Integrated Aquaponic Co‐Culture of Nile Tilapia and African Catfish: Effects on Plant Yield, Fish Growth, Nutrient Profile, and Heavy Metal Accumulation
Metadata
- Cite key: fidjiIntegratedAquaponicCoCulture2026
- Item type: Journal Article
- Authors: Jude Kifufu Gafidji Fidji, Réné Wamene Baranger, Gaétan Kalala Bolokango, Patrick Mafwila Kinkela, Honoré Kiatoko Mangeye
- Affiliation: Département de Zootechnie, Université de Kinshasa, Faculté des Sciences Agronomiques et Environnement, Kinshasa, Democratic Republic of the Congo (Bolokango, Kinkela, Mangeye, and Fidji as a joint affiliation); Institut Supérieur Pédagogique de Bandundu, Mention Santé et Productions Animales, Bandundu, Democratic Republic of the Congo (Fidji, Baranger)
- Journal: Food Science & Nutrition 14(1) (2026) e71333
- Date: 01/2026
- Date added: 2026-07-15
- DOI: 10.1002/fsn3.71333
- Funding: [not reported]
- URL: https://doi.org/10.1002/fsn3.71333
- PDF:
Fidji et al. - 2026 - Integrated Aquaponic Co‐Culture of Nile Tilapia and African Catfish Effects on Plant Yield, Fish Gr.pdf
Opinion
A clean, well-instrumented 106-day, 5-treatment co-culture trial with real replication language (“each replicate”) and appropriate stats (ANOVA+Tukey, Chi-squared, PCA, GLMM) — but the write-up has an unusual number of internal numeric slips for a paper this length: two irreconcilable Nile tilapia weight-gain figures (TC2, TC3) that recur across narrative text, a table cell, and a second independently-stated daily-growth-rate figure that itself implies yet a third total. None of this touches the paper’s headline conclusion (co-culture ratio doesn’t hurt tilapia, does affect catfish), but it does mean the exact TC2/TC3 tilapia growth numbers should not be cited without going back to the authors’ raw data. Also worth flagging: FCR is never reported despite full feed composition being given, and the paper never states a fish tank/system water volume, exchange rate, or replicate count per treatment — all common items that would normally appear in an aquaponics methods section.
Abstract
A 106-day experiment was conducted to evaluate the co-culture of Nile tilapia (Oreochromis niloticus, 20.4 ± 0.29 g) and African catfish (Clarias gariepinus, 19.8 ± 0.46 g) in an aquaponic system using eggplant (Solanum melongena) as vegetable specie. Fish were stocked at an initial biomass of 10 kg·m⁻³ with five different Nile tilapia: African catfish stocking ratios: 2:8 (TC1), 4:6 (TC2), 5:5 (TC3), 6:4 (TC4), and 8:2 (TC5), totalling 100 fish per tank. Results showed no significant differences (p > 0.05) in eggplant yield, Nile tilapia growth performance, or survival rate among treatments. However, African catfish growth performance declined with decreasing proportion of catfish, with a statistically significant difference observed only between the highest (TC1) and lowest (TC5) catfish ratios. The lowest survival rate (90.0%) was recorded in TC5. Proximate analysis revealed no significant variation (p > 0.05) in the nutritional composition of either fish species across treatments, and heavy metal levels in fish flesh remained below toxic thresholds. Water quality parameters remained within optimal ranges for both species and did not significantly influence fish growth. These findings support the feasibility of co-culturing Nile tilapia and African catfish in aquaponic systems without compromising water quality, yield plant, fish performance, or nutritional value.
Summary
Five identical replicate-backed coupled aquaponic units (one per treatment; “ten identical aquaponic units” total across the study) in Bandundu, DRC, grew eggplant on water shared by tanks stocked with a fixed 10 kg/m³ combined biomass of Nile tilapia and African catfish, split across five stocking ratios (2:8 through 8:2, tilapia:catfish) over 106 days. Eggplant fruit yield (count, weight, dimensions, biomass) did not differ significantly across ratios. Nile tilapia growth, survival, and body composition were statistically unaffected by co-culture ratio, but African catfish growth performance declined as their proportion in the tank fell, with the high-catfish (TC1) and low-catfish (TC5) treatments differing significantly by the trial’s end. Fish-flesh proximate composition and seven screened heavy metals (three non-detected: Pb, Cd, As) were similar across treatments and within safety limits for both species. Water quality (pH, DO, temperature, EC, ammonia, nitrite, nitrate) stayed within ranges the authors consider suitable for both fish species and was not a significant driver of the growth differences observed. The authors conclude tilapia–catfish co-culture in aquaponics is feasible at the ratios tested without sacrificing plant yield, fish welfare, or product safety.
Experiment data
- Location: Bandundu, Democratic Republic of the Congo (p.2)
- Design: 5 treatments (TC1–TC5), each a distinct Nile tilapia:African catfish stocking ratio (2:8, 4:6, 5:5, 6:4, 8:2) at a fixed total initial biomass of 10 kg/m³ and 100 fish/tank; “ten identical aquaponic units were used” (p.2), implying replication, but the per-treatment replicate count is never stated as a number
- Replicates / n: UNCLEAR — not stated as a number (see Extraction notes)
- Duration: 106 days (8 July–22 October 2024), fish sampled every 26 days; eggplant harvest began day 56 and ran weekly for 7 weeks to trial end
- Organisms: Nile Tilapia (Oreochromis niloticus) / African Catfish (Clarias gariepinus) / Eggplant (Solanum melongena)
- Statistics: Levene’s test, one-way ANOVA + Tukey’s HSD, Chi-squared (survival), PCA, GLMM (Sabwa et al. 2021 approach); R v4.3.2 + Excel/XLSTAT; α = 0.05
- Marketable eggplant yield: 22334.4 ± 78.4 to 28052.6 ± 65.9 g/m²/year across TC1–TC5, ns (p > 0.05)
- Nile tilapia survival: 92.5 ± 5.7% (TC1) to 96.8 ± 4.2% (TC5), ns (p = 0.093)
- African catfish survival: 90.0 ± 4.5% (TC5, lowest) to 98.7 ± 2.9% (TC1), ns overall (p = 0.158) but a clear directional trend
- African catfish overall weight gain (0–106d): 47.8 ± 6.8 g (TC1) down to 34.1 ± 7.2 g (TC5), p = 0.001 — significant decline as catfish proportion falls
Eggplant (Solanum melongena) yield
This paper: Number of fruits/plant (7.0–7.9 ± ~1.3–3.6), average fruit weight (51.9–57.2 ± 5.7–8.3 g/plant/week), and marketable yield (22334.4–28052.6 g/m²/year) did not differ significantly across the five tilapia:catfish stocking ratios (p > 0.05 for the annualised totals). Individual weekly slices sometimes reached significance (e.g. Week 7 fruit count p = 0.002) without a consistent directional pattern, which the authors read as normal harvest-to-harvest variation rather than a treatment effect.
Compared with:
- todo Coulibaly 2021 — recorded up to 80 fruits/plant in soil-based eggplant cultivation, well above this paper’s ~7–8 fruits/plant; attributed by the authors to soil vs. soilless systems and to a shorter (2–3 day) harvest interval used by Coulibaly vs. this study’s >5-day interval. (p.14)
- todo Aglinglo et al. 2018 — frequent harvesting at 2–3 day intervals stimulates fruiting; cited as a likely explanation for this paper’s comparatively low per-plant yield. (p.14)
Nile tilapia and African catfish growth performance
This paper: Nile tilapia growth, survival, and body condition were statistically indistinguishable across all five co-culture ratios (GLMM and ANOVA, Table 5, p > 0.05 throughout except a few individual sub-period SGR/K comparisons). African catfish, by contrast, grew significantly less as their numerical share of the tank fell: weight gain (0–106d) dropped from 47.8 ± 6.8 g (TC1, 80 catfish/tank) to 34.1 ± 7.2 g (TC5, 20 catfish/tank), p = 0.001, with GLMM (Table 6) confirming a significant co-culture effect on catfish weight, length, DGR and K. The authors attribute this to catfish specifically benefiting from higher intraspecific density, consistent with Toko et al. (2007).
Compared with:
- todo Elegbe et al. 2015 — pond-based Clarias gariepinus–Oreochromis niloticus co-culture; cited to support that co-culture does not negatively affect tilapia growth in either pond or aquaponic systems. (p.15)
- todo Toko et al. 2007 — reported that African catfish grow better at higher stocking densities, a trend the authors say their own data corroborate. (p.15)
- todo Licamele 2009 — survival rates reported are lower than those recorded in the present study despite this study’s lower stocking densities. (p.15)
- todo Sabwa et al. 2021 — aquaponic Nile tilapia survival rates cited as consistent with this paper’s own results; GLMM analytical approach also adopted from this source. (p.3, p.15)
Fish body composition and heavy metals
This paper: Proximate composition (moisture, crude protein, crude lipid, ash, crude fiber) of both species’ flesh was statistically indistinguishable across treatments (p > 0.05 for all five parameters, both species, Table 7). Seven heavy metals were screened (Table 8): Pb, Cd, and As were non-detected in all samples of both species; Hg, Mn, Cu, and Zn were present at trace levels well below FAO (2020)/WHO (2006) safety thresholds, with no significant treatment effect.
Compared with:
- todo Adeniyi et al. 2012 — cited for the general link between fish body composition (protein/lipid/ash/moisture) and nutritional quality, and to support that composition differences here are species-specific rather than treatment-driven. (p.15)
- todo Sarkar et al. 2021 and Salam et al. 2002 — reported high Pb/Cd/Cr/Cu/Zn in some commercial feeds elsewhere; cited as context for why heavy-metal screening matters, though this paper’s own feed/water were apparently free of Pb/Cd/As contamination. (p.15)
Water quality
This paper: DO, pH, water temperature, EC, TAN/ammonium, nitrite, and nitrate were all reported as trial means (with SD) separately for the hydroponic (“HP”) and aquaponic/fish-tank (“AP”) compartments of each treatment’s coupled system (Table 9), rather than as a single system-wide mean. Values were described by the authors as within optimal/acceptable ranges for both fish species throughout. A PCA (Figure 5) found ammonia, nitrite, and temperature had the strongest positive association with fish growth, and EC a negative one, in the two principal components explaining ~74% of variance.
Compared with: (no external water-quality benchmarks cited by name in this section beyond the general FAO/WHO/aquaculture-suitability ranges already noted in Extraction notes)
Linked claims
- Fish co-culture ratio does not affect Nile tilapia growth or survival in aquaponics
- African catfish growth declines with lower stocking proportion in co-culture
- Heavy metals in aquaponic fish flesh can remain below safety thresholds across a range of co-culture ratios
- Aquaponic eggplant yield is insensitive to fish co-culture ratio when total biomass is held constant
Citations to chase
- todo Coulibaly, N.D. et al. (2021) — soil-grown eggplant reported ~80 fruits/plant vs. this paper’s ~7–8/plant in aquaponics; useful for a soil-vs-soilless eggplant yield comparison
- todo Aglinglo, A.L. et al. (2018) — harvest-interval effects on eggplant fruiting frequency
- todo Elegbe, A. et al. (2015) — pond-based tilapia–catfish co-culture zootechnical/economic performance
- todo Toko, I. et al. (2007) — African catfish stocking-density growth effects
- todo Licamele, J. (2009) — aquaponic biomass/nutrient dynamics, survival-rate benchmark
- todo Sabwa, A. et al. (2021) — GLMM approach for aquaponic tilapia growth analysis; survival-rate benchmark
- todo Adeniyi, S.A. et al. (2012) — species-specific fish body composition benchmarks (Clarias gariepinus, Tilapia guineensis)
- todo Andriani, Y. et al. (2017) — aquaponic tilapia–catfish co-culture with stocking-density ratio effects on plant productivity (methodologically close precedent, condition-factor formula also cited from this source)
Extraction notes
Type classification: experiment — a controlled, randomised(-implied) 5-treatment design with replication language (“each replicate”), formal hypothesis tests (ANOVA, Tukey HSD, Chi-squared, GLMM), and original data collected by the authors (fish biometrics, plant harvest data, proximate/heavy-metal lab analysis). No review or modelling content beyond routine literature comparison in the Discussion.
Contradiction severity tally: 2 × ⚠️BLOCK, 1 × ⚠️MATERIAL, 2 × ⚠️CHECK (recurring across all 5 trials), 3 × ⚠️MINOR. Per SCHEMA.md scoring (BLOCK/MATERIAL only): 2+ BLOCK → quality: suspect.
- ⚠️BLOCK — Nile tilapia overall (0–106d) weight gain, TC2. Section 3.2.2 narrative (p.4) states “mean live weight increased from 20.09 ± 4.5 to 94.19 ± 13.9 g in TC2” (implied gain 74.10 g), matching Table 4’s Weight gain cell for TC2 (74.1 ± 3.3 g). A separate sentence in the same section states “cumulative weight gains were … 68.1 ± 3.3 g (TC2)”, and Table 4’s Daily weight gain cell for TC2 (0.64 ± 0.02 g/day) × 106 days = 67.84 g, matching the 68.1 figure rather than 74.1. Two internally self-consistent clusters (~74 vs ~68), each independently corroborated within the paper; no textual basis to prefer one. Recorded UNCLEAR in the note’s per-species breakdown (trials.csv leaves the Fish weight gain cell NR for schema reasons independent of this flag — see below). Does not change the paper’s own conclusion that tilapia growth was unaffected by ratio, since TC2 falls in the “no significant difference” group under either reading.
- ⚠️BLOCK — Nile tilapia overall (0–106d) weight gain, TC3. Same shape: narrative live-weight progression implies 68.10 g (matching Table 4’s Weight gain cell, 68.5 ± 4.7 g), but a separate narrative sentence states 66.4 ± 4.7 g, and Table 4’s Daily weight gain (0.62 ± 0.03 g/day) × 106 = 65.72 g, closer to 66.4. Recorded UNCLEAR.
- ⚠️MATERIAL — Nile tilapia SGR (0–106d) SD, TC2. Table 4 gives “0.61 ± 0.4”%/day; the p.6 narrative independently states “0.61%/day±0.04%/day (TC2)”. The table’s SD (0.4) is an order of magnitude larger than every sibling treatment’s SD in the same row (0.02–0.05) and than TC2’s own SD in every other period row; almost certainly a dropped-decimal typo in the table. Narrative’s 0.04 recorded as the defensible value in the note; not carried to a trials.csv cell (SGR is NR there for the co-culture schema-limitation reason below), but flagged since it would otherwise mislead a reader of Table 4 directly.
- ⚠️CHECK — NO3-N vs NO3 (all 5 trials). Table 9’s column header reads “NO3⁻” without stating whether the colorimetric method reports total nitrate ion or nitrate-nitrogen; the trials.csv schema column is explicitly NO3-N. Recorded the table’s value as-is in each trial row; if the paper meant total NO3⁻, true NO3-N would be roughly 4.43× smaller. Unresolved from the text.
- ⚠️CHECK — water-quality compartment, HP vs AP (all 5 trials). Table 9 reports DO, pH, EC, ammonium, nitrite, and nitrate separately for the “HP” (hydroponic/plant-bed) and “AP” (fish-tank) sampling points of the same coupled system per treatment, rather than one system mean. Per SCHEMA.md’s instruction to prefer the plant-bed/hydroponic-unit value when compartmentalised, the HP figures were entered into trials.csv; the AP figures (often materially higher, e.g. ammonia and nitrate roughly 2–8× higher at the fish tank) are recorded as the alternate candidate in each trial row’s Experimental Remarks.
- ⚠️MINOR — Marketable eggplant yield SD, TC4. Running text (p.3) truncates TC4’s SD mid-sentence (“TC4: 24823.4 ± and TC5: …”); Table 2’s own cell gives the complete figure (24823.4 ± 85.2), used in trials.csv.
- ⚠️MINOR — Length gain unit label. p.6 narrative labels Nile tilapia length-gain values as “cm/day” via a stray “g/day” unit (“8.45±2.62g/day (TC1)…”), inconsistent with Table 4’s correct “cm” header for the same numbers; values themselves match exactly, so this is a labelling slip with no cell impact.
- ⚠️MINOR — TC1 Week 3 fruit-count SD. p.3 narrative states “25±1.29 fruits in TC1 (Week 3)”; Table 2’s own TC1 Week 3 cell is 25±1.89 (Week 2’s SD is 1.29) — apparent cross-week conflation in the narrative. No trials.csv cell affected (per-week fruit counts are not extracted).
Schema-limitation note (not a contradiction): FCR, SGR, Fish size final, Fish survival rate, and Fish weight gain are NR in every trials.csv row because this is a two-species co-culture and the paper never states a combined/system-level figure for any of these metrics — only separate Nile tilapia and African catfish values (Tables 3–4). Full species-level breakdowns are recorded per-trial in each row’s Experimental Remarks rather than fabricated into a single cell.
[not reported] fields (grouped): Funding; Lat/Long (Bandundu, DRC — no coordinates given); Replicates (n) (paper implies replication — “ten identical aquaponic units”, “each replicate” — but never states the per-treatment count, and 10÷5 is not written in the paper so is not entered per the no-derivation rule); FCR (never mentioned anywhere, despite full feed composition being given); Total Feed (kg); Fish biomass created (kg); Water recycle rate; Water volume in the system; Water type; Water classification; Daily water exchange rate; SPAD; Plant height; Leaf count; Plant dry matter; Tissue nitrate AP/HYD; FUE AP/HYD; WUE; feed N (%) and K (%) (only crude protein and phosphorus are broken out; fat/fiber/ash/sodium/calcium are NO COLUMN items, see below); Biological system already in use, Air supplement, Iron supplemented, Remineralization, pH Buffers, Climate control, Artificial Lighting, Nutrient supplemented, and all paired ”…Details” columns.
[unclear] fields: System type — Methods 2.1 never names the grow-bed design (media bed / DWC / NFT); Figure 1a’s photograph shows a substrate/gravel-filled bed consistent with a media-bed system, but this is not stated in the text, so the cell is UNCLEAR rather than inferred from the image.
NO COLUMN items: Feed composition beyond protein/phosphorus (11.5% crude fat, 4% crude fiber, 8% ash, 0.2% sodium, 1.5% calcium, p.2). Per-treatment fish-flesh proximate composition (moisture, crude protein, crude lipid, ash, crude fiber) for both species (Table 7) — this is fish tissue, not plant tissue, so it also does not qualify for plant.csv; no fish-tissue-proximate column exists anywhere in the schema. Per-treatment heavy metals in fish flesh (Hg, Mn, Cu, Zn; Pb/Cd/As non-detected; Table 8) — same reasoning, no dedicated column.
Judgment call on tags: Used the vault’s existing Meta/Fish/Tilapia (generic, already applied to Oreochromis niloticus in other notes, e.g. lenzCommonChicoryProduction2021, ahmedImpactBiofilmSupport2026) and Meta/Fish/African-catfish (existing spelling, e.g. caloneImprovingWaterManagement2019) rather than inventing new Nile-Tilapia/African-Catfish leaves, per CLAUDE.md’s instruction to check existing facets before adding new ones. Meta/Plant/Eggplant is a new leaf — no existing Eggplant/Aubergine/Solanum melongena facet found in the vault (checked against Meta/Plant/Tomato, Basil, Chicory, Lettuce, Rocket, Cucumber, Coriander, Parsley, Barley, Watercress, Arugula, Nasturtium, Indian-Spinach, Spinach, Lemna-Minor). Meta/Region/Africa reused as-is (existing leaf, e.g. used for other DRC/African-continent papers in the vault).
Water panel note: No water-quality data was excluded from plant.csv — none was ever a candidate, since it belongs in trials.csv per SCHEMA.md regardless. plant.csv is 0 rows: this paper’s only biochemical/proximate/heavy-metal analysis was performed on fish flesh (Nile tilapia and African catfish), never on the eggplant plant tissue. No leaf/fruit tissue nitrate, mineral, biochemistry, microbiology, or proximate analysis of Solanum melongena was performed or reported anywhere in the paper — only fruit production parameters (count, length, diameter, weight, biomass), which are yield data belonging in trials.csv, not plant.csv.
Source: Fidji et al. - 2026 - Integrated Aquaponic Co‐Culture of Nile Tilapia and African Catfish Effects on Plant Yield, Fish Gr.pdf
Data Tables
Structured data extracted from this paper into the vault's
trials.csv/plant_measurements.csvdatasets. Fields the paper didn't report are omitted. Download the full datasets (measurements).
Trial Parameters
fidjiIntegratedAquaponicCoCulture2026-T1
Fish
| Field | Value |
|---|---|
| Fish | Nile tilapia (Oreochromis niloticus) + African catfish (Clarias gariepinus), co-culture ratio 2:8 Nile tilapia:African catfish (TC1; 20 tilapia + 80 catfish per 100-fish tank, Table 1) |
| Fish Category | Warm-water species (p.9) |
| Initial Stock density | 10 |
| Protein | 35 |
| P | 0.8 |
| Fish size initial | 20.21 (average of both species, Table 1) |
| Feed routine | Y |
| Feed regime | Commercial floating pellet, 35% crude protein, 11.5% crude fat, 4% crude fiber, 8% ash, 0.2% sodium, 1.5% calcium, 0.8% phosphorus; fed twice daily at 09:00 and 15:00 (p.2) |
| Fish trial duration (days) | 106 |
Water
| Field | Value |
|---|---|
| Aq pH | 7.43 +/- 1.07 |
| pHOptimal | 6.5-8.5 (stated as optimal range for freshwater aquaculture, p.9 Discussion) |
| Dissolved Oxigen | 6.43 +/- 1.17 |
| EC | 0.9529 +/- 0.0346 |
| Water temperature | 30.62 +/- 5.76 |
| TAN / NH4-N | 0.09 +/- 0.03 |
| NO2-N | <0.05 |
| NO3-N | 1.43 +/- 0.07 |
Plant
| Field | Value |
|---|---|
| Plant | Eggplant (Solanum melongena) |
| Details | Fruit parameters only recorded (number, length, diameter, individual weight, total biomass); seedlings transplanted 30 days post-germination at 9 plants/m2; harvest began 56 days after transplanting and continued weekly (7 weekly harvests, Table 2) to trial end; fruits harvested as they reached maturity (p.2) |
| Plant Category | Fruit-bearing vegetable (p.14) |
| Days Plant after transplant | 56 (harvest start, stated, p.2); harvested weekly for 7 weeks to trial end (~day 106) |
| Plants/m2 | 9 |
| Plant fresh weight | 53.5 +/- 6.4 |
System & Setup
| Field | Value |
|---|---|
| System type | UNCLEAR (media/system design not named in Methods 2.1; Figure 1a photo shows a substrate/gravel-filled grow bed, consistent with a media-bed design but not explicitly stated as such in text) |
| Equipment | Hanna HI98194 multiparameter meter (water temperature, dissolved oxygen, pH); commercial colorimetric test kits (TAN, NO2-); mercury thermometer (air temperature); ichthyometer, 0.1mm precision (fish standard length); digital scale, 0.1g precision (fish body weight); R software v4.3.2 with XLSTAT add-in (p.3) |
| Control Parameters | Fixed feeding schedule, twice daily at 09:00 and 15:00; fixed initial total biomass of 10 kg/m3 per tank across all five treatments; fish biometric sampling every 26 days over the 106-day trial (p.2-3) |
| Combination | Nile tilapia (Oreochromis niloticus) + African catfish (Clarias gariepinus) co-culture at five stocking ratios, with eggplant (Solanum melongena) in the same coupled aquaponic unit |
Site
| Field | Value |
|---|---|
| Region | Africa |
| Country | Democratic Republic of the Congo |
| Average room Temperature | 30.85 +/- 4.52 |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | Fruit count (Nb); fruit length (cm); fruit diameter (cm); individual fruit weight (g); total fruit biomass (g); yield expressed as Nb/plant, g/plant/week, g/m2/week, Nb/m2/year, g/m2/year (Table 2) |
| Statistic Details | Levene’s test (variance homogeneity); one-way ANOVA + Tukey’s HSD post hoc; Chi-squared test (survival, Table 3); Principal Component Analysis (water quality vs fish growth, Fig. 5); generalized linear mixed-effects models (GLMM, Tables 5-6, following Sabwa et al. 2021); R v4.3.2 and Excel + XLSTAT add-in; alpha=0.05 (p.3) |
| Statistically analysed | Y |
| AP | 22334.4 +/- 78.4 |
Experimental Remarks: TRIAL DEFINITION: T1 = TC1, Nile tilapia:African catfish stocking ratio 2:8 (20 tilapia + 80 catfish per 100-fish tank; initial biomass split accordingly within the fixed 10 kg/m3 total, Table 1). All five trials (T1-T5) share the same eggplant (Solanum melongena) co-culture system at constant 10 kg/m3 total fish biomass and 9 plants/m2; only the tilapia:catfish stocking ratio differs across trials. No hydroponic-only control arm exists in this study (fully coupled aquaponic system; HYD=NA throughout) — the comparison of interest is between co-culture ratios, not AP vs HYD. | SCHEMA LIMITATION, not a contradiction: FCR, SGR, Fish size final, Fish survival rate, and Fish weight gain are left NR in this row because the paper tracks Nile tilapia and African catfish as two separate species within the same co-culture tank and never states a combined/system-level figure for any of these metrics — only per-species values (Table 3, Table 4). Species-level breakdown for TC1: Nile tilapia final weight 92.13 +/- 16.3 g (initial 20.93 +/- 5.8 g), weight gain 71.2 +/- 5.2 g (0-106d), DWG 0.67 +/- 0.01 g/day, length gain 8.45 +/- 2.62 cm, SGR 0.59 +/- 0.05 %/day, K 2.1 +/- 0.48, survival 92.5 +/- 5.7%. African catfish final weight 67.24 +/- 11.7 g (initial 20.04 +/- 4.1 g), weight gain 47.8 +/- 6.8 g, DWG 0.45 +/- 0.07 g/day, length gain 8.21 +/- 1.35 cm, SGR 0.27 +/- 0.01 %/day, K 1.3 +/- 0.19, survival 98.7 +/- 2.9% (Tables 3-4). | WARN-CHECK NO3-N vs NO3 (applies to all 5 trials): Table 9 header reads ‘NO3-’ (nitrate ion) without stating whether the colorimetric method reports total NO3- or NO3—N; the schema column expects NO3-N specifically. Recorded the table’s HP-compartment value as-is (1.43 +/- 0.07 mg/L); if the paper intended total nitrate ion rather than nitrate-N, true NO3-N would be roughly 4.43x smaller. Not resolvable from the text (Methods 2.2 only names colorimetric kits generically for TAN and NO2-, no method/brand named for nitrate itself). | WARN-CHECK water-quality compartment (applies to all 5 trials): Table 9 reports DO/pH/EC/NH3/NO3 separately for ‘HP’ (hydroponic/plant-bed) and ‘AP’ (fish-tank) sampling points within the same coupled system. For TC1: DO HP 6.43 +/- 1.17 vs AP 7.80 +/- 2.57 mg/L; pH HP 7.43 +/- 1.07 vs AP 7.27 +/- 1.12; EC HP 0.9529 +/- 0.0346 vs AP 0.7653 +/- 0.021 dS/m (raw uS/cm: HP and AP as in Table 9); Water temperature HP 30.62 +/- 5.76 vs AP 30.18 +/- 7.21 degC; TAN/NH4-N HP 0.09 +/- 0.03 vs AP 0.76 +/- 0.04 mg/L; NO3-N HP 1.43 +/- 0.07 vs AP 4.86 +/- 1.28 mg/L. Per SCHEMA.md guidance to take the plant-bed/hydroponic-unit value when a paper reports separate per-compartment readings, the HP column is recorded in the Aq pH/Dissolved Oxigen/EC/Water temperature/TAN/NO2/NO3 cells above; the AP (fish-tank) compartment values given here are the alternate candidate, often materially higher (e.g. NH3/NH4+ and NO3-). | WARN-MINOR: Table 9’s ‘T Air’ column shows a dash (’—’) under HP for TC1 and a single value under AP only (30.85 +/- 4.52 degC) — air/ambient temperature was evidently measured once per day (Methods 2.2: ‘ambient air temperature was recorded daily using a mercury thermometer’), not per compartment; the AP-column placement appears to be a table-formatting artifact rather than a true fish-tank-specific air reading. Recorded in Average room Temperature as given. | WARN-CHECK Plant fresh weight: the paper never states a single cumulative fresh weight per plant for the full 7-week harvest window — only a per-week average (‘Marketable eggplant yield, g/plant/week’, Table 2: 53.5 +/- 6.4 for TC1) and a projected annualised area yield (‘g/m2/year’: 22334.4 +/- 78.4). Recorded the weekly per-plant average in Plant fresh weight and the annualised area yield in the AP column; both are the paper’s own stated figures, not derived. | NOT DERIVED, left NR: Replicates (n) — paper states ‘ten identical aquaponic units’ and ‘five treatment groups’ (implying up to 2 tanks/treatment) and separately refers to ‘each replicate’ when sampling fish (p.3), confirming replication exists, but never states the per-treatment replicate count as a number; 10 divided by 5 is not written in the paper and is not entered per the no-derivation rule. FCR (never reported for either species, anywhere in the paper — no total feed intake figure is given either). Total Feed (kg), Fish biomass created (kg), Water recycle rate, Water volume, Water type/classification, Daily water exchange rate, SPAD, Plant height, Leaf count, Plant dry matter, Tissue nitrate AP/HYD, FUE AP/HYD, WUE, Lat/Long (Bandundu, DRC — no coordinates given), N (feed nitrogen, %), K (feed potassium, %) — none stated. | NO COLUMN: feed composition beyond protein/phosphorus — 11.5% crude fat, 4% crude fiber, 8% ash, 0.2% sodium, 1.5% calcium (p.2, no dedicated columns). Fish flesh proximate composition (TC1): Nile tilapia moisture 86.9 +/- 1.2%, crude protein 25.0 +/- 1.2%, crude lipid 3.57 +/- 0.3%, ash 0.81 +/- 0.0%, crude fiber 1.39 +/- 0.1%; African catfish moisture 87.2 +/- 2.6%, crude protein 18.6 +/- 1.5%, crude lipid 6.72 +/- 0.2%, ash 1.64 +/- 0.2%, crude fiber 1.78 +/- 0.3% (Table 7; both species ns across treatments, p>0.05) — this is FISH tissue proximate data, not plant tissue, so it does not belong in plant.csv either; no fish-tissue-proximate column exists in trials.csv. Heavy metals in fish flesh (TC1, mg/kg): Nile tilapia Hg 0.014, Mn 0.091, Cu 0.039, Zn 0.025 (Pb/Cd/As non-detect); African catfish Hg 0.011, Mn 0.059, Cu 0.026, Zn 0.033 (Pb/Cd/As non-detect); all well below FAO (2020)/WHO (2006) safety limits (Table 8) — no dedicated heavy-metal column exists.
fidjiIntegratedAquaponicCoCulture2026-T2
Fish
| Field | Value |
|---|---|
| Fish | Nile tilapia (Oreochromis niloticus) + African catfish (Clarias gariepinus), co-culture ratio 4:6 Nile tilapia:African catfish (TC2; 40 tilapia + 60 catfish per 100-fish tank, Table 1) |
| Fish Category | Warm-water species (p.9) |
| Initial Stock density | 10 |
| Protein | 35 |
| P | 0.8 |
| Fish size initial | 20.23 (average of both species, Table 1) |
| Feed routine | Y |
| Feed regime | Commercial floating pellet, 35% crude protein, 11.5% crude fat, 4% crude fiber, 8% ash, 0.2% sodium, 1.5% calcium, 0.8% phosphorus; fed twice daily at 09:00 and 15:00 (p.2) |
| Fish trial duration (days) | 106 |
Water
| Field | Value |
|---|---|
| Aq pH | 7.51 +/- 1.64 |
| pHOptimal | 6.5-8.5 (stated as optimal range for freshwater aquaculture, p.9 Discussion) |
| Dissolved Oxigen | 6.68 +/- 1.06 |
| EC | 0.8108 +/- 0.0174 |
| Water temperature | 30.17 +/- 4.95 |
| TAN / NH4-N | 0.13 +/- 0.02 |
| NO2-N | <0.05 |
| NO3-N | 1.45 +/- 0.04 |
Plant
| Field | Value |
|---|---|
| Plant | Eggplant (Solanum melongena) |
| Details | Fruit parameters only recorded (number, length, diameter, individual weight, total biomass); seedlings transplanted 30 days post-germination at 9 plants/m2; harvest began 56 days after transplanting and continued weekly (7 weekly harvests, Table 2) to trial end; fruits harvested as they reached maturity (p.2) |
| Plant Category | Fruit-bearing vegetable (p.14) |
| Days Plant after transplant | 56 (harvest start, stated, p.2); harvested weekly for 7 weeks to trial end (~day 106) |
| Plants/m2 | 9 |
| Plant fresh weight | 54.2 +/- 8.3 |
System & Setup
| Field | Value |
|---|---|
| System type | UNCLEAR (media/system design not named in Methods 2.1; Figure 1a photo shows a substrate/gravel-filled grow bed, consistent with a media-bed design but not explicitly stated as such in text) |
| Equipment | Hanna HI98194 multiparameter meter (water temperature, dissolved oxygen, pH); commercial colorimetric test kits (TAN, NO2-); mercury thermometer (air temperature); ichthyometer, 0.1mm precision (fish standard length); digital scale, 0.1g precision (fish body weight); R software v4.3.2 with XLSTAT add-in (p.3) |
| Control Parameters | Fixed feeding schedule, twice daily at 09:00 and 15:00; fixed initial total biomass of 10 kg/m3 per tank across all five treatments; fish biometric sampling every 26 days over the 106-day trial (p.2-3) |
| Combination | Nile tilapia (Oreochromis niloticus) + African catfish (Clarias gariepinus) co-culture at five stocking ratios, with eggplant (Solanum melongena) in the same coupled aquaponic unit |
Site
| Field | Value |
|---|---|
| Region | Africa |
| Country | Democratic Republic of the Congo |
| Average room Temperature | 30.81 +/- 5.79 |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | Fruit count (Nb); fruit length (cm); fruit diameter (cm); individual fruit weight (g); total fruit biomass (g); yield expressed as Nb/plant, g/plant/week, g/m2/week, Nb/m2/year, g/m2/year (Table 2) |
| Statistic Details | Levene’s test (variance homogeneity); one-way ANOVA + Tukey’s HSD post hoc; Chi-squared test (survival, Table 3); Principal Component Analysis (water quality vs fish growth, Fig. 5); generalized linear mixed-effects models (GLMM, Tables 5-6, following Sabwa et al. 2021); R v4.3.2 and Excel + XLSTAT add-in; alpha=0.05 (p.3) |
| Statistically analysed | Y |
| AP | 28052.6 +/- 65.9 |
Experimental Remarks: TRIAL DEFINITION: T2 = TC2, Nile tilapia:African catfish stocking ratio 4:6 (40 tilapia + 60 catfish per 100-fish tank; initial biomass split accordingly within the fixed 10 kg/m3 total, Table 1). All five trials (T1-T5) share the same eggplant (Solanum melongena) co-culture system at constant 10 kg/m3 total fish biomass and 9 plants/m2; only the tilapia:catfish stocking ratio differs across trials. No hydroponic-only control arm exists in this study (fully coupled aquaponic system; HYD=NA throughout) — the comparison of interest is between co-culture ratios, not AP vs HYD. | SCHEMA LIMITATION, not a contradiction: FCR, SGR, Fish size final, Fish survival rate, and Fish weight gain are left NR in this row because the paper tracks Nile tilapia and African catfish as two separate species within the same co-culture tank and never states a combined/system-level figure for any of these metrics — only per-species values (Table 3, Table 4). Species-level breakdown for TC2: Nile tilapia final weight 94.19 +/- 13.9 g (initial 20.09 +/- 4.5 g), weight gain UNCLEAR (68.1 or 74.1, see WARN-BLOCK below) g (0-106d), DWG 0.64 +/- 0.02 g/day, length gain 9.83 +/- 2.49 cm, SGR 0.61 +/- 0.04 (text) / 0.61 +/- 0.4 (Table 4, likely typo, see WARN-MATERIAL) %/day, K 1.9 +/- 0.37, survival 95.0 +/- 6.3%. African catfish final weight 65.22 +/- 12.5 g (initial 20.32 +/- 5.2 g), weight gain 44.9 +/- 7.2 g, DWG 0.44 +/- 0.03 g/day, length gain 7.55 +/- 1.79 cm, SGR 0.24 +/- 0.01 %/day, K 1.5 +/- 0.28, survival 95.0 +/- 3.8% (Tables 3-4). | WARN-BLOCK Nile tilapia overall (0-106d) weight gain, TC2: three internally-inconsistent figures for the same quantity. (a) Section 3.2.2 narrative (p.4): ‘mean live weight increased from 20.09 +/- 4.5 to 94.19 +/- 13.9 g in TC2’ -> implied gain = 74.10 g. (b) Table 4, Weight gain (g), 0-106 days, TC2 cell: ‘74.1 +/- 3.3 g’ (matches (a) via recomputation, verification only, not entered as a cell value). (c) Section 3.2.2 narrative (p.4), separate sentence: ‘cumulative weight gains were … 68.1 +/- 3.3 g (TC2)’. (d) Table 4, Daily weight gain (g/day), 0-106 days, TC2 cell: ‘0.64 +/- 0.02 g/day’, which x106 days = 67.84 g (close to (c), not to (a)/(b)); the same 0.64 figure is independently restated in the p.6 narrative. Values (a)+(b) form one self-consistent cluster (~74 g); (c)+(d) form a second self-consistent cluster (~68 g). No basis in the text to prefer one cluster over the other — both are corroborated by an independent second figure within the paper. Recorded UNCLEAR (see species-level breakdown above, Weight gain cell for Nile tilapia). Affects: interpretation of whether TC2 tilapia growth was similar to or higher than TC1/TC4 (72.5 g). | WARN-CHECK NO3-N vs NO3 (applies to all 5 trials): Table 9 header reads ‘NO3-’ (nitrate ion) without stating whether the colorimetric method reports total NO3- or NO3—N; the schema column expects NO3-N specifically. Recorded the table’s HP-compartment value as-is (1.45 +/- 0.04 mg/L); if the paper intended total nitrate ion rather than nitrate-N, true NO3-N would be roughly 4.43x smaller. Not resolvable from the text (Methods 2.2 only names colorimetric kits generically for TAN and NO2-, no method/brand named for nitrate itself). | WARN-CHECK water-quality compartment (applies to all 5 trials): Table 9 reports DO/pH/EC/NH3/NO3 separately for ‘HP’ (hydroponic/plant-bed) and ‘AP’ (fish-tank) sampling points within the same coupled system. For TC2: DO HP 6.68 +/- 1.06 vs AP 7.55 +/- 2.29 mg/L; pH HP 7.51 +/- 1.64 vs AP 7.05 +/- 1.02; EC HP 0.8108 +/- 0.0174 vs AP 0.9323 +/- 0.0197 dS/m (raw uS/cm: HP and AP as in Table 9); Water temperature HP 30.17 +/- 4.95 vs AP 30.24 +/- 3.67 degC; TAN/NH4-N HP 0.13 +/- 0.02 vs AP 0.37 +/- 0.07 mg/L; NO3-N HP 1.45 +/- 0.04 vs AP 5.15 +/- 1.51 mg/L. Per SCHEMA.md guidance to take the plant-bed/hydroponic-unit value when a paper reports separate per-compartment readings, the HP column is recorded in the Aq pH/Dissolved Oxigen/EC/Water temperature/TAN/NO2/NO3 cells above; the AP (fish-tank) compartment values given here are the alternate candidate, often materially higher (e.g. NH3/NH4+ and NO3-). | WARN-MINOR: Table 9’s ‘T Air’ column shows a dash (’—’) under HP for TC2 and a single value under AP only (30.81 +/- 5.79 degC) — air/ambient temperature was evidently measured once per day (Methods 2.2: ‘ambient air temperature was recorded daily using a mercury thermometer’), not per compartment; the AP-column placement appears to be a table-formatting artifact rather than a true fish-tank-specific air reading. Recorded in Average room Temperature as given. | WARN-CHECK Plant fresh weight: the paper never states a single cumulative fresh weight per plant for the full 7-week harvest window — only a per-week average (‘Marketable eggplant yield, g/plant/week’, Table 2: 54.2 +/- 8.3 for TC2) and a projected annualised area yield (‘g/m2/year’: 28052.6 +/- 65.9). Recorded the weekly per-plant average in Plant fresh weight and the annualised area yield in the AP column; both are the paper’s own stated figures, not derived. | NOT DERIVED, left NR: Replicates (n) — paper states ‘ten identical aquaponic units’ and ‘five treatment groups’ (implying up to 2 tanks/treatment) and separately refers to ‘each replicate’ when sampling fish (p.3), confirming replication exists, but never states the per-treatment replicate count as a number; 10 divided by 5 is not written in the paper and is not entered per the no-derivation rule. FCR (never reported for either species, anywhere in the paper — no total feed intake figure is given either). Total Feed (kg), Fish biomass created (kg), Water recycle rate, Water volume, Water type/classification, Daily water exchange rate, SPAD, Plant height, Leaf count, Plant dry matter, Tissue nitrate AP/HYD, FUE AP/HYD, WUE, Lat/Long (Bandundu, DRC — no coordinates given), N (feed nitrogen, %), K (feed potassium, %) — none stated. | NO COLUMN: feed composition beyond protein/phosphorus — 11.5% crude fat, 4% crude fiber, 8% ash, 0.2% sodium, 1.5% calcium (p.2, no dedicated columns). Fish flesh proximate composition (TC2): Nile tilapia moisture 79.8 +/- 1.1%, crude protein 25.0 +/- 2.5%, crude lipid 3.29 +/- 0.1%, ash 0.82 +/- 0.0%, crude fiber 1.42 +/- 0.1%; African catfish moisture 83.5 +/- 1.8%, crude protein 17.2 +/- 1.1%, crude lipid 6.99 +/- 0.6%, ash 0.97 +/- 0.1%, crude fiber 1.84 +/- 0.1% (Table 7; both species ns across treatments, p>0.05) — this is FISH tissue proximate data, not plant tissue, so it does not belong in plant.csv either; no fish-tissue-proximate column exists in trials.csv. Heavy metals in fish flesh (TC2, mg/kg): Nile tilapia Hg 0.013, Mn 0.084, Cu 0.034, Zn 0.029 (Pb/Cd/As non-detect); African catfish Hg 0.010, Mn 0.07, Cu 0.031, Zn 0.045 (Pb/Cd/As non-detect); all well below FAO (2020)/WHO (2006) safety limits (Table 8) — no dedicated heavy-metal column exists.
fidjiIntegratedAquaponicCoCulture2026-T3
Fish
| Field | Value |
|---|---|
| Fish | Nile tilapia (Oreochromis niloticus) + African catfish (Clarias gariepinus), co-culture ratio 5:5 Nile tilapia:African catfish (TC3; 50 tilapia + 50 catfish per 100-fish tank, Table 1) |
| Fish Category | Warm-water species (p.9) |
| Initial Stock density | 10 |
| Protein | 35 |
| P | 0.8 |
| Fish size initial | 20.29 (average of both species, Table 1) |
| Feed routine | Y |
| Feed regime | Commercial floating pellet, 35% crude protein, 11.5% crude fat, 4% crude fiber, 8% ash, 0.2% sodium, 1.5% calcium, 0.8% phosphorus; fed twice daily at 09:00 and 15:00 (p.2) |
| Fish trial duration (days) | 106 |
Water
| Field | Value |
|---|---|
| Aq pH | 7.21 +/- 1.73 |
| pHOptimal | 6.5-8.5 (stated as optimal range for freshwater aquaculture, p.9 Discussion) |
| Dissolved Oxigen | 6.54 +/- 1.74 |
| EC | 0.8715 +/- 0.0167 |
| Water temperature | 30.25 +/- 4.77 |
| TAN / NH4-N | 0.19 +/- 0.03 |
| NO2-N | <0.05 |
| NO3-N | 1.15 +/- 0.00 |
Plant
| Field | Value |
|---|---|
| Plant | Eggplant (Solanum melongena) |
| Details | Fruit parameters only recorded (number, length, diameter, individual weight, total biomass); seedlings transplanted 30 days post-germination at 9 plants/m2; harvest began 56 days after transplanting and continued weekly (7 weekly harvests, Table 2) to trial end; fruits harvested as they reached maturity (p.2) |
| Plant Category | Fruit-bearing vegetable (p.14) |
| Days Plant after transplant | 56 (harvest start, stated, p.2); harvested weekly for 7 weeks to trial end (~day 106) |
| Plants/m2 | 9 |
| Plant fresh weight | 57.2 +/- 7.8 |
System & Setup
| Field | Value |
|---|---|
| System type | UNCLEAR (media/system design not named in Methods 2.1; Figure 1a photo shows a substrate/gravel-filled grow bed, consistent with a media-bed design but not explicitly stated as such in text) |
| Equipment | Hanna HI98194 multiparameter meter (water temperature, dissolved oxygen, pH); commercial colorimetric test kits (TAN, NO2-); mercury thermometer (air temperature); ichthyometer, 0.1mm precision (fish standard length); digital scale, 0.1g precision (fish body weight); R software v4.3.2 with XLSTAT add-in (p.3) |
| Control Parameters | Fixed feeding schedule, twice daily at 09:00 and 15:00; fixed initial total biomass of 10 kg/m3 per tank across all five treatments; fish biometric sampling every 26 days over the 106-day trial (p.2-3) |
| Combination | Nile tilapia (Oreochromis niloticus) + African catfish (Clarias gariepinus) co-culture at five stocking ratios, with eggplant (Solanum melongena) in the same coupled aquaponic unit |
Site
| Field | Value |
|---|---|
| Region | Africa |
| Country | Democratic Republic of the Congo |
| Average room Temperature | 30.20 +/- 5.42 |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | Fruit count (Nb); fruit length (cm); fruit diameter (cm); individual fruit weight (g); total fruit biomass (g); yield expressed as Nb/plant, g/plant/week, g/m2/week, Nb/m2/year, g/m2/year (Table 2) |
| Statistic Details | Levene’s test (variance homogeneity); one-way ANOVA + Tukey’s HSD post hoc; Chi-squared test (survival, Table 3); Principal Component Analysis (water quality vs fish growth, Fig. 5); generalized linear mixed-effects models (GLMM, Tables 5-6, following Sabwa et al. 2021); R v4.3.2 and Excel + XLSTAT add-in; alpha=0.05 (p.3) |
| Statistically analysed | Y |
| AP | 26372.9 +/- 82.7 |
Experimental Remarks: TRIAL DEFINITION: T3 = TC3, Nile tilapia:African catfish stocking ratio 5:5 (50 tilapia + 50 catfish per 100-fish tank; initial biomass split accordingly within the fixed 10 kg/m3 total, Table 1). All five trials (T1-T5) share the same eggplant (Solanum melongena) co-culture system at constant 10 kg/m3 total fish biomass and 9 plants/m2; only the tilapia:catfish stocking ratio differs across trials. No hydroponic-only control arm exists in this study (fully coupled aquaponic system; HYD=NA throughout) — the comparison of interest is between co-culture ratios, not AP vs HYD. | SCHEMA LIMITATION, not a contradiction: FCR, SGR, Fish size final, Fish survival rate, and Fish weight gain are left NR in this row because the paper tracks Nile tilapia and African catfish as two separate species within the same co-culture tank and never states a combined/system-level figure for any of these metrics — only per-species values (Table 3, Table 4). Species-level breakdown for TC3: Nile tilapia final weight 88.51 +/- 15.4 g (initial 20.41 +/- 5.1 g), weight gain UNCLEAR (66.4 or 68.5, see WARN-BLOCK below) g (0-106d), DWG 0.62 +/- 0.03 g/day, length gain 6.48 +/- 2.83 cm, SGR 0.57 +/- 0.03 %/day, K 1.8 +/- 0.42, survival 95.0 +/- 6.9%. African catfish final weight 71.38 +/- 13.6 g (initial 20.18 +/- 4.9 g), weight gain 51.2 +/- 8.6 g, DWG 0.42 +/- 0.02 g/day, length gain 6.53 +/- 2.02 cm, SGR 0.33 +/- 0.02 %/day, K 1.3 +/- 0.12, survival 97.0 +/- 4.5% (Tables 3-4). | WARN-BLOCK Nile tilapia overall (0-106d) weight gain, TC3: same pattern as TC2 (see T2 remarks for the general shape of this issue). (a) Section 3.2.2 narrative (p.4): ‘mean live weight increased from 20.41 +/- 5.1 to 88.51 +/- 15.4 g in TC3’ -> implied gain = 68.10 g. (b) Table 4, Weight gain (g), 0-106 days, TC3 cell: ‘68.5 +/- 4.7 g’ (matches (a), within rounding; verification only). (c) Section 3.2.2 narrative (p.4): ‘cumulative weight gains were … 66.4 +/- 4.7 g (TC3)’. (d) Table 4, Daily weight gain (g/day), 0-106 days, TC3 cell: ‘0.62 +/- 0.03 g/day’ x106 = 65.72 g (closer to (c) than to (a)/(b)); independently restated in p.6 narrative as 0.62 +/- 0.03. Two self-consistent clusters (~68 vs ~66), no textual basis to prefer either. Recorded UNCLEAR. Affects: whether TC3 (highest-catfish-number-among-mid-ratio) tilapia growth reading is ~66 or ~68 g, relevant to the paper’s own claim that tilapia growth was unaffected by co-culture ratio (both readings are close enough that this conclusion is unlikely to change). | WARN-CHECK NO3-N vs NO3 (applies to all 5 trials): Table 9 header reads ‘NO3-’ (nitrate ion) without stating whether the colorimetric method reports total NO3- or NO3—N; the schema column expects NO3-N specifically. Recorded the table’s HP-compartment value as-is (1.15 +/- 0.00 mg/L); if the paper intended total nitrate ion rather than nitrate-N, true NO3-N would be roughly 4.43x smaller. Not resolvable from the text (Methods 2.2 only names colorimetric kits generically for TAN and NO2-, no method/brand named for nitrate itself). | WARN-CHECK water-quality compartment (applies to all 5 trials): Table 9 reports DO/pH/EC/NH3/NO3 separately for ‘HP’ (hydroponic/plant-bed) and ‘AP’ (fish-tank) sampling points within the same coupled system. For TC3: DO HP 6.54 +/- 1.74 vs AP 7.27 +/- 2.42 mg/L; pH HP 7.21 +/- 1.73 vs AP 7.04 +/- 1.22; EC HP 0.8715 +/- 0.0167 vs AP 0.8391 +/- 0.0213 dS/m (raw uS/cm: HP and AP as in Table 9); Water temperature HP 30.25 +/- 4.77 vs AP 30.81 +/- 4.02 degC; TAN/NH4-N HP 0.19 +/- 0.03 vs AP 1.03 +/- 0.64 mg/L; NO3-N HP 1.15 +/- 0.00 vs AP 5.10 +/- 1.36 mg/L. Per SCHEMA.md guidance to take the plant-bed/hydroponic-unit value when a paper reports separate per-compartment readings, the HP column is recorded in the Aq pH/Dissolved Oxigen/EC/Water temperature/TAN/NO2/NO3 cells above; the AP (fish-tank) compartment values given here are the alternate candidate, often materially higher (e.g. NH3/NH4+ and NO3-). | WARN-MINOR: Table 9’s ‘T Air’ column shows a dash (’—’) under HP for TC3 and a single value under AP only (30.20 +/- 5.42 degC) — air/ambient temperature was evidently measured once per day (Methods 2.2: ‘ambient air temperature was recorded daily using a mercury thermometer’), not per compartment; the AP-column placement appears to be a table-formatting artifact rather than a true fish-tank-specific air reading. Recorded in Average room Temperature as given. | WARN-CHECK Plant fresh weight: the paper never states a single cumulative fresh weight per plant for the full 7-week harvest window — only a per-week average (‘Marketable eggplant yield, g/plant/week’, Table 2: 57.2 +/- 7.8 for TC3) and a projected annualised area yield (‘g/m2/year’: 26372.9 +/- 82.7). Recorded the weekly per-plant average in Plant fresh weight and the annualised area yield in the AP column; both are the paper’s own stated figures, not derived. | NOT DERIVED, left NR: Replicates (n) — paper states ‘ten identical aquaponic units’ and ‘five treatment groups’ (implying up to 2 tanks/treatment) and separately refers to ‘each replicate’ when sampling fish (p.3), confirming replication exists, but never states the per-treatment replicate count as a number; 10 divided by 5 is not written in the paper and is not entered per the no-derivation rule. FCR (never reported for either species, anywhere in the paper — no total feed intake figure is given either). Total Feed (kg), Fish biomass created (kg), Water recycle rate, Water volume, Water type/classification, Daily water exchange rate, SPAD, Plant height, Leaf count, Plant dry matter, Tissue nitrate AP/HYD, FUE AP/HYD, WUE, Lat/Long (Bandundu, DRC — no coordinates given), N (feed nitrogen, %), K (feed potassium, %) — none stated. | NO COLUMN: feed composition beyond protein/phosphorus — 11.5% crude fat, 4% crude fiber, 8% ash, 0.2% sodium, 1.5% calcium (p.2, no dedicated columns). Fish flesh proximate composition (TC3): Nile tilapia moisture 85.7 +/- 1.4%, crude protein 24.3 +/- 1.8%, crude lipid 3.05 +/- 0.1%, ash 0.86 +/- 0.0%, crude fiber 1.27 +/- 0.1%; African catfish moisture 84.9 +/- 2.6%, crude protein 19.8 +/- 1.6%, crude lipid 7.32 +/- 1.0%, ash 1.05 +/- 0.1%, crude fiber 1.69 +/- 0.1% (Table 7; both species ns across treatments, p>0.05) — this is FISH tissue proximate data, not plant tissue, so it does not belong in plant.csv either; no fish-tissue-proximate column exists in trials.csv. Heavy metals in fish flesh (TC3, mg/kg): Nile tilapia Hg 0.016, Mn 0.079, Cu 0.029, Zn 0.022 (Pb/Cd/As non-detect); African catfish Hg 0.014, Mn 0.062, Cu 0.029, Zn 0.032 (Pb/Cd/As non-detect); all well below FAO (2020)/WHO (2006) safety limits (Table 8) — no dedicated heavy-metal column exists.
fidjiIntegratedAquaponicCoCulture2026-T4
Fish
| Field | Value |
|---|---|
| Fish | Nile tilapia (Oreochromis niloticus) + African catfish (Clarias gariepinus), co-culture ratio 6:4 Nile tilapia:African catfish (TC4; 60 tilapia + 40 catfish per 100-fish tank, Table 1) |
| Fish Category | Warm-water species (p.9) |
| Initial Stock density | 10 |
| Protein | 35 |
| P | 0.8 |
| Fish size initial | 20.14 (average of both species, Table 1) |
| Feed routine | Y |
| Feed regime | Commercial floating pellet, 35% crude protein, 11.5% crude fat, 4% crude fiber, 8% ash, 0.2% sodium, 1.5% calcium, 0.8% phosphorus; fed twice daily at 09:00 and 15:00 (p.2) |
| Fish trial duration (days) | 106 |
Water
| Field | Value |
|---|---|
| Aq pH | 7.25 +/- 1.86 |
| pHOptimal | 6.5-8.5 (stated as optimal range for freshwater aquaculture, p.9 Discussion) |
| Dissolved Oxigen | 5.95 +/- 2.06 |
| EC | 0.9644 +/- 0.0185 |
| Water temperature | 30.05 +/- 3.27 |
| TAN / NH4-N | 0.59 +/- 0.05 |
| NO2-N | <0.05 |
| NO3-N | 1.75 +/- 0.03 |
Plant
| Field | Value |
|---|---|
| Plant | Eggplant (Solanum melongena) |
| Details | Fruit parameters only recorded (number, length, diameter, individual weight, total biomass); seedlings transplanted 30 days post-germination at 9 plants/m2; harvest began 56 days after transplanting and continued weekly (7 weekly harvests, Table 2) to trial end; fruits harvested as they reached maturity (p.2) |
| Plant Category | Fruit-bearing vegetable (p.14) |
| Days Plant after transplant | 56 (harvest start, stated, p.2); harvested weekly for 7 weeks to trial end (~day 106) |
| Plants/m2 | 9 |
| Plant fresh weight | 51.9 +/- 5.7 |
System & Setup
| Field | Value |
|---|---|
| System type | UNCLEAR (media/system design not named in Methods 2.1; Figure 1a photo shows a substrate/gravel-filled grow bed, consistent with a media-bed design but not explicitly stated as such in text) |
| Equipment | Hanna HI98194 multiparameter meter (water temperature, dissolved oxygen, pH); commercial colorimetric test kits (TAN, NO2-); mercury thermometer (air temperature); ichthyometer, 0.1mm precision (fish standard length); digital scale, 0.1g precision (fish body weight); R software v4.3.2 with XLSTAT add-in (p.3) |
| Control Parameters | Fixed feeding schedule, twice daily at 09:00 and 15:00; fixed initial total biomass of 10 kg/m3 per tank across all five treatments; fish biometric sampling every 26 days over the 106-day trial (p.2-3) |
| Combination | Nile tilapia (Oreochromis niloticus) + African catfish (Clarias gariepinus) co-culture at five stocking ratios, with eggplant (Solanum melongena) in the same coupled aquaponic unit |
Site
| Field | Value |
|---|---|
| Region | Africa |
| Country | Democratic Republic of the Congo |
| Average room Temperature | 30.27 +/- 3.86 |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | Fruit count (Nb); fruit length (cm); fruit diameter (cm); individual fruit weight (g); total fruit biomass (g); yield expressed as Nb/plant, g/plant/week, g/m2/week, Nb/m2/year, g/m2/year (Table 2) |
| Statistic Details | Levene’s test (variance homogeneity); one-way ANOVA + Tukey’s HSD post hoc; Chi-squared test (survival, Table 3); Principal Component Analysis (water quality vs fish growth, Fig. 5); generalized linear mixed-effects models (GLMM, Tables 5-6, following Sabwa et al. 2021); R v4.3.2 and Excel + XLSTAT add-in; alpha=0.05 (p.3) |
| Statistically analysed | Y |
| AP | 24823.4 +/- 85.2 |
Experimental Remarks: TRIAL DEFINITION: T4 = TC4, Nile tilapia:African catfish stocking ratio 6:4 (60 tilapia + 40 catfish per 100-fish tank; initial biomass split accordingly within the fixed 10 kg/m3 total, Table 1). All five trials (T1-T5) share the same eggplant (Solanum melongena) co-culture system at constant 10 kg/m3 total fish biomass and 9 plants/m2; only the tilapia:catfish stocking ratio differs across trials. No hydroponic-only control arm exists in this study (fully coupled aquaponic system; HYD=NA throughout) — the comparison of interest is between co-culture ratios, not AP vs HYD. | SCHEMA LIMITATION, not a contradiction: FCR, SGR, Fish size final, Fish survival rate, and Fish weight gain are left NR in this row because the paper tracks Nile tilapia and African catfish as two separate species within the same co-culture tank and never states a combined/system-level figure for any of these metrics — only per-species values (Table 3, Table 4). Species-level breakdown for TC4: Nile tilapia final weight 93.16 +/- 18.6 g (initial 20.6 +/- 3.9 g), weight gain 72.5 +/- 6.7 g (0-106d), DWG 0.68 +/- 0.02 g/day, length gain 8.93 +/- 3.65 cm, SGR 0.56 +/- 0.04 %/day, K 2.3 +/- 1.03, survival 96.7 +/- 5.5%. African catfish final weight 55.05 +/- 10.8 g (initial 19.45 +/- 5.1 g), weight gain 35.6 +/- 6.5 g, DWG 0.33 +/- 0.02 g/day, length gain 4.58 +/- 1.24 cm, SGR 0.31 +/- 0.02 %/day, K 1.4 +/- 0.14, survival 97.5 +/- 3.1% (Tables 3-4). | WARN-MINOR Marketable eggplant yield (g/m2/year), TC4: running text (p.3) lists the five treatment values as ‘TC1: 22334.4 +/- 78.4; TC2: 28052.6 +/- 65.9; TC3: 26372.9 +/- 82.7; TC4: 24823.4 +/- and TC5: 26073.0 +/- 83.6 g/m2/year’ — TC4’s SD is truncated/missing in the running-text sentence (apparent typesetting error, sentence runs straight into ‘and TC5’). Table 2’s own tabulated cell for TC4 gives the complete figure ‘24823.4 +/- 85.2’, used here instead of the truncated narrative version. No other candidate value exists, so this is recorded as MINOR (table cell is unambiguous) rather than BLOCK/MATERIAL. | WARN-CHECK NO3-N vs NO3 (applies to all 5 trials): Table 9 header reads ‘NO3-’ (nitrate ion) without stating whether the colorimetric method reports total NO3- or NO3—N; the schema column expects NO3-N specifically. Recorded the table’s HP-compartment value as-is (1.75 +/- 0.03 mg/L); if the paper intended total nitrate ion rather than nitrate-N, true NO3-N would be roughly 4.43x smaller. Not resolvable from the text (Methods 2.2 only names colorimetric kits generically for TAN and NO2-, no method/brand named for nitrate itself). | WARN-CHECK water-quality compartment (applies to all 5 trials): Table 9 reports DO/pH/EC/NH3/NO3 separately for ‘HP’ (hydroponic/plant-bed) and ‘AP’ (fish-tank) sampling points within the same coupled system. For TC4: DO HP 5.95 +/- 2.06 vs AP 7.70 +/- 2.29 mg/L; pH HP 7.25 +/- 1.86 vs AP 7.16 +/- 1.34; EC HP 0.9644 +/- 0.0185 vs AP 0.8541 +/- 0.0159 dS/m (raw uS/cm: HP and AP as in Table 9); Water temperature HP 30.05 +/- 3.27 vs AP 30.31 +/- 3.12 degC; TAN/NH4-N HP 0.59 +/- 0.05 vs AP 0.46 +/- 0.03 mg/L; NO3-N HP 1.75 +/- 0.03 vs AP 2.92 +/- 0.03 mg/L. Per SCHEMA.md guidance to take the plant-bed/hydroponic-unit value when a paper reports separate per-compartment readings, the HP column is recorded in the Aq pH/Dissolved Oxigen/EC/Water temperature/TAN/NO2/NO3 cells above; the AP (fish-tank) compartment values given here are the alternate candidate, often materially higher (e.g. NH3/NH4+ and NO3-). | WARN-MINOR: Table 9’s ‘T Air’ column shows a dash (’—’) under HP for TC4 and a single value under AP only (30.27 +/- 3.86 degC) — air/ambient temperature was evidently measured once per day (Methods 2.2: ‘ambient air temperature was recorded daily using a mercury thermometer’), not per compartment; the AP-column placement appears to be a table-formatting artifact rather than a true fish-tank-specific air reading. Recorded in Average room Temperature as given. | WARN-CHECK Plant fresh weight: the paper never states a single cumulative fresh weight per plant for the full 7-week harvest window — only a per-week average (‘Marketable eggplant yield, g/plant/week’, Table 2: 51.9 +/- 5.7 for TC4) and a projected annualised area yield (‘g/m2/year’: 24823.4 +/- 85.2). Recorded the weekly per-plant average in Plant fresh weight and the annualised area yield in the AP column; both are the paper’s own stated figures, not derived. | NOT DERIVED, left NR: Replicates (n) — paper states ‘ten identical aquaponic units’ and ‘five treatment groups’ (implying up to 2 tanks/treatment) and separately refers to ‘each replicate’ when sampling fish (p.3), confirming replication exists, but never states the per-treatment replicate count as a number; 10 divided by 5 is not written in the paper and is not entered per the no-derivation rule. FCR (never reported for either species, anywhere in the paper — no total feed intake figure is given either). Total Feed (kg), Fish biomass created (kg), Water recycle rate, Water volume, Water type/classification, Daily water exchange rate, SPAD, Plant height, Leaf count, Plant dry matter, Tissue nitrate AP/HYD, FUE AP/HYD, WUE, Lat/Long (Bandundu, DRC — no coordinates given), N (feed nitrogen, %), K (feed potassium, %) — none stated. | NO COLUMN: feed composition beyond protein/phosphorus — 11.5% crude fat, 4% crude fiber, 8% ash, 0.2% sodium, 1.5% calcium (p.2, no dedicated columns). Fish flesh proximate composition (TC4): Nile tilapia moisture 77.4 +/- 1.1%, crude protein 24.0 +/- 1.6%, crude lipid 3.57 +/- 0.2%, ash 0.91 +/- 0.1%, crude fiber 1.06 +/- 0.0%; African catfish moisture 82.5 +/- 1.4%, crude protein 15.5 +/- 1.4%, crude lipid 6.15 +/- 0.7%, ash 1.18 +/- 0.1%, crude fiber 1.90 +/- 0.2% (Table 7; both species ns across treatments, p>0.05) — this is FISH tissue proximate data, not plant tissue, so it does not belong in plant.csv either; no fish-tissue-proximate column exists in trials.csv. Heavy metals in fish flesh (TC4, mg/kg): Nile tilapia Hg 0.014, Mn 0.087, Cu 0.033, Zn 0.028 (Pb/Cd/As non-detect); African catfish Hg 0.009, Mn 0.067, Cu 0.034, Zn 0.036 (Pb/Cd/As non-detect); all well below FAO (2020)/WHO (2006) safety limits (Table 8) — no dedicated heavy-metal column exists.
fidjiIntegratedAquaponicCoCulture2026-T5
Fish
| Field | Value |
|---|---|
| Fish | Nile tilapia (Oreochromis niloticus) + African catfish (Clarias gariepinus), co-culture ratio 8:2 Nile tilapia:African catfish (TC5; 80 tilapia + 20 catfish per 100-fish tank, Table 1) |
| Fish Category | Warm-water species (p.9) |
| Initial Stock density | 10 |
| Protein | 35 |
| P | 0.8 |
| Fish size initial | 20.21 (average of both species, Table 1) |
| Feed routine | Y |
| Feed regime | Commercial floating pellet, 35% crude protein, 11.5% crude fat, 4% crude fiber, 8% ash, 0.2% sodium, 1.5% calcium, 0.8% phosphorus; fed twice daily at 09:00 and 15:00 (p.2) |
| Fish trial duration (days) | 106 |
Water
| Field | Value |
|---|---|
| Aq pH | 7.45 +/- 1.94 |
| pHOptimal | 6.5-8.5 (stated as optimal range for freshwater aquaculture, p.9 Discussion) |
| Dissolved Oxigen | 6.39 +/- 1.13 |
| EC | 0.8517 +/- 0.0214 |
| Water temperature | 30.28 +/- 2.59 |
| TAN / NH4-N | 0.53 +/- 0.06 |
| NO2-N | <0.05 |
| NO3-N | 1.25 +/- 0.08 |
Plant
| Field | Value |
|---|---|
| Plant | Eggplant (Solanum melongena) |
| Details | Fruit parameters only recorded (number, length, diameter, individual weight, total biomass); seedlings transplanted 30 days post-germination at 9 plants/m2; harvest began 56 days after transplanting and continued weekly (7 weekly harvests, Table 2) to trial end; fruits harvested as they reached maturity (p.2) |
| Plant Category | Fruit-bearing vegetable (p.14) |
| Days Plant after transplant | 56 (harvest start, stated, p.2); harvested weekly for 7 weeks to trial end (~day 106) |
| Plants/m2 | 9 |
| Plant fresh weight | 55.1 +/- 6.3 |
System & Setup
| Field | Value |
|---|---|
| System type | UNCLEAR (media/system design not named in Methods 2.1; Figure 1a photo shows a substrate/gravel-filled grow bed, consistent with a media-bed design but not explicitly stated as such in text) |
| Equipment | Hanna HI98194 multiparameter meter (water temperature, dissolved oxygen, pH); commercial colorimetric test kits (TAN, NO2-); mercury thermometer (air temperature); ichthyometer, 0.1mm precision (fish standard length); digital scale, 0.1g precision (fish body weight); R software v4.3.2 with XLSTAT add-in (p.3) |
| Control Parameters | Fixed feeding schedule, twice daily at 09:00 and 15:00; fixed initial total biomass of 10 kg/m3 per tank across all five treatments; fish biometric sampling every 26 days over the 106-day trial (p.2-3) |
| Combination | Nile tilapia (Oreochromis niloticus) + African catfish (Clarias gariepinus) co-culture at five stocking ratios, with eggplant (Solanum melongena) in the same coupled aquaponic unit |
Site
| Field | Value |
|---|---|
| Region | Africa |
| Country | Democratic Republic of the Congo |
| Average room Temperature | 30.50 +/- 4.52 |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | Fruit count (Nb); fruit length (cm); fruit diameter (cm); individual fruit weight (g); total fruit biomass (g); yield expressed as Nb/plant, g/plant/week, g/m2/week, Nb/m2/year, g/m2/year (Table 2) |
| Statistic Details | Levene’s test (variance homogeneity); one-way ANOVA + Tukey’s HSD post hoc; Chi-squared test (survival, Table 3); Principal Component Analysis (water quality vs fish growth, Fig. 5); generalized linear mixed-effects models (GLMM, Tables 5-6, following Sabwa et al. 2021); R v4.3.2 and Excel + XLSTAT add-in; alpha=0.05 (p.3) |
| Statistically analysed | Y |
| AP | 26073.0 +/- 83.6 |
Experimental Remarks: TRIAL DEFINITION: T5 = TC5, Nile tilapia:African catfish stocking ratio 8:2 (80 tilapia + 20 catfish per 100-fish tank; initial biomass split accordingly within the fixed 10 kg/m3 total, Table 1). All five trials (T1-T5) share the same eggplant (Solanum melongena) co-culture system at constant 10 kg/m3 total fish biomass and 9 plants/m2; only the tilapia:catfish stocking ratio differs across trials. No hydroponic-only control arm exists in this study (fully coupled aquaponic system; HYD=NA throughout) — the comparison of interest is between co-culture ratios, not AP vs HYD. | SCHEMA LIMITATION, not a contradiction: FCR, SGR, Fish size final, Fish survival rate, and Fish weight gain are left NR in this row because the paper tracks Nile tilapia and African catfish as two separate species within the same co-culture tank and never states a combined/system-level figure for any of these metrics — only per-species values (Table 3, Table 4). Species-level breakdown for TC5: Nile tilapia final weight 98.0 +/- 12.5 g (initial 20.6 +/- 6.1 g), weight gain 77.4 +/- 5.8 (Table 4) / 77.4 +/- 5.0 (p.6 narrative, SD mismatch only, mean identical — WARN-MINOR) g (0-106d), DWG 0.73 +/- 0.03 g/day, length gain 8.38 +/- 3.5 cm, SGR 0.52 +/- 0.02 %/day, K 1.9 +/- 0.72, survival 96.8 +/- 4.2%. African catfish final weight 53.39 +/- 13.9 g (initial 19.29 +/- 4.8 g), weight gain 34.1 +/- 7.2 g, DWG 0.32 +/- 0.04 g/day, length gain 4.68 +/- 1.85 cm, SGR 0.24 +/- 0.01 %/day, K 1.3 +/- 0.19, survival 90.0 +/- 4.5% (Tables 3-4). | WARN-CHECK NO3-N vs NO3 (applies to all 5 trials): Table 9 header reads ‘NO3-’ (nitrate ion) without stating whether the colorimetric method reports total NO3- or NO3—N; the schema column expects NO3-N specifically. Recorded the table’s HP-compartment value as-is (1.25 +/- 0.08 mg/L); if the paper intended total nitrate ion rather than nitrate-N, true NO3-N would be roughly 4.43x smaller. Not resolvable from the text (Methods 2.2 only names colorimetric kits generically for TAN and NO2-, no method/brand named for nitrate itself). | WARN-CHECK water-quality compartment (applies to all 5 trials): Table 9 reports DO/pH/EC/NH3/NO3 separately for ‘HP’ (hydroponic/plant-bed) and ‘AP’ (fish-tank) sampling points within the same coupled system. For TC5: DO HP 6.39 +/- 1.13 vs AP 7.85 +/- 2.09 mg/L; pH HP 7.45 +/- 1.94 vs AP 7.01 +/- 1.42; EC HP 0.8517 +/- 0.0214 vs AP 0.8459 +/- 0.0149 dS/m (raw uS/cm: HP and AP as in Table 9); Water temperature HP 30.28 +/- 2.59 vs AP 30.13 +/- 3.78 degC; TAN/NH4-N HP 0.53 +/- 0.06 vs AP 0.97 +/- 0.05 mg/L; NO3-N HP 1.25 +/- 0.08 vs AP 5.95 +/- 1.30 mg/L. Per SCHEMA.md guidance to take the plant-bed/hydroponic-unit value when a paper reports separate per-compartment readings, the HP column is recorded in the Aq pH/Dissolved Oxigen/EC/Water temperature/TAN/NO2/NO3 cells above; the AP (fish-tank) compartment values given here are the alternate candidate, often materially higher (e.g. NH3/NH4+ and NO3-). | WARN-MINOR: Table 9’s ‘T Air’ column shows a dash (’—’) under HP for TC5 and a single value under AP only (30.50 +/- 4.52 degC) — air/ambient temperature was evidently measured once per day (Methods 2.2: ‘ambient air temperature was recorded daily using a mercury thermometer’), not per compartment; the AP-column placement appears to be a table-formatting artifact rather than a true fish-tank-specific air reading. Recorded in Average room Temperature as given. | WARN-CHECK Plant fresh weight: the paper never states a single cumulative fresh weight per plant for the full 7-week harvest window — only a per-week average (‘Marketable eggplant yield, g/plant/week’, Table 2: 55.1 +/- 6.3 for TC5) and a projected annualised area yield (‘g/m2/year’: 26073.0 +/- 83.6). Recorded the weekly per-plant average in Plant fresh weight and the annualised area yield in the AP column; both are the paper’s own stated figures, not derived. | NOT DERIVED, left NR: Replicates (n) — paper states ‘ten identical aquaponic units’ and ‘five treatment groups’ (implying up to 2 tanks/treatment) and separately refers to ‘each replicate’ when sampling fish (p.3), confirming replication exists, but never states the per-treatment replicate count as a number; 10 divided by 5 is not written in the paper and is not entered per the no-derivation rule. FCR (never reported for either species, anywhere in the paper — no total feed intake figure is given either). Total Feed (kg), Fish biomass created (kg), Water recycle rate, Water volume, Water type/classification, Daily water exchange rate, SPAD, Plant height, Leaf count, Plant dry matter, Tissue nitrate AP/HYD, FUE AP/HYD, WUE, Lat/Long (Bandundu, DRC — no coordinates given), N (feed nitrogen, %), K (feed potassium, %) — none stated. | NO COLUMN: feed composition beyond protein/phosphorus — 11.5% crude fat, 4% crude fiber, 8% ash, 0.2% sodium, 1.5% calcium (p.2, no dedicated columns). Fish flesh proximate composition (TC5): Nile tilapia moisture 87.6 +/- 1.3%, crude protein 24.2 +/- 2.3%, crude lipid 3.19 +/- 0.1%, ash 0.91 +/- 0.1%, crude fiber 1.34 +/- 0.1%; African catfish moisture 85.9 +/- 1.7%, crude protein 17.9 +/- 1.6%, crude lipid 6.28 +/- 0.5%, ash 0.94 +/- 0.0%, crude fiber 1.56 +/- 0.1% (Table 7; both species ns across treatments, p>0.05) — this is FISH tissue proximate data, not plant tissue, so it does not belong in plant.csv either; no fish-tissue-proximate column exists in trials.csv. Heavy metals in fish flesh (TC5, mg/kg): Nile tilapia Hg 0.015, Mn 0.093, Cu 0.035, Zn 0.024 (Pb/Cd/As non-detect); African catfish Hg 0.013, Mn 0.063, Cu 0.031, Zn 0.039 (Pb/Cd/As non-detect); all well below FAO (2020)/WHO (2006) safety limits (Table 8) — no dedicated heavy-metal column exists.