotechnology Development Agency (NABDA) Headquarters, Umar Musa Yar’adua Express, Airport Road, Lugbe, Abuja, Nigeria

Design: 4 grow-media treatments (Gravel/GRV — control; Palm Kernel Shell/PKS; Periwinkle Shell/PWS; Mixed substrate at 33.33% each) x 8 planting-trough replicates/treatment (32 troughs total); each treatment has its own independent fish tank (A1–A4, 50 catfish/tank, 200 fish total) and complete RAS loop; no hydroponic treatment

Replicates / n: 8 troughs per treatment

Duration: “Four months” of fish culture ([UNIT CONVERSION ONLY, approximate] ~120 days); plant growth monitored biweekly from week 4 to week 16 post-germination

Organisms: Fluted pumpkin (Telfairia occidentalis) / African catfish (Clarias gariepinus)

Statistics: One-way ANOVA + Fisher’s least significant difference (LSD) post-hoc, P<0.05; Minitab 14 software

Plant yield: PKS and PWS significantly higher than GRV and MIXED (Fig. 7; letter groups only, no table/text values — see Extraction notes)

Fish weight gain: 655.89 ± 0.98 g/fish over the trial (pooled across all 4 tanks/media, not broken out)

Water quality (NH3/NO2/NO3): significantly different by compartment (fish tank > trough inlet > trough outlet) and, more weakly, by media; ⚠️CHECK species basis (NH3 vs NH3-N, etc.) unstated — see Extraction notes

Growth media and plant yield

This paper: Pumpkin grown in PKS and PWS significantly outperformed gravel-grown and mixed-substrate-grown plants across vine length, leaf number, leaf area, branch number, and bi-weekly yield (Figs. 3–7); PKS and PWS never differed significantly from each other. Fig. 7’s final “Plant yield (kg)” bar chart marks PWS and PKS with significance letter “a” and GRV and MIXED with “b” (i.e. GRV and the mixed substrate were statistically indistinguishable from each other and both significantly lower than PKS/PWS) — no absolute kg values are given in text or a table, only the chart, so no numeric yield value was extracted (never read a value off a figure). The authors attribute the gravel shortfall to its poor water/nutrient retention around the root zone (citing Mader 2012 and Rakocy et al. 2006) and link the PKS/PWS advantage to differences in nitrification/nutrient availability among the media (discussed below).

Compared with:

todo Mader 2012 — lettuce grew significantly better in coconut-husk media than gravel, attributed to gravel’s poor water/nutrient retention; cited as the direct analogue for this paper’s own GRV shortfall (p.6)
todo Rakocy et al. 2006 — cites gravel’s inability to retain sufficient water/nutrients as a known disadvantage of the medium in aquaponics (p.6)
todo Akoroda and Adejoro 1990 — sigmoidal vegetative growth pattern reported for Telfairia occidentalis, cited to explain a growth slowdown observed at weeks 10–12 across all four treatments (p.6)
todo Trang et al. 2010 — different growth media affect plant nutrient uptake in aquaponics, cited as the general mechanism behind this paper’s own media effect (p.5)
todo Lennard and Leonard 2006 — comparison of gravel-bed, floating and NFT hydroponic subsystems in aquaponics, cited alongside Trang et al. 2010 for the media-affects-uptake argument (p.5)

Water quality and nitrogenous-compound removal

This paper: Temperature, pH, alkalinity and hardness were statistically similar across the fish tank, trough inlet, and all four media’s trough outlets. NH3, NO2 and NO3 all decreased significantly along the flow path (fish tank > trough inlet > trough outlet), consistent with active nitrification; DO showed the same declining pattern but did not differ significantly among the four outlet media. Percentage removal of NH3 was consistently much greater than for NO2 or NO3 (Fig. 2), which the authors attribute (citing Yamamoto et al. 2008) to ammonia-oxidising bacteria proliferating faster than nitrite-oxidising bacteria. The mixed substrate achieved the highest ammonia percentage reduction; the paper argues this supports mixing all three substrates for the best combined water quality, even though the mix did not produce the best plant growth (PKS/PWS alone did). See Extraction notes for a ⚠️CHECK flag on whether the paper’s NH3/NO2/NO3 figures are nitrogen-equivalent or whole-ion values, which affects comparability with papers reporting NO3-N etc. directly.

Compared with:

todo Maucieri et al. 2017 — observed insignificant differences in all water qualities except NH4 among compartments of their aquaponics system, broadly consistent with this paper’s own compartment comparison, though this paper found NH3/NO2/NO3 all significant (p.3-4)
todo Akinwole and Dauda 2014 — palm kernel shell gave better nitrification efficiency than imported polypropylene blocks for aquaculture wastewater treatment, cited as directly corroborating this paper’s PKS result (p.4)
todo Fang et al. 2017 — media-based aquaponics with varying aeration levels; cited re: low ammonia/NO2 levels indicating a well-developed microbial community, consistent with this paper’s findings (p.4)
todo Wahyuningsih et al. 2015 — nitrogen removal of aquaculture wastewater in an aquaponic recirculation system using lettuce; cited as a favourably comparable result (p.4)
todo Schmautz et al. 2017 — microbial diversity across aquaponics system compartments, cited for the general claim that microbial communities drive nutrient dynamics (p.4)
todo Tyson et al. 2004; Tyson et al. 2008 — reconciling pH for ammonia biofiltration/nitrification vs. plant and fish requirements; cited for the trade-off pH range this study’s own pH (6.8–6.9) fell below, apparently without detrimental effect (p.3)
todo Yamamoto et al. 2008 — proliferation of ammonia-oxidising bacteria is faster than nitrite-oxidising bacteria, cited as the mechanism behind this paper’s much larger ammonia (vs. nitrite/nitrate) percentage removal (p.4)

Fish growth

This paper: Fish (African catfish, Clarias gariepinus, 200 juveniles total, 50/tank across 4 independent tanks) gained 655.89 ± 0.98 g per fish over “the four months of culture” from an initial mean weight of 9.97 g, fed a commercial diet (Coppens, 45% crude protein) at 5% of body weight. This is reported as a single pooled figure for the whole study — never broken out by media treatment/tank — despite each treatment running its own separate fish population, so no media-specific fish-growth comparison is possible from this paper. No FCR, SGR, survival rate, or stocking density (kg/m3) is stated.

Compared with:

No fish-growth literature comparisons are made in this paper; the Discussion’s citations focus entirely on water quality and plant/media performance.

Linked claims

Citations to chase

Extraction notes

Metadata judgment call — publication year: the PDF itself is an Elsevier “Article in Press” proof (header: “Received 6 January 2018, Accepted 4 February 2018, Available online xxxx,” copyright line ”© 2018 The Authors”), and the source filename correspondingly reads “…2018…”. However, Crossref (https://api.crossref.org/works/10.1016/j.jksus.2018.02.001) gives published-print: January 2020, volume 32, issue 1, pages 60–66, with issued: [2020, 1] — i.e. the article was formally assigned to the January 2020 print issue after a roughly two-year “in press” period. Per CLAUDE.md (“Resolve title, authors, journal, year, volume, pages from crossref… PDF headers are frequently wrong”), this note uses year=2020 and Paper ID oladimejiEffectsDifferentGrowth2020, diverging from the in-press-year filename. Flagged explicitly since it affects the citekey/filename and any pooled by-year analysis; the paper’s own data-collection period is unaffected either way.

⚠️MINOR initial fish stocking weight, SD conflict (p.2 vs p.3): Methods states juvenile catfish “weight = 9.97 ± 0.21 g; length = 12.07 ± 1.92 cm.” Results restates fish “gained 655.89 ± 0.98 g … from an initial weight of 9.97 ± 0.55 g.” The mean is identical (9.97 g) in both places; only the SD differs (0.21 vs 0.55). No interpretive impact. trials.csv Fish size initial records 9.97 ± 0.21 (the original Methods stocking description); the Results-restated 0.55 is noted as the alternate in Experimental Remarks.

⚠️MINOR Table 2 caption significance threshold (p.3): the caption reads “Mean in the same row with different superscripts differ significantly (P < 0.0)” — almost certainly a typo for “P < 0.05,” matching the Methods statistics paragraph and the Fig. 2/Fig. 7 captions (“P ≤ 0.05”) used consistently elsewhere in the paper. No cell affected.

⚠️CHECK ammonia/nitrite/nitrate species basis unstated (Table 2, p.3): the water-quality table labels its nitrogen rows simply “NH3 (ppm),” “NO2 (mg l⁻¹)” and “NO3 (ppm),” never stating whether these are nitrogen-equivalent species (NH3-N / NO2-N / NO3-N — the basis this vault’s schema columns are defined in) or whole-ion/compound concentrations, and the instrument used (Hanna HL 98126 multi-parameter tester) is not independently documented here as reporting one basis or the other. The paper’s own ppm figures were entered directly into trials.csv’s TAN/NH4-N, NO2-N and NO3-N columns (treating ppm as numerically equivalent to mg/L, which requires no conversion), but the N-only-vs-whole-ion question remains open and could matter by a factor of several fold if this paper is pooled against studies that explicitly report the -N basis (cf. SCHEMA.md’s NO3-vs-NO3-N example, a factor of 4.43). Added to REVIEW.md by the batch merge step.

UNIT CONVERSION ONLY: Water recycle — Table 1’s “Hydraulic loading rate: 7.5 L/hr” → 7.5 / 60 = 0.125 L/min. Fish trial duration — Results’ “the four months of culture” has no exact day count stated anywhere in the paper; recorded ~120 (4 × 30 = 120, assuming a 30-day month). Unlike a weeks-to-days conversion, a months-to-days conversion is not exact, so this figure carries acknowledged imprecision beyond what the authors themselves stated.

Trial structure judgment call: this paper has no hydroponic treatment of any kind — all four rows (Gravel/GRV, Palm Kernel Shell/PKS, Periwinkle Shell/PWS, Mixed substrate) are aquaponic treatments, each with its own independent fish tank and complete recirculating loop. The authors describe Gravel as “the control growth media” (p.2), but this is a substrate baseline within the aquaponics design, not a hydroponic control in the schema’s sense — so HYD-labelled columns are NA throughout, per SCHEMA.md’s rule for papers missing half the schema, rather than being populated with GRV’s own values. Recorded as oladimejiEffectsDifferentGrowth2020-T1 (GRV) through -T4 (Mixed), one row per media treatment, per SCHEMA.md’s “one row per aquaponic treatment” rule.

Fish data not disaggregated by treatment: although each of the four media treatments runs its own separate fish tank and population (A1–A4, 50 fish each), the paper reports fish growth (weight gain, initial weight) as a single pooled figure for “the fish reared in the aquaponics system” as a whole, never broken out per tank/media. The same pooled Fish weight gain (655.89 ± 0.98 g) and Fish size initial (9.97 ± 0.21 g) values are therefore recorded identically across all four trials.csv rows — this is a limitation of the paper’s own reporting, not an extraction error.

Plant Category judgment call: recorded as “Leafy vegetable and edible-seed crop / tropical vine,” quoting the Introduction’s own descriptive phrase (“a tropical vine propagated in West Africa mainly for its leafy vegetables and edible seeds,” p.1) rather than a formal category label the paper explicitly assigns to its trial design — flagged as a softer judgment call than, e.g., a paper that heads a table column “Leafy vegetables.”

System type left NR: the paper never applies a categorical term (e.g. “media bed,” “ebb-and-flow,” “NFT”) to its own design, describing it only as planting troughs/bowls filled with growth medium and continuously drip-fed at a stated hydraulic loading rate; the Introduction’s background discussion of NFT/EAF/DWC bed types describes aquaponics generally, not this system specifically. Per SCHEMA.md, no substitute taxonomy was applied.

[not reported] fields, grouped:

Fish: Fish Category, Initial Stock density (kg/m3 — fish count and tank volume given, but SCHEMA.md disallows deriving a density from these), FCR, SGR, feed N/P/K composition (only crude protein/fiber/moisture/ash given), Fish size final, Feed routine (frequency), Total Feed (kg), Fish biomass created (kg), Fish survival rate.
Water: Water type, Water classification, pHOptimal, FUE AP, WUE, EC.
Plant: Days Plant after transplant (direct-seeded, not transplanted; no single total days-to-harvest stated), Plants/m2, SPAD, Plant height, Leaf count, Plant fresh weight, Plant dry matter, Tissue nitrate AP (no plant tissue analysis performed), AP (yield figure exists only in a bar chart — never read off a figure).
Site: Lat, Long (no coordinates stated anywhere in the paper; not looked up externally), Average room Temperature.
Design/management: Artificial Lighting, Climate control, pH Buffers, Iron supplemented, Remineralization, Nutrient supplemented (none of these are mentioned anywhere in the paper; genuine silence recorded as NR, not N, per the N-vs-NR distinction in SCHEMA.md — note the Introduction’s general claim that aquaponics “allows sustainable growth … without the use of chemical fertilizer” describes aquaponics broadly and was not treated as a Methods-level confirmation for this specific system).

No water panel excluded from plant.csv: this paper’s water-quality panel (Table 2) is fully captured as trial-mean values in trials.csv per SCHEMA.md (aquaponic-loop water chemistry belongs there, not in plant_measurements.csv).

plant_measurements.csv left header-only: the paper reports no plant tissue analyte data of any kind — no biochemistry (chlorophyll, phenols, etc.), mineral/elemental tissue content, microbiology, or proximate composition for the pumpkin. All plant data collected (vine length, leaf number, leaf area, branch number, yield) are morphological/yield growth metrics that belong in trials.csv’s dedicated columns, not plant_measurements.csv’s four analyte categories, and in any case are only available as figures (see above), not extractable text/table values. Confirmed by a full read of Results and Figs. 3–7 before deciding to leave this file header-only.

New tags introduced: Meta/Plant/Fluted-Pumpkin (new leaf — no existing pumpkin/Telfairia facet anywhere in the vault; follows the existing hyphenated Title-Case convention, e.g. Meta/Plant/Water-Spinach). Meta/Region/Africa and Meta/Fish/African-Catfish reused exactly as already spelled in existing vault notes (ahmedImpactBiofilmSupport2026/goddekFullyIntegratedSimulation2019/magwazaPartiallyTreatedDomestic2020/godaOptimizingNutrientUtilization2024/goddekNecessityDesalinationTechnology2018/fidjiIntegratedAquaponicCoCulture2026 for Africa; knausAquaponicGrowthBasil2024 for African-Catfish, matching this paper’s own species, Clarias gariepinus). Meta/Type/Experiment per SCHEMA.md.

PDF quality: clean text layer throughout (Elsevier “Article in Press” typeset proof), no OCR issues, 7 pages, all figures/tables legible as prose+captions; only the underlying numeric data in Figs. 3–7 is inaccessible (bar/line chart only, no table), which is a reporting limitation of the source, not a PDF-quality problem.

Source: Oladimeji et al. - 2018 - Effects of different growth media on water quality.pdf

Data Tables

Structured data extracted from this paper into the vault's trials.csv / plant_measurements.csv datasets. Fields the paper didn't report are omitted. Download the full datasets (measurements).

Trial Parameters

oladimejiEffectsDifferentGrowth2020-T1

Fish

Field	Value
Fish	African catfish (Clarias gariepinus), juveniles
Protein	45
% of body weight	5
Fish size initial	9.97 +/- 0.21
Feed regime	Commercial diet Coppens (45% crude protein, 1.5% fiber, 8.2% moisture, 9.5% ash) fed at 5% of total body weight (Results, p.3)
Fish weight gain	655.89 +/- 0.98
Fish trial duration (days)	~120

Water

Field	Value
Water recycle	0.125
Water volume in the system	800 (0.8 m3); Table 1 gives this as a single generic system figure — assumed to describe one treatment’s complete loop (fish tank+filter+sump+troughs), not summed across all 4 treatments; see Extraction notes
Daily Water exchange rate	5
Aq pH	6.82 +/- 0.51
Dissolved Oxigen	4.39 +/- 0.49
Water temperature	28.79 +/- 1.50
TAN / NH4-N	2.30 +/- 0.80
NO2-N	0.52 +/- 0.04
NO3-N	104.2 +/- 5.20

Plant

Field	Value
Plant	Fluted pumpkin (Telfairia occidentalis)
Details	Pumpkin pods purchased alongside the PKS from Imo State, Nigeria; pods cut to expose seeds, 2 seeds planted per trough; yield/growth data (vine length, leaf number, branch number, bi-weekly plant yield) collected from week 4 post-germination, every 2 weeks thereafter through week 16 (Cornelissen et al. 2003 protocol, p.3); leaf area also tracked (Fig. 4). All growth-curve and yield data reported only in Figs. 3-7 (line/bar charts, Mean +/- SE), no accompanying table or text values.
Plant Category	Leafy vegetable and edible-seed crop / tropical vine (descriptive wording from Introduction, p.1: ‘a tropical vine propagated in West Africa mainly for its leafy vegetables and edible seeds’ — not a formal categorical label applied to the trial design; judgment call, see Extraction notes)

System & Setup

Field	Value
Media Details	Gravel (GRV), conventional substrate, control growth media (p.2); troughs filled to 50% capacity
Biological system already in use	Y (Fig. 1 depicts a distinct biofilter component (separate from the 20 L mechanical filter tank listed in Table 1) in each treatment’s recirculating loop, between the fish tank/mechanical filter and the sump; nitrifying bacterial action is inferred by the authors from the water NH3->NO2->NO3 concentration trend (p.4); no separate biofilter volume or media stated.)
Air supplement	Y (Air blower and air valve shown in the system diagram (Fig. 1); no aeration rate/flow stated.)
Equipment	Hanna water tester Model HL 98126 (digital multi-parameter water checker, for temperature/pH/DO/alkalinity/hardness/NH3/NO2/NO3); autoclave (100 C, 1 h, substrate sterilization); Minitab 14 statistical software
Control Parameters	Feeding rate 5% of total body weight; hydraulic loading rate 7.5 L/hr (Table 1); stated daily water top-up of 5% of total system water to offset evaporation/transpiration losses (design assumption, citing Maucieri et al. 2017, p.3)
Combination	African catfish (Clarias gariepinus) and fluted pumpkin (Telfairia occidentalis); comparison of 4 grow-bed media (gravel/control, palm kernel shell, periwinkle shell, and a 33.33%-each mix of all three) within independent, media-based aquaponic loops — no hydroponic treatment in this study

Site

Field	Value
Region	Africa
Country	Nigeria

Results & Statistics

Field	Value
Measured Unit	C (temperature); ppm (Dissolved Oxygen, NH3, NO2, NO3 — treated as mg/L equivalent per the paper’s own units, see WARN-CHECK in remarks); % (nitrogenous-compound reduction, Fig. 2, NO COLUMN)
Statistic Details	Descriptive statistics; one-way analysis of variance (ANOVA); Fisher’s least significant difference (LSD) post-hoc when P<0.05; Minitab 14 (Methods, p.3)
Statistically analysed	Y
Replicates (n)	8

Experimental Remarks: TRIAL DEFINITION: T1 = Gravel (GRV) growth medium — the conventional substrate, described by the authors as ‘the control growth media’ (p.2) for comparison against the two waste-derived media and their mix. Own independent fish tank (A1, 50 catfish) and RAS loop; 8 planting troughs filled to 50% capacity with gravel. This ‘control’ is a substrate baseline within the aquaponics design, not a hydroponic control — no hydroponic treatment exists in this study, so HYD columns = NA. | No hydroponic control exists anywhere in this study — all four rows are aquaponic treatments compared against each other; Gravel (GRV) is explicitly described by the authors as ‘the control growth media’ (p.2) — i.e. the conventional-substrate baseline against which the two waste-derived media (PKS, PWS) and their mixture are judged — but this is a substrate control, not a hydroponic control, so HYD columns = NA throughout per SCHEMA.md. Design: 4 treatments (Gravel/GRV, Palm Kernel Shell/PKS, Periwinkle Shell/PWS, Mixed substrate/MIXED at 33.33% each of GRV+PKS+PWS) x 8 planting-trough replicates each (32 troughs total, Fig. 1/Table 1); each treatment served by its own independent fish tank (A1-A4, 50 African catfish juveniles/tank, 200 fish total across the study) and its own complete recirculating loop (fish tank -> mechanical filter -> biofilter -> sump -> pump -> planting troughs -> effluent back to sump, Fig. 1). One-way ANOVA + Fisher’s least significant difference post-hoc (P<0.05); Minitab 14 software (Methods, p.3). | TAN/NH4-N, NO2-N, NO3-N and Aq pH/Dissolved Oxigen/Water temperature below are this treatment’s own Water-outlet (post-plant-bed) figures from Table 2 (p.3), per SCHEMA.md’s instruction to take the plant-bed/hydroponic-unit compartment value; see WARN-CHECK below re: NH3/NO2/NO3 species basis, and NO COLUMN below for the pre-plant-bed (fish-effluent, water-inlet) compartment figures. | WARN-MINOR initial fish stocking weight SD: Methods (p.2) states juvenile catfish ‘weight = 9.97 +/- 0.21 g; length = 12.07 +/- 1.92 cm’. Results (p.3) restates fish ‘gained 655.89 +/- 0.98 g … from an initial weight of 9.97 +/- 0.55 g’. Mean identical (9.97 g) in both places; only the SD differs (0.21 vs 0.55). No interpretive impact — Fish size initial recorded as 9.97 +/- 0.21 (Methods, the original stocking description); the Results-restated 0.55 is given here as the alternate. | WARN-MINOR Table 2 caption significance threshold: caption reads ‘Mean in the same row with different superscripts differ significantly (P < 0.0)’ (p.3) — almost certainly a typo for ‘P < 0.05’, matching the Methods statistics statement and the Fig. 2/Fig. 7 captions (‘P <= 0.05’) used consistently elsewhere in the same paper. No cell affected (Statistically analysed = Y regardless; the stated P<0.05 threshold from Methods is used). | WARN-CHECK ammonia/nitrite/nitrate species basis unclear, Table 2 (p.3): rows are labelled simply ‘NH3 (ppm)’, ‘NO2 (mg l-1)’ and ‘NO3 (ppm)’, with no statement of whether these are nitrogen-equivalent species (NH3-N / NO2-N / NO3-N, the basis implied by this schema’s TAN/NH4-N, NO2-N, NO3-N column definitions) or the whole-ion/compound concentration. The digital multi-parameter tester used (Hanna HL 98126, p.2) is not independently documented here as reporting one basis or the other. The paper’s own ppm figures are entered directly into the TAN/NH4-N, NO2-N and NO3-N columns below (ppm treated as numerically equivalent to mg/L for a dilute aqueous solution, which introduces no conversion factor), but the N-only-vs-whole-ion basis is unconfirmed and could differ by a factor of several fold depending on which convention the instrument uses (cf. SCHEMA.md’s NO3-vs-NO3-N example, factor 4.43). Added to REVIEW.md by the batch merge step. | UNIT CONVERSION ONLY: Water recycle — Table 1 (p.2) states ‘Hydraulic loading rate: 7.5 L/hr’ -> 7.5 / 60 = 0.125 L/min. | UNIT CONVERSION ONLY (approximate): Fish trial duration — Results (p.3) states fish were reared and monitored for ‘the four months of culture’; no exact day count is given anywhere in the paper. Recorded ~120 (4 x 30 = 120, assuming a 30-day month — unlike weeks-to-days, a month-to-days conversion is not exact, so this figure carries acknowledged imprecision the authors did not state). | NOT DERIVED, left NR: Initial Stock density (50 fish per 200 L tank and a mean initial weight of 9.97 g would allow back-calculating a density of roughly 2.5 kg/m3, but this exact figure is never stated by the paper as a density, and SCHEMA.md explicitly disallows computing kg/m3 from fish count + tank volume); FCR (total feed consumed is never stated as a cumulative figure, only the 5%-of-body-weight daily ration rate — cannot compute without a stated feed total); SGR (not stated, not computed); feed N/P/K composition (Coppens diet given only as 45% CP / 1.5% fiber / 8.2% moisture / 9.5% ash, no N/P/K breakdown); Fish size final (only the per-fish weight GAIN, 655.89 +/- 0.98 g, is stated; final weight = initial + gain would be a derived sum); Total Feed (kg) (ration given only as a %-of-body-weight rate, never as a cumulative kg total); Fish biomass created (kg) (only a per-fish weight-gain figure is given, not a population total; multiplying by 200 fish would be derivation); Fish survival rate (no mortality/survival figure stated anywhere in the paper, despite fish being tracked over 4 months); Feed routine (feeding frequency, e.g. times/day, is not stated — only the 5% ration rate); pHOptimal (Discussion, p.3, cites Tyson et al. 2008’s pH trade-off range as a literature benchmark this study’s own pH fell short of — a secondary comparison, not a target/setpoint adopted by this study’s own design); Plants/m2 (Table 1 gives only whole-system RAS/hydroponic land areas, 12 m2 / 48 m2, with no stated plant density per m2, and it is not stated whether these areas describe one treatment’s footprint or all four combined — no safe division exists); Days Plant after transplant (pumpkin was direct-seeded into troughs, never transplanted; the paper states only a periodic monitoring schedule from week 4 to week 16 post-germination, never a single total days-to-harvest figure); SPAD, Plant height, Leaf count, Plant fresh weight, Plant dry matter, AP (system yield) (Figs. 3-7 give vine length / leaf number / leaf area / branch number / yield only as line or bar charts across weeks, with no accompanying table or text values anywhere in Results — never read a value off a figure, per SCHEMA.md); Tissue nitrate AP (no plant tissue analysis of any kind was performed in this study); FUE AP, WUE (no fertilizer-use or water-use efficiency metric calculated or stated); Water type, Water classification, Fish Category (paper never applies a categorising term to the fish beyond ‘African Catfish Clarias gariepinus juveniles’); Lat/Long (no coordinates stated anywhere in the paper — the NABDA facility is given only as a street/area description, ‘Umar Musa Yar’adua Express, Airport road Lugbe Abuja, Nigeria’; not looked up externally per the prime directive); Average room Temperature (only water temperature is monitored; no separate ambient/greenhouse air temperature is reported); Artificial Lighting, Climate control, pH Buffers, Iron supplemented, Remineralization (none of these are mentioned anywhere in the paper — genuine silence, recorded NR rather than N per the N-vs-NR distinction in SCHEMA.md). | NO COLUMN: Alkalinity and Hardness (Table 2, p.3, ppm) — Fish effluent 66.23 +/- 0.47 / 84.9 +/- 0.66; Water inlet 66.14 +/- 17.9 / 84.2 +/- 1.30; GRV outlet 66.32 +/- 18.0 / 84.2 +/- 1.30; PKS outlet 66.11 +/- 17.9 / 84.01 +/- 1.30; PWS outlet 66.53 +/- 18.0 / 84.50 +/- 1.40; MIXED outlet 66.20 +/- 17.9 / 84.03 +/- 1.30 — no dedicated column exists for either parameter; Table 2 gives no significance letters for these two rows (i.e. no significant differences reported among compartments). | Fish-effluent and water-inlet compartment values (Table 2, p.3) for Temperature/pH/DO/NH3/NO2/NO3 — these are pre-plant-bed compartment values (fish-tank outflow, and post-biofilter/pre-trough inflow), not the plant-bed/hydroponic-unit value used for this row’s Aq pH / Dissolved Oxigen / Water temperature / TAN-NH4-N / NO2-N / NO3-N columns per SCHEMA.md’s compartment rule; both are single pooled figures not broken out by treatment, so cannot be attributed to any one row. Recorded here for reference (same for all 4 rows): Fish effluent — Temp 27.83 +/- 0.20 C, pH 6.80 +/- 0.01, DO 6.68 +/- 0.03 ppm(a), NH3 8.81 +/- 0.11 ppm(a), NO2 0.31 +/- 0.35 mg/L(d), NO3 53.18 +/- 2.40 ppm(d). Water inlet — Temp 28.90 +/- 1.50 C, pH 6.85 +/- 0.10, DO 4.41 +/- 0.60 ppm(b), NH3 6.80 +/- 0.60 ppm(a), NO2 0.66 +/- 0.06 mg/L(a), NO3 122.3 +/- 9.30 ppm(a). | Percentage nitrogenous-compound reduction (Fig. 2, p.3; formula given, p.3: %Reduction = (a-b)/a x 100, a=inlet concentration, b=outlet concentration) — reported only as a bar chart per substrate (GRV/PKS/PWS/MIX) for NH3, NO2, NO3, with significance letter groups but no printed data table or text values anywhere; qualitatively, ammonia %-reduction was highest in MIX and lowest in GRV; nitrite %-reduction was highest in GRV/PKS and lower in PWS/MIX; nitrate %-reduction was highest in MIX. No numeric values extracted (never read off a figure, per SCHEMA.md). | Table 1 system dimensions not otherwise captured: Mechanical filter tank 20 L; Sump tank 250 L; Planting bowls (troughs) 0.045 m3 each, filled to 50% capacity with the treatment’s growth medium; RAS land area occupied 12 m2; Hydroponic land area occupied 48 m2 — not stated whether these two land-area figures describe one treatment’s footprint or the whole 4-treatment facility combined; ambiguous, not resolved. | Feed proximate composition beyond crude protein: Coppens commercial diet, 45% CP, 1.5% fiber, 8.2% moisture, 9.5% ash (Results, p.3). | Growth-media preparation protocol: PKS sourced from a palm-oil milling industry in Imo State; PWS sourced from a fish-waste dumping site in Ogoja, Cross River State; both sorted to remove debris, autoclaved at 100 C for 1 h, rinsed in clean water, and sun-dried for 12 h before placement in troughs (Methods, p.2). | Design assumptions stated by the authors as explicit hypotheses (p.3), not measured results: ‘the number of fish reared and vegetable seedling propagated matches between nutrient input and requirements’; ‘a hydraulics loading rate 7.5 L/hr was sufficient for the hydroponics system of this setup’; ‘daily addition of 5% of the total water in the aquaponics system was adequate to compensate for evaporation and transpiration losses’ (citing Maucieri et al. 2017) — none of these are validated/measured outcomes in this paper, only stated design assumptions. | Secondary/comparative literature figures (#todo, not this paper’s own data): Akinwole and Dauda (2014) — PKS gave better nitrification efficiency than imported polypropylene blocks for aquaculture wastewater treatment, cited as corroborating this paper’s PKS result; Mader (2012) — lettuce grew significantly better in coconut husk media than gravel, cited as a parallel media-comparison finding; Maucieri et al. (2017) — insignificant differences in all water qualities except NH4 across compartments of their aquaponics system, cited as consistent with this paper’s own compartment comparison; Fang et al. (2017) — media-based aquaponics with varying aeration levels, cited re: low ammonia/NO2 as evidence of a well-developed microbial community; Wahyuningsih et al. (2015) — nitrogen removal in aquaponic recirculation using lettuce, cited as a favourable comparison.

oladimejiEffectsDifferentGrowth2020-T2

Fish

Field	Value
Fish	African catfish (Clarias gariepinus), juveniles
Protein	45
% of body weight	5
Fish size initial	9.97 +/- 0.21
Feed regime	Commercial diet Coppens (45% crude protein, 1.5% fiber, 8.2% moisture, 9.5% ash) fed at 5% of total body weight (Results, p.3)
Fish weight gain	655.89 +/- 0.98
Fish trial duration (days)	~120

Water

Field	Value
Water recycle	0.125
Water volume in the system	800 (0.8 m3); Table 1 gives this as a single generic system figure — assumed to describe one treatment’s complete loop (fish tank+filter+sump+troughs), not summed across all 4 treatments; see Extraction notes
Daily Water exchange rate	5
Aq pH	6.85 +/- 0.50
Dissolved Oxigen	4.18 +/- 0.50
Water temperature	28.61 +/- 1.2
TAN / NH4-N	1.40 +/- 0.50
NO2-N	0.54 +/- 0.03
NO3-N	78.90 +/- 2.58

Plant

Field	Value
Plant	Fluted pumpkin (Telfairia occidentalis)
Details	Pumpkin pods purchased alongside the PKS from Imo State, Nigeria; pods cut to expose seeds, 2 seeds planted per trough; yield/growth data (vine length, leaf number, branch number, bi-weekly plant yield) collected from week 4 post-germination, every 2 weeks thereafter through week 16 (Cornelissen et al. 2003 protocol, p.3); leaf area also tracked (Fig. 4). All growth-curve and yield data reported only in Figs. 3-7 (line/bar charts, Mean +/- SE), no accompanying table or text values.
Plant Category	Leafy vegetable and edible-seed crop / tropical vine (descriptive wording from Introduction, p.1: ‘a tropical vine propagated in West Africa mainly for its leafy vegetables and edible seeds’ — not a formal categorical label applied to the trial design; judgment call, see Extraction notes)

System & Setup

Field	Value
Media Details	Palm Kernel Shell (PKS), agricultural waste from palm-oil milling, Imo State (p.2); autoclaved 100C/1h, rinsed, sun-dried 12h; troughs filled to 50% capacity
Biological system already in use	Y (Fig. 1 depicts a distinct biofilter component (separate from the 20 L mechanical filter tank listed in Table 1) in each treatment’s recirculating loop, between the fish tank/mechanical filter and the sump; nitrifying bacterial action is inferred by the authors from the water NH3->NO2->NO3 concentration trend (p.4); no separate biofilter volume or media stated.)
Air supplement	Y (Air blower and air valve shown in the system diagram (Fig. 1); no aeration rate/flow stated.)
Equipment	Hanna water tester Model HL 98126 (digital multi-parameter water checker, for temperature/pH/DO/alkalinity/hardness/NH3/NO2/NO3); autoclave (100 C, 1 h, substrate sterilization); Minitab 14 statistical software
Control Parameters	Feeding rate 5% of total body weight; hydraulic loading rate 7.5 L/hr (Table 1); stated daily water top-up of 5% of total system water to offset evaporation/transpiration losses (design assumption, citing Maucieri et al. 2017, p.3)
Combination	African catfish (Clarias gariepinus) and fluted pumpkin (Telfairia occidentalis); comparison of 4 grow-bed media (gravel/control, palm kernel shell, periwinkle shell, and a 33.33%-each mix of all three) within independent, media-based aquaponic loops — no hydroponic treatment in this study

Site

Field	Value
Region	Africa
Country	Nigeria

Results & Statistics

Field	Value
Measured Unit	C (temperature); ppm (Dissolved Oxygen, NH3, NO2, NO3 — treated as mg/L equivalent per the paper’s own units, see WARN-CHECK in remarks); % (nitrogenous-compound reduction, Fig. 2, NO COLUMN)
Statistic Details	Descriptive statistics; one-way analysis of variance (ANOVA); Fisher’s least significant difference (LSD) post-hoc when P<0.05; Minitab 14 (Methods, p.3)
Statistically analysed	Y
Replicates (n)	8

Experimental Remarks: TRIAL DEFINITION: T2 = Palm Kernel Shell (PKS) growth medium, an agricultural by-product from palm-oil milling (Imo State), evaluated as an alternative to gravel. Own independent fish tank (A2, 50 catfish) and RAS loop; 8 planting troughs filled to 50% capacity with PKS. Compared against GRV (T1) as the paper’s own within-study baseline; no hydroponic treatment exists in this study, so HYD columns = NA. | No hydroponic control exists anywhere in this study — all four rows are aquaponic treatments compared against each other; Gravel (GRV) is explicitly described by the authors as ‘the control growth media’ (p.2) — i.e. the conventional-substrate baseline against which the two waste-derived media (PKS, PWS) and their mixture are judged — but this is a substrate control, not a hydroponic control, so HYD columns = NA throughout per SCHEMA.md. Design: 4 treatments (Gravel/GRV, Palm Kernel Shell/PKS, Periwinkle Shell/PWS, Mixed substrate/MIXED at 33.33% each of GRV+PKS+PWS) x 8 planting-trough replicates each (32 troughs total, Fig. 1/Table 1); each treatment served by its own independent fish tank (A1-A4, 50 African catfish juveniles/tank, 200 fish total across the study) and its own complete recirculating loop (fish tank -> mechanical filter -> biofilter -> sump -> pump -> planting troughs -> effluent back to sump, Fig. 1). One-way ANOVA + Fisher’s least significant difference post-hoc (P<0.05); Minitab 14 software (Methods, p.3). | TAN/NH4-N, NO2-N, NO3-N and Aq pH/Dissolved Oxigen/Water temperature below are this treatment’s own Water-outlet (post-plant-bed) figures from Table 2 (p.3), per SCHEMA.md’s instruction to take the plant-bed/hydroponic-unit compartment value; see WARN-CHECK below re: NH3/NO2/NO3 species basis, and NO COLUMN below for the pre-plant-bed (fish-effluent, water-inlet) compartment figures. | WARN-MINOR initial fish stocking weight SD: Methods (p.2) states juvenile catfish ‘weight = 9.97 +/- 0.21 g; length = 12.07 +/- 1.92 cm’. Results (p.3) restates fish ‘gained 655.89 +/- 0.98 g … from an initial weight of 9.97 +/- 0.55 g’. Mean identical (9.97 g) in both places; only the SD differs (0.21 vs 0.55). No interpretive impact — Fish size initial recorded as 9.97 +/- 0.21 (Methods, the original stocking description); the Results-restated 0.55 is given here as the alternate. | WARN-MINOR Table 2 caption significance threshold: caption reads ‘Mean in the same row with different superscripts differ significantly (P < 0.0)’ (p.3) — almost certainly a typo for ‘P < 0.05’, matching the Methods statistics statement and the Fig. 2/Fig. 7 captions (‘P <= 0.05’) used consistently elsewhere in the same paper. No cell affected (Statistically analysed = Y regardless; the stated P<0.05 threshold from Methods is used). | WARN-CHECK ammonia/nitrite/nitrate species basis unclear, Table 2 (p.3): rows are labelled simply ‘NH3 (ppm)’, ‘NO2 (mg l-1)’ and ‘NO3 (ppm)’, with no statement of whether these are nitrogen-equivalent species (NH3-N / NO2-N / NO3-N, the basis implied by this schema’s TAN/NH4-N, NO2-N, NO3-N column definitions) or the whole-ion/compound concentration. The digital multi-parameter tester used (Hanna HL 98126, p.2) is not independently documented here as reporting one basis or the other. The paper’s own ppm figures are entered directly into the TAN/NH4-N, NO2-N and NO3-N columns below (ppm treated as numerically equivalent to mg/L for a dilute aqueous solution, which introduces no conversion factor), but the N-only-vs-whole-ion basis is unconfirmed and could differ by a factor of several fold depending on which convention the instrument uses (cf. SCHEMA.md’s NO3-vs-NO3-N example, factor 4.43). Added to REVIEW.md by the batch merge step. | UNIT CONVERSION ONLY: Water recycle — Table 1 (p.2) states ‘Hydraulic loading rate: 7.5 L/hr’ -> 7.5 / 60 = 0.125 L/min. | UNIT CONVERSION ONLY (approximate): Fish trial duration — Results (p.3) states fish were reared and monitored for ‘the four months of culture’; no exact day count is given anywhere in the paper. Recorded ~120 (4 x 30 = 120, assuming a 30-day month — unlike weeks-to-days, a month-to-days conversion is not exact, so this figure carries acknowledged imprecision the authors did not state). | NOT DERIVED, left NR: Initial Stock density (50 fish per 200 L tank and a mean initial weight of 9.97 g would allow back-calculating a density of roughly 2.5 kg/m3, but this exact figure is never stated by the paper as a density, and SCHEMA.md explicitly disallows computing kg/m3 from fish count + tank volume); FCR (total feed consumed is never stated as a cumulative figure, only the 5%-of-body-weight daily ration rate — cannot compute without a stated feed total); SGR (not stated, not computed); feed N/P/K composition (Coppens diet given only as 45% CP / 1.5% fiber / 8.2% moisture / 9.5% ash, no N/P/K breakdown); Fish size final (only the per-fish weight GAIN, 655.89 +/- 0.98 g, is stated; final weight = initial + gain would be a derived sum); Total Feed (kg) (ration given only as a %-of-body-weight rate, never as a cumulative kg total); Fish biomass created (kg) (only a per-fish weight-gain figure is given, not a population total; multiplying by 200 fish would be derivation); Fish survival rate (no mortality/survival figure stated anywhere in the paper, despite fish being tracked over 4 months); Feed routine (feeding frequency, e.g. times/day, is not stated — only the 5% ration rate); pHOptimal (Discussion, p.3, cites Tyson et al. 2008’s pH trade-off range as a literature benchmark this study’s own pH fell short of — a secondary comparison, not a target/setpoint adopted by this study’s own design); Plants/m2 (Table 1 gives only whole-system RAS/hydroponic land areas, 12 m2 / 48 m2, with no stated plant density per m2, and it is not stated whether these areas describe one treatment’s footprint or all four combined — no safe division exists); Days Plant after transplant (pumpkin was direct-seeded into troughs, never transplanted; the paper states only a periodic monitoring schedule from week 4 to week 16 post-germination, never a single total days-to-harvest figure); SPAD, Plant height, Leaf count, Plant fresh weight, Plant dry matter, AP (system yield) (Figs. 3-7 give vine length / leaf number / leaf area / branch number / yield only as line or bar charts across weeks, with no accompanying table or text values anywhere in Results — never read a value off a figure, per SCHEMA.md); Tissue nitrate AP (no plant tissue analysis of any kind was performed in this study); FUE AP, WUE (no fertilizer-use or water-use efficiency metric calculated or stated); Water type, Water classification, Fish Category (paper never applies a categorising term to the fish beyond ‘African Catfish Clarias gariepinus juveniles’); Lat/Long (no coordinates stated anywhere in the paper — the NABDA facility is given only as a street/area description, ‘Umar Musa Yar’adua Express, Airport road Lugbe Abuja, Nigeria’; not looked up externally per the prime directive); Average room Temperature (only water temperature is monitored; no separate ambient/greenhouse air temperature is reported); Artificial Lighting, Climate control, pH Buffers, Iron supplemented, Remineralization (none of these are mentioned anywhere in the paper — genuine silence, recorded NR rather than N per the N-vs-NR distinction in SCHEMA.md). | NO COLUMN: Alkalinity and Hardness (Table 2, p.3, ppm) — Fish effluent 66.23 +/- 0.47 / 84.9 +/- 0.66; Water inlet 66.14 +/- 17.9 / 84.2 +/- 1.30; GRV outlet 66.32 +/- 18.0 / 84.2 +/- 1.30; PKS outlet 66.11 +/- 17.9 / 84.01 +/- 1.30; PWS outlet 66.53 +/- 18.0 / 84.50 +/- 1.40; MIXED outlet 66.20 +/- 17.9 / 84.03 +/- 1.30 — no dedicated column exists for either parameter; Table 2 gives no significance letters for these two rows (i.e. no significant differences reported among compartments). | Fish-effluent and water-inlet compartment values (Table 2, p.3) for Temperature/pH/DO/NH3/NO2/NO3 — these are pre-plant-bed compartment values (fish-tank outflow, and post-biofilter/pre-trough inflow), not the plant-bed/hydroponic-unit value used for this row’s Aq pH / Dissolved Oxigen / Water temperature / TAN-NH4-N / NO2-N / NO3-N columns per SCHEMA.md’s compartment rule; both are single pooled figures not broken out by treatment, so cannot be attributed to any one row. Recorded here for reference (same for all 4 rows): Fish effluent — Temp 27.83 +/- 0.20 C, pH 6.80 +/- 0.01, DO 6.68 +/- 0.03 ppm(a), NH3 8.81 +/- 0.11 ppm(a), NO2 0.31 +/- 0.35 mg/L(d), NO3 53.18 +/- 2.40 ppm(d). Water inlet — Temp 28.90 +/- 1.50 C, pH 6.85 +/- 0.10, DO 4.41 +/- 0.60 ppm(b), NH3 6.80 +/- 0.60 ppm(a), NO2 0.66 +/- 0.06 mg/L(a), NO3 122.3 +/- 9.30 ppm(a). | Percentage nitrogenous-compound reduction (Fig. 2, p.3; formula given, p.3: %Reduction = (a-b)/a x 100, a=inlet concentration, b=outlet concentration) — reported only as a bar chart per substrate (GRV/PKS/PWS/MIX) for NH3, NO2, NO3, with significance letter groups but no printed data table or text values anywhere; qualitatively, ammonia %-reduction was highest in MIX and lowest in GRV; nitrite %-reduction was highest in GRV/PKS and lower in PWS/MIX; nitrate %-reduction was highest in MIX. No numeric values extracted (never read off a figure, per SCHEMA.md). | Table 1 system dimensions not otherwise captured: Mechanical filter tank 20 L; Sump tank 250 L; Planting bowls (troughs) 0.045 m3 each, filled to 50% capacity with the treatment’s growth medium; RAS land area occupied 12 m2; Hydroponic land area occupied 48 m2 — not stated whether these two land-area figures describe one treatment’s footprint or the whole 4-treatment facility combined; ambiguous, not resolved. | Feed proximate composition beyond crude protein: Coppens commercial diet, 45% CP, 1.5% fiber, 8.2% moisture, 9.5% ash (Results, p.3). | Growth-media preparation protocol: PKS sourced from a palm-oil milling industry in Imo State; PWS sourced from a fish-waste dumping site in Ogoja, Cross River State; both sorted to remove debris, autoclaved at 100 C for 1 h, rinsed in clean water, and sun-dried for 12 h before placement in troughs (Methods, p.2). | Design assumptions stated by the authors as explicit hypotheses (p.3), not measured results: ‘the number of fish reared and vegetable seedling propagated matches between nutrient input and requirements’; ‘a hydraulics loading rate 7.5 L/hr was sufficient for the hydroponics system of this setup’; ‘daily addition of 5% of the total water in the aquaponics system was adequate to compensate for evaporation and transpiration losses’ (citing Maucieri et al. 2017) — none of these are validated/measured outcomes in this paper, only stated design assumptions. | Secondary/comparative literature figures (#todo, not this paper’s own data): Akinwole and Dauda (2014) — PKS gave better nitrification efficiency than imported polypropylene blocks for aquaculture wastewater treatment, cited as corroborating this paper’s PKS result; Mader (2012) — lettuce grew significantly better in coconut husk media than gravel, cited as a parallel media-comparison finding; Maucieri et al. (2017) — insignificant differences in all water qualities except NH4 across compartments of their aquaponics system, cited as consistent with this paper’s own compartment comparison; Fang et al. (2017) — media-based aquaponics with varying aeration levels, cited re: low ammonia/NO2 as evidence of a well-developed microbial community; Wahyuningsih et al. (2015) — nitrogen removal in aquaponic recirculation using lettuce, cited as a favourable comparison.

oladimejiEffectsDifferentGrowth2020-T3

Fish

Field	Value
Fish	African catfish (Clarias gariepinus), juveniles
Protein	45
% of body weight	5
Fish size initial	9.97 +/- 0.21
Feed regime	Commercial diet Coppens (45% crude protein, 1.5% fiber, 8.2% moisture, 9.5% ash) fed at 5% of total body weight (Results, p.3)
Fish weight gain	655.89 +/- 0.98
Fish trial duration (days)	~120

Water

Field	Value
Water recycle	0.125
Water volume in the system	800 (0.8 m3); Table 1 gives this as a single generic system figure — assumed to describe one treatment’s complete loop (fish tank+filter+sump+troughs), not summed across all 4 treatments; see Extraction notes
Daily Water exchange rate	5
Aq pH	6.94 +/- 0.51
Dissolved Oxigen	4.38 +/- 0.45
Water temperature	28.69 +/- 1.50
TAN / NH4-N	1.53 +/- 0.33
NO2-N	0.59 +/- 0.04
NO3-N	80.31 +/- 3.24

Plant

Field	Value
Plant	Fluted pumpkin (Telfairia occidentalis)
Details	Pumpkin pods purchased alongside the PKS from Imo State, Nigeria; pods cut to expose seeds, 2 seeds planted per trough; yield/growth data (vine length, leaf number, branch number, bi-weekly plant yield) collected from week 4 post-germination, every 2 weeks thereafter through week 16 (Cornelissen et al. 2003 protocol, p.3); leaf area also tracked (Fig. 4). All growth-curve and yield data reported only in Figs. 3-7 (line/bar charts, Mean +/- SE), no accompanying table or text values.
Plant Category	Leafy vegetable and edible-seed crop / tropical vine (descriptive wording from Introduction, p.1: ‘a tropical vine propagated in West Africa mainly for its leafy vegetables and edible seeds’ — not a formal categorical label applied to the trial design; judgment call, see Extraction notes)

System & Setup

Field	Value
Media Details	Periwinkle Shell (PWS), fish-waste by-product, Ogoja, Cross River State (p.2); autoclaved 100C/1h, rinsed, sun-dried 12h; troughs filled to 50% capacity
Biological system already in use	Y (Fig. 1 depicts a distinct biofilter component (separate from the 20 L mechanical filter tank listed in Table 1) in each treatment’s recirculating loop, between the fish tank/mechanical filter and the sump; nitrifying bacterial action is inferred by the authors from the water NH3->NO2->NO3 concentration trend (p.4); no separate biofilter volume or media stated.)
Air supplement	Y (Air blower and air valve shown in the system diagram (Fig. 1); no aeration rate/flow stated.)
Equipment	Hanna water tester Model HL 98126 (digital multi-parameter water checker, for temperature/pH/DO/alkalinity/hardness/NH3/NO2/NO3); autoclave (100 C, 1 h, substrate sterilization); Minitab 14 statistical software
Control Parameters	Feeding rate 5% of total body weight; hydraulic loading rate 7.5 L/hr (Table 1); stated daily water top-up of 5% of total system water to offset evaporation/transpiration losses (design assumption, citing Maucieri et al. 2017, p.3)
Combination	African catfish (Clarias gariepinus) and fluted pumpkin (Telfairia occidentalis); comparison of 4 grow-bed media (gravel/control, palm kernel shell, periwinkle shell, and a 33.33%-each mix of all three) within independent, media-based aquaponic loops — no hydroponic treatment in this study

Site

Field	Value
Region	Africa
Country	Nigeria

Results & Statistics

Field	Value
Measured Unit	C (temperature); ppm (Dissolved Oxygen, NH3, NO2, NO3 — treated as mg/L equivalent per the paper’s own units, see WARN-CHECK in remarks); % (nitrogenous-compound reduction, Fig. 2, NO COLUMN)
Statistic Details	Descriptive statistics; one-way analysis of variance (ANOVA); Fisher’s least significant difference (LSD) post-hoc when P<0.05; Minitab 14 (Methods, p.3)
Statistically analysed	Y
Replicates (n)	8

Experimental Remarks: TRIAL DEFINITION: T3 = Periwinkle Shell (PWS) growth medium, a fish-waste by-product sourced from a fish-waste dumping site (Ogoja, Cross River State), evaluated as an alternative to gravel. Own independent fish tank (A3, 50 catfish) and RAS loop; 8 planting troughs filled to 50% capacity with PWS. Compared against GRV (T1) as the paper’s own within-study baseline; no hydroponic treatment exists in this study, so HYD columns = NA. | No hydroponic control exists anywhere in this study — all four rows are aquaponic treatments compared against each other; Gravel (GRV) is explicitly described by the authors as ‘the control growth media’ (p.2) — i.e. the conventional-substrate baseline against which the two waste-derived media (PKS, PWS) and their mixture are judged — but this is a substrate control, not a hydroponic control, so HYD columns = NA throughout per SCHEMA.md. Design: 4 treatments (Gravel/GRV, Palm Kernel Shell/PKS, Periwinkle Shell/PWS, Mixed substrate/MIXED at 33.33% each of GRV+PKS+PWS) x 8 planting-trough replicates each (32 troughs total, Fig. 1/Table 1); each treatment served by its own independent fish tank (A1-A4, 50 African catfish juveniles/tank, 200 fish total across the study) and its own complete recirculating loop (fish tank -> mechanical filter -> biofilter -> sump -> pump -> planting troughs -> effluent back to sump, Fig. 1). One-way ANOVA + Fisher’s least significant difference post-hoc (P<0.05); Minitab 14 software (Methods, p.3). | TAN/NH4-N, NO2-N, NO3-N and Aq pH/Dissolved Oxigen/Water temperature below are this treatment’s own Water-outlet (post-plant-bed) figures from Table 2 (p.3), per SCHEMA.md’s instruction to take the plant-bed/hydroponic-unit compartment value; see WARN-CHECK below re: NH3/NO2/NO3 species basis, and NO COLUMN below for the pre-plant-bed (fish-effluent, water-inlet) compartment figures. | WARN-MINOR initial fish stocking weight SD: Methods (p.2) states juvenile catfish ‘weight = 9.97 +/- 0.21 g; length = 12.07 +/- 1.92 cm’. Results (p.3) restates fish ‘gained 655.89 +/- 0.98 g … from an initial weight of 9.97 +/- 0.55 g’. Mean identical (9.97 g) in both places; only the SD differs (0.21 vs 0.55). No interpretive impact — Fish size initial recorded as 9.97 +/- 0.21 (Methods, the original stocking description); the Results-restated 0.55 is given here as the alternate. | WARN-MINOR Table 2 caption significance threshold: caption reads ‘Mean in the same row with different superscripts differ significantly (P < 0.0)’ (p.3) — almost certainly a typo for ‘P < 0.05’, matching the Methods statistics statement and the Fig. 2/Fig. 7 captions (‘P <= 0.05’) used consistently elsewhere in the same paper. No cell affected (Statistically analysed = Y regardless; the stated P<0.05 threshold from Methods is used). | WARN-CHECK ammonia/nitrite/nitrate species basis unclear, Table 2 (p.3): rows are labelled simply ‘NH3 (ppm)’, ‘NO2 (mg l-1)’ and ‘NO3 (ppm)’, with no statement of whether these are nitrogen-equivalent species (NH3-N / NO2-N / NO3-N, the basis implied by this schema’s TAN/NH4-N, NO2-N, NO3-N column definitions) or the whole-ion/compound concentration. The digital multi-parameter tester used (Hanna HL 98126, p.2) is not independently documented here as reporting one basis or the other. The paper’s own ppm figures are entered directly into the TAN/NH4-N, NO2-N and NO3-N columns below (ppm treated as numerically equivalent to mg/L for a dilute aqueous solution, which introduces no conversion factor), but the N-only-vs-whole-ion basis is unconfirmed and could differ by a factor of several fold depending on which convention the instrument uses (cf. SCHEMA.md’s NO3-vs-NO3-N example, factor 4.43). Added to REVIEW.md by the batch merge step. | UNIT CONVERSION ONLY: Water recycle — Table 1 (p.2) states ‘Hydraulic loading rate: 7.5 L/hr’ -> 7.5 / 60 = 0.125 L/min. | UNIT CONVERSION ONLY (approximate): Fish trial duration — Results (p.3) states fish were reared and monitored for ‘the four months of culture’; no exact day count is given anywhere in the paper. Recorded ~120 (4 x 30 = 120, assuming a 30-day month — unlike weeks-to-days, a month-to-days conversion is not exact, so this figure carries acknowledged imprecision the authors did not state). | NOT DERIVED, left NR: Initial Stock density (50 fish per 200 L tank and a mean initial weight of 9.97 g would allow back-calculating a density of roughly 2.5 kg/m3, but this exact figure is never stated by the paper as a density, and SCHEMA.md explicitly disallows computing kg/m3 from fish count + tank volume); FCR (total feed consumed is never stated as a cumulative figure, only the 5%-of-body-weight daily ration rate — cannot compute without a stated feed total); SGR (not stated, not computed); feed N/P/K composition (Coppens diet given only as 45% CP / 1.5% fiber / 8.2% moisture / 9.5% ash, no N/P/K breakdown); Fish size final (only the per-fish weight GAIN, 655.89 +/- 0.98 g, is stated; final weight = initial + gain would be a derived sum); Total Feed (kg) (ration given only as a %-of-body-weight rate, never as a cumulative kg total); Fish biomass created (kg) (only a per-fish weight-gain figure is given, not a population total; multiplying by 200 fish would be derivation); Fish survival rate (no mortality/survival figure stated anywhere in the paper, despite fish being tracked over 4 months); Feed routine (feeding frequency, e.g. times/day, is not stated — only the 5% ration rate); pHOptimal (Discussion, p.3, cites Tyson et al. 2008’s pH trade-off range as a literature benchmark this study’s own pH fell short of — a secondary comparison, not a target/setpoint adopted by this study’s own design); Plants/m2 (Table 1 gives only whole-system RAS/hydroponic land areas, 12 m2 / 48 m2, with no stated plant density per m2, and it is not stated whether these areas describe one treatment’s footprint or all four combined — no safe division exists); Days Plant after transplant (pumpkin was direct-seeded into troughs, never transplanted; the paper states only a periodic monitoring schedule from week 4 to week 16 post-germination, never a single total days-to-harvest figure); SPAD, Plant height, Leaf count, Plant fresh weight, Plant dry matter, AP (system yield) (Figs. 3-7 give vine length / leaf number / leaf area / branch number / yield only as line or bar charts across weeks, with no accompanying table or text values anywhere in Results — never read a value off a figure, per SCHEMA.md); Tissue nitrate AP (no plant tissue analysis of any kind was performed in this study); FUE AP, WUE (no fertilizer-use or water-use efficiency metric calculated or stated); Water type, Water classification, Fish Category (paper never applies a categorising term to the fish beyond ‘African Catfish Clarias gariepinus juveniles’); Lat/Long (no coordinates stated anywhere in the paper — the NABDA facility is given only as a street/area description, ‘Umar Musa Yar’adua Express, Airport road Lugbe Abuja, Nigeria’; not looked up externally per the prime directive); Average room Temperature (only water temperature is monitored; no separate ambient/greenhouse air temperature is reported); Artificial Lighting, Climate control, pH Buffers, Iron supplemented, Remineralization (none of these are mentioned anywhere in the paper — genuine silence, recorded NR rather than N per the N-vs-NR distinction in SCHEMA.md). | NO COLUMN: Alkalinity and Hardness (Table 2, p.3, ppm) — Fish effluent 66.23 +/- 0.47 / 84.9 +/- 0.66; Water inlet 66.14 +/- 17.9 / 84.2 +/- 1.30; GRV outlet 66.32 +/- 18.0 / 84.2 +/- 1.30; PKS outlet 66.11 +/- 17.9 / 84.01 +/- 1.30; PWS outlet 66.53 +/- 18.0 / 84.50 +/- 1.40; MIXED outlet 66.20 +/- 17.9 / 84.03 +/- 1.30 — no dedicated column exists for either parameter; Table 2 gives no significance letters for these two rows (i.e. no significant differences reported among compartments). | Fish-effluent and water-inlet compartment values (Table 2, p.3) for Temperature/pH/DO/NH3/NO2/NO3 — these are pre-plant-bed compartment values (fish-tank outflow, and post-biofilter/pre-trough inflow), not the plant-bed/hydroponic-unit value used for this row’s Aq pH / Dissolved Oxigen / Water temperature / TAN-NH4-N / NO2-N / NO3-N columns per SCHEMA.md’s compartment rule; both are single pooled figures not broken out by treatment, so cannot be attributed to any one row. Recorded here for reference (same for all 4 rows): Fish effluent — Temp 27.83 +/- 0.20 C, pH 6.80 +/- 0.01, DO 6.68 +/- 0.03 ppm(a), NH3 8.81 +/- 0.11 ppm(a), NO2 0.31 +/- 0.35 mg/L(d), NO3 53.18 +/- 2.40 ppm(d). Water inlet — Temp 28.90 +/- 1.50 C, pH 6.85 +/- 0.10, DO 4.41 +/- 0.60 ppm(b), NH3 6.80 +/- 0.60 ppm(a), NO2 0.66 +/- 0.06 mg/L(a), NO3 122.3 +/- 9.30 ppm(a). | Percentage nitrogenous-compound reduction (Fig. 2, p.3; formula given, p.3: %Reduction = (a-b)/a x 100, a=inlet concentration, b=outlet concentration) — reported only as a bar chart per substrate (GRV/PKS/PWS/MIX) for NH3, NO2, NO3, with significance letter groups but no printed data table or text values anywhere; qualitatively, ammonia %-reduction was highest in MIX and lowest in GRV; nitrite %-reduction was highest in GRV/PKS and lower in PWS/MIX; nitrate %-reduction was highest in MIX. No numeric values extracted (never read off a figure, per SCHEMA.md). | Table 1 system dimensions not otherwise captured: Mechanical filter tank 20 L; Sump tank 250 L; Planting bowls (troughs) 0.045 m3 each, filled to 50% capacity with the treatment’s growth medium; RAS land area occupied 12 m2; Hydroponic land area occupied 48 m2 — not stated whether these two land-area figures describe one treatment’s footprint or the whole 4-treatment facility combined; ambiguous, not resolved. | Feed proximate composition beyond crude protein: Coppens commercial diet, 45% CP, 1.5% fiber, 8.2% moisture, 9.5% ash (Results, p.3). | Growth-media preparation protocol: PKS sourced from a palm-oil milling industry in Imo State; PWS sourced from a fish-waste dumping site in Ogoja, Cross River State; both sorted to remove debris, autoclaved at 100 C for 1 h, rinsed in clean water, and sun-dried for 12 h before placement in troughs (Methods, p.2). | Design assumptions stated by the authors as explicit hypotheses (p.3), not measured results: ‘the number of fish reared and vegetable seedling propagated matches between nutrient input and requirements’; ‘a hydraulics loading rate 7.5 L/hr was sufficient for the hydroponics system of this setup’; ‘daily addition of 5% of the total water in the aquaponics system was adequate to compensate for evaporation and transpiration losses’ (citing Maucieri et al. 2017) — none of these are validated/measured outcomes in this paper, only stated design assumptions. | Secondary/comparative literature figures (#todo, not this paper’s own data): Akinwole and Dauda (2014) — PKS gave better nitrification efficiency than imported polypropylene blocks for aquaculture wastewater treatment, cited as corroborating this paper’s PKS result; Mader (2012) — lettuce grew significantly better in coconut husk media than gravel, cited as a parallel media-comparison finding; Maucieri et al. (2017) — insignificant differences in all water qualities except NH4 across compartments of their aquaponics system, cited as consistent with this paper’s own compartment comparison; Fang et al. (2017) — media-based aquaponics with varying aeration levels, cited re: low ammonia/NO2 as evidence of a well-developed microbial community; Wahyuningsih et al. (2015) — nitrogen removal in aquaponic recirculation using lettuce, cited as a favourable comparison.

oladimejiEffectsDifferentGrowth2020-T4

Fish

Field	Value
Fish	African catfish (Clarias gariepinus), juveniles
Protein	45
% of body weight	5
Fish size initial	9.97 +/- 0.21
Feed regime	Commercial diet Coppens (45% crude protein, 1.5% fiber, 8.2% moisture, 9.5% ash) fed at 5% of total body weight (Results, p.3)
Fish weight gain	655.89 +/- 0.98
Fish trial duration (days)	~120

Water

Field	Value
Water recycle	0.125
Water volume in the system	800 (0.8 m3); Table 1 gives this as a single generic system figure — assumed to describe one treatment’s complete loop (fish tank+filter+sump+troughs), not summed across all 4 treatments; see Extraction notes
Daily Water exchange rate	5
Aq pH	6.92 +/- 0.30
Dissolved Oxigen	4.65 +/- 0.32
Water temperature	28.35 +/- 1.90
TAN / NH4-N	0.82 +/- 0.09
NO2-N	0.58 +/- 0.04
NO3-N	71.5 +/- 4.72

Plant

Field	Value
Plant	Fluted pumpkin (Telfairia occidentalis)
Details	Pumpkin pods purchased alongside the PKS from Imo State, Nigeria; pods cut to expose seeds, 2 seeds planted per trough; yield/growth data (vine length, leaf number, branch number, bi-weekly plant yield) collected from week 4 post-germination, every 2 weeks thereafter through week 16 (Cornelissen et al. 2003 protocol, p.3); leaf area also tracked (Fig. 4). All growth-curve and yield data reported only in Figs. 3-7 (line/bar charts, Mean +/- SE), no accompanying table or text values.
Plant Category	Leafy vegetable and edible-seed crop / tropical vine (descriptive wording from Introduction, p.1: ‘a tropical vine propagated in West Africa mainly for its leafy vegetables and edible seeds’ — not a formal categorical label applied to the trial design; judgment call, see Extraction notes)

System & Setup

Field	Value
Media Details	Mixed substrate: Gravel + Palm Kernel Shell + Periwinkle Shell, 33.33% each by volume (p.2); troughs filled to 50% capacity
Biological system already in use	Y (Fig. 1 depicts a distinct biofilter component (separate from the 20 L mechanical filter tank listed in Table 1) in each treatment’s recirculating loop, between the fish tank/mechanical filter and the sump; nitrifying bacterial action is inferred by the authors from the water NH3->NO2->NO3 concentration trend (p.4); no separate biofilter volume or media stated.)
Air supplement	Y (Air blower and air valve shown in the system diagram (Fig. 1); no aeration rate/flow stated.)
Equipment	Hanna water tester Model HL 98126 (digital multi-parameter water checker, for temperature/pH/DO/alkalinity/hardness/NH3/NO2/NO3); autoclave (100 C, 1 h, substrate sterilization); Minitab 14 statistical software
Control Parameters	Feeding rate 5% of total body weight; hydraulic loading rate 7.5 L/hr (Table 1); stated daily water top-up of 5% of total system water to offset evaporation/transpiration losses (design assumption, citing Maucieri et al. 2017, p.3)
Combination	African catfish (Clarias gariepinus) and fluted pumpkin (Telfairia occidentalis); comparison of 4 grow-bed media (gravel/control, palm kernel shell, periwinkle shell, and a 33.33%-each mix of all three) within independent, media-based aquaponic loops — no hydroponic treatment in this study

Site

Field	Value
Region	Africa
Country	Nigeria

Results & Statistics

Field	Value
Measured Unit	C (temperature); ppm (Dissolved Oxygen, NH3, NO2, NO3 — treated as mg/L equivalent per the paper’s own units, see WARN-CHECK in remarks); % (nitrogenous-compound reduction, Fig. 2, NO COLUMN)
Statistic Details	Descriptive statistics; one-way analysis of variance (ANOVA); Fisher’s least significant difference (LSD) post-hoc when P<0.05; Minitab 14 (Methods, p.3)
Statistically analysed	Y
Replicates (n)	8

Experimental Remarks: TRIAL DEFINITION: T4 = Mixed substrate, 33.33% each of Gravel + Palm Kernel Shell + Periwinkle Shell by volume. Own independent fish tank (A4, 50 catfish) and RAS loop; 8 planting troughs filled to 50% capacity with the mixed medium. Compared against GRV (T1) as the paper’s own within-study baseline; no hydroponic treatment exists in this study, so HYD columns = NA. | No hydroponic control exists anywhere in this study — all four rows are aquaponic treatments compared against each other; Gravel (GRV) is explicitly described by the authors as ‘the control growth media’ (p.2) — i.e. the conventional-substrate baseline against which the two waste-derived media (PKS, PWS) and their mixture are judged — but this is a substrate control, not a hydroponic control, so HYD columns = NA throughout per SCHEMA.md. Design: 4 treatments (Gravel/GRV, Palm Kernel Shell/PKS, Periwinkle Shell/PWS, Mixed substrate/MIXED at 33.33% each of GRV+PKS+PWS) x 8 planting-trough replicates each (32 troughs total, Fig. 1/Table 1); each treatment served by its own independent fish tank (A1-A4, 50 African catfish juveniles/tank, 200 fish total across the study) and its own complete recirculating loop (fish tank -> mechanical filter -> biofilter -> sump -> pump -> planting troughs -> effluent back to sump, Fig. 1). One-way ANOVA + Fisher’s least significant difference post-hoc (P<0.05); Minitab 14 software (Methods, p.3). | TAN/NH4-N, NO2-N, NO3-N and Aq pH/Dissolved Oxigen/Water temperature below are this treatment’s own Water-outlet (post-plant-bed) figures from Table 2 (p.3), per SCHEMA.md’s instruction to take the plant-bed/hydroponic-unit compartment value; see WARN-CHECK below re: NH3/NO2/NO3 species basis, and NO COLUMN below for the pre-plant-bed (fish-effluent, water-inlet) compartment figures. | WARN-MINOR initial fish stocking weight SD: Methods (p.2) states juvenile catfish ‘weight = 9.97 +/- 0.21 g; length = 12.07 +/- 1.92 cm’. Results (p.3) restates fish ‘gained 655.89 +/- 0.98 g … from an initial weight of 9.97 +/- 0.55 g’. Mean identical (9.97 g) in both places; only the SD differs (0.21 vs 0.55). No interpretive impact — Fish size initial recorded as 9.97 +/- 0.21 (Methods, the original stocking description); the Results-restated 0.55 is given here as the alternate. | WARN-MINOR Table 2 caption significance threshold: caption reads ‘Mean in the same row with different superscripts differ significantly (P < 0.0)’ (p.3) — almost certainly a typo for ‘P < 0.05’, matching the Methods statistics statement and the Fig. 2/Fig. 7 captions (‘P <= 0.05’) used consistently elsewhere in the same paper. No cell affected (Statistically analysed = Y regardless; the stated P<0.05 threshold from Methods is used). | WARN-CHECK ammonia/nitrite/nitrate species basis unclear, Table 2 (p.3): rows are labelled simply ‘NH3 (ppm)’, ‘NO2 (mg l-1)’ and ‘NO3 (ppm)’, with no statement of whether these are nitrogen-equivalent species (NH3-N / NO2-N / NO3-N, the basis implied by this schema’s TAN/NH4-N, NO2-N, NO3-N column definitions) or the whole-ion/compound concentration. The digital multi-parameter tester used (Hanna HL 98126, p.2) is not independently documented here as reporting one basis or the other. The paper’s own ppm figures are entered directly into the TAN/NH4-N, NO2-N and NO3-N columns below (ppm treated as numerically equivalent to mg/L for a dilute aqueous solution, which introduces no conversion factor), but the N-only-vs-whole-ion basis is unconfirmed and could differ by a factor of several fold depending on which convention the instrument uses (cf. SCHEMA.md’s NO3-vs-NO3-N example, factor 4.43). Added to REVIEW.md by the batch merge step. | UNIT CONVERSION ONLY: Water recycle — Table 1 (p.2) states ‘Hydraulic loading rate: 7.5 L/hr’ -> 7.5 / 60 = 0.125 L/min. | UNIT CONVERSION ONLY (approximate): Fish trial duration — Results (p.3) states fish were reared and monitored for ‘the four months of culture’; no exact day count is given anywhere in the paper. Recorded ~120 (4 x 30 = 120, assuming a 30-day month — unlike weeks-to-days, a month-to-days conversion is not exact, so this figure carries acknowledged imprecision the authors did not state). | NOT DERIVED, left NR: Initial Stock density (50 fish per 200 L tank and a mean initial weight of 9.97 g would allow back-calculating a density of roughly 2.5 kg/m3, but this exact figure is never stated by the paper as a density, and SCHEMA.md explicitly disallows computing kg/m3 from fish count + tank volume); FCR (total feed consumed is never stated as a cumulative figure, only the 5%-of-body-weight daily ration rate — cannot compute without a stated feed total); SGR (not stated, not computed); feed N/P/K composition (Coppens diet given only as 45% CP / 1.5% fiber / 8.2% moisture / 9.5% ash, no N/P/K breakdown); Fish size final (only the per-fish weight GAIN, 655.89 +/- 0.98 g, is stated; final weight = initial + gain would be a derived sum); Total Feed (kg) (ration given only as a %-of-body-weight rate, never as a cumulative kg total); Fish biomass created (kg) (only a per-fish weight-gain figure is given, not a population total; multiplying by 200 fish would be derivation); Fish survival rate (no mortality/survival figure stated anywhere in the paper, despite fish being tracked over 4 months); Feed routine (feeding frequency, e.g. times/day, is not stated — only the 5% ration rate); pHOptimal (Discussion, p.3, cites Tyson et al. 2008’s pH trade-off range as a literature benchmark this study’s own pH fell short of — a secondary comparison, not a target/setpoint adopted by this study’s own design); Plants/m2 (Table 1 gives only whole-system RAS/hydroponic land areas, 12 m2 / 48 m2, with no stated plant density per m2, and it is not stated whether these areas describe one treatment’s footprint or all four combined — no safe division exists); Days Plant after transplant (pumpkin was direct-seeded into troughs, never transplanted; the paper states only a periodic monitoring schedule from week 4 to week 16 post-germination, never a single total days-to-harvest figure); SPAD, Plant height, Leaf count, Plant fresh weight, Plant dry matter, AP (system yield) (Figs. 3-7 give vine length / leaf number / leaf area / branch number / yield only as line or bar charts across weeks, with no accompanying table or text values anywhere in Results — never read a value off a figure, per SCHEMA.md); Tissue nitrate AP (no plant tissue analysis of any kind was performed in this study); FUE AP, WUE (no fertilizer-use or water-use efficiency metric calculated or stated); Water type, Water classification, Fish Category (paper never applies a categorising term to the fish beyond ‘African Catfish Clarias gariepinus juveniles’); Lat/Long (no coordinates stated anywhere in the paper — the NABDA facility is given only as a street/area description, ‘Umar Musa Yar’adua Express, Airport road Lugbe Abuja, Nigeria’; not looked up externally per the prime directive); Average room Temperature (only water temperature is monitored; no separate ambient/greenhouse air temperature is reported); Artificial Lighting, Climate control, pH Buffers, Iron supplemented, Remineralization (none of these are mentioned anywhere in the paper — genuine silence, recorded NR rather than N per the N-vs-NR distinction in SCHEMA.md). | NO COLUMN: Alkalinity and Hardness (Table 2, p.3, ppm) — Fish effluent 66.23 +/- 0.47 / 84.9 +/- 0.66; Water inlet 66.14 +/- 17.9 / 84.2 +/- 1.30; GRV outlet 66.32 +/- 18.0 / 84.2 +/- 1.30; PKS outlet 66.11 +/- 17.9 / 84.01 +/- 1.30; PWS outlet 66.53 +/- 18.0 / 84.50 +/- 1.40; MIXED outlet 66.20 +/- 17.9 / 84.03 +/- 1.30 — no dedicated column exists for either parameter; Table 2 gives no significance letters for these two rows (i.e. no significant differences reported among compartments). | Fish-effluent and water-inlet compartment values (Table 2, p.3) for Temperature/pH/DO/NH3/NO2/NO3 — these are pre-plant-bed compartment values (fish-tank outflow, and post-biofilter/pre-trough inflow), not the plant-bed/hydroponic-unit value used for this row’s Aq pH / Dissolved Oxigen / Water temperature / TAN-NH4-N / NO2-N / NO3-N columns per SCHEMA.md’s compartment rule; both are single pooled figures not broken out by treatment, so cannot be attributed to any one row. Recorded here for reference (same for all 4 rows): Fish effluent — Temp 27.83 +/- 0.20 C, pH 6.80 +/- 0.01, DO 6.68 +/- 0.03 ppm(a), NH3 8.81 +/- 0.11 ppm(a), NO2 0.31 +/- 0.35 mg/L(d), NO3 53.18 +/- 2.40 ppm(d). Water inlet — Temp 28.90 +/- 1.50 C, pH 6.85 +/- 0.10, DO 4.41 +/- 0.60 ppm(b), NH3 6.80 +/- 0.60 ppm(a), NO2 0.66 +/- 0.06 mg/L(a), NO3 122.3 +/- 9.30 ppm(a). | Percentage nitrogenous-compound reduction (Fig. 2, p.3; formula given, p.3: %Reduction = (a-b)/a x 100, a=inlet concentration, b=outlet concentration) — reported only as a bar chart per substrate (GRV/PKS/PWS/MIX) for NH3, NO2, NO3, with significance letter groups but no printed data table or text values anywhere; qualitatively, ammonia %-reduction was highest in MIX and lowest in GRV; nitrite %-reduction was highest in GRV/PKS and lower in PWS/MIX; nitrate %-reduction was highest in MIX. No numeric values extracted (never read off a figure, per SCHEMA.md). | Table 1 system dimensions not otherwise captured: Mechanical filter tank 20 L; Sump tank 250 L; Planting bowls (troughs) 0.045 m3 each, filled to 50% capacity with the treatment’s growth medium; RAS land area occupied 12 m2; Hydroponic land area occupied 48 m2 — not stated whether these two land-area figures describe one treatment’s footprint or the whole 4-treatment facility combined; ambiguous, not resolved. | Feed proximate composition beyond crude protein: Coppens commercial diet, 45% CP, 1.5% fiber, 8.2% moisture, 9.5% ash (Results, p.3). | Growth-media preparation protocol: PKS sourced from a palm-oil milling industry in Imo State; PWS sourced from a fish-waste dumping site in Ogoja, Cross River State; both sorted to remove debris, autoclaved at 100 C for 1 h, rinsed in clean water, and sun-dried for 12 h before placement in troughs (Methods, p.2). | Design assumptions stated by the authors as explicit hypotheses (p.3), not measured results: ‘the number of fish reared and vegetable seedling propagated matches between nutrient input and requirements’; ‘a hydraulics loading rate 7.5 L/hr was sufficient for the hydroponics system of this setup’; ‘daily addition of 5% of the total water in the aquaponics system was adequate to compensate for evaporation and transpiration losses’ (citing Maucieri et al. 2017) — none of these are validated/measured outcomes in this paper, only stated design assumptions. | Secondary/comparative literature figures (#todo, not this paper’s own data): Akinwole and Dauda (2014) — PKS gave better nitrification efficiency than imported polypropylene blocks for aquaculture wastewater treatment, cited as corroborating this paper’s PKS result; Mader (2012) — lettuce grew significantly better in coconut husk media than gravel, cited as a parallel media-comparison finding; Maucieri et al. (2017) — insignificant differences in all water qualities except NH4 across compartments of their aquaponics system, cited as consistent with this paper’s own compartment comparison; Fang et al. (2017) — media-based aquaponics with varying aeration levels, cited re: low ammonia/NO2 as evidence of a well-developed microbial community; Wahyuningsih et al. (2015) — nitrogen removal in aquaponic recirculation using lettuce, cited as a favourable comparison.

Quartz 5

Explorer

Effects of different growth media on water quality and plant yield in a catfish-pumpkin aquaponics system

Effects of different growth media on water quality and plant yield in a catfish-pumpkin aquaponics system

Metadata

Abstract

Summary

Growth media and plant yield

Water quality and nitrogenous-compound removal

Fish growth

Linked claims

Citations to chase

Extraction notes

Data Tables

Trial Parameters

oladimejiEffectsDifferentGrowth2020-T1

oladimejiEffectsDifferentGrowth2020-T2

oladimejiEffectsDifferentGrowth2020-T3

oladimejiEffectsDifferentGrowth2020-T4

Graph View

Table of Contents

Backlinks