Simultaneous removal of toxic ammonia and lettuce cultivation in aquaponic system using microwave pyrolysis biochar
Metadata
- Cite key: suSimultaneousRemovalToxic2020
- Item type: Journal Article
- Authors: Man Huan Su, Elfina Azwar, YaFeng Yang, Christian Sonne, Peter Nai Yuh Yek, Rock Keey Liew, Chin Kui Cheng, Pau Loke Show, Su Shiung Lam
- Affiliation: Henan Province Engineering Research Center for Biomass Value-added Products, School of Forestry, Henan Agricultural University, Zhengzhou, China (Su, Yang, Lam); Pyrolysis Technology Research Group, Institute of Tropical Aquaculture and Fisheries (AKUATROP), Universiti Malaysia Terengganu, Kuala Nerus, Terengganu, Malaysia (Su, Azwar, Yek, Lam); Aarhus University, Department of Bioscience, Arctic Research Centre, Roskilde, Denmark (Sonne); University College of Technology Sarawak, Sibu, Sarawak, Malaysia (Yek); NV Western PLT, Georgetown, Pulau Pinang, Malaysia (Liew); Universiti Malaysia Pahang, Gambang Kuantan, Pahang, Malaysia (Cheng); University of Nottingham Malaysia, Semenyih, Selangor, Malaysia (Show)
- Journal: Journal of Hazardous Materials 396 (2020) 122610
- Date: 09/2020
- Date added: 2021-02-04 (per zotero-export.csv)
- DOI: 10.1016/j.jhazmat.2020.122610
- Funding: Henan Agricultural University, Universiti Malaysia Terengganu, and University College of Technology Sarawak (financial/technical support); Ministry of Education Malaysia, Fundamental Research Grant Scheme (Vot 59512); Universiti Malaysia Terengganu, Golden Goose Research Grant Scheme (Vot 55191)
- URL: https://doi.org/10.1016/j.jhazmat.2020.122610
- PDF:
Su et al. - 2020 - Simultaneous removal of toxic ammonia and lettuce .pdf
Opinion
A materials-science paper (microwave pyrolysis biochar as a novel biofilter carrier) wearing an aquaponics hat: the fish and lettuce are instrumentation for validating the biochar, not the object of study, and the paper is honest about that framing throughout. The core engineering result — higher biochar dose gives more biofilm mass, better ammonia/TSS removal, and better lettuce growth, in a clean dose-response across four levels — is plausible and internally coherent in its headline direction. But the design has a real limitation the authors never name: each of the four biochar levels was run in exactly one physical trough with one filter column (no independent replicate systems), so the “triplicate” (n=3) and “n=5” figures in the table footnotes are almost certainly repeated measurements/subsamples within that single unit, not independent experimental replicates — classic pseudoreplication for a paper this confident in its ANOVA/Tukey significance claims. On top of that, the paper contradicts its own numbers in two places that matter: catfish “weight gain” is stated as 1.3 g in the same paragraph that gives a 93 g final mean weight from a 10-15 g start (impossible as stated — almost certainly a mislabeled per-day rate), and Treatment C’s nitrate concentration is given as both 19.7 mg/L and 29.7 mg/L within the same paragraph. Neither undermines the paper’s central biochar-characterization contribution (BET surface area, pore structure, SEM before/after), but both should be checked against the Supplementary Material before citing the biology-side numbers.
Abstract
This study examined an aquaponic approach of circulating water containing ammonia excretions from African catfish grown in an aquaculture tank for bacterial conversion into nitrates, which then acted as a nutrient substance to cultivate lettuce in hydroponic tank. We found that microwave pyrolysis biochar (450 g) having microporous (1.803 nm) and high BET surface area (419 m2/g) was suitable for use as biological carrier to grow nitrifying bacteria (63 g of biofilm mass) that treated the water quality through removing the ammonia (67%) and total suspended solids (68%), resulting in low concentration of remaining ammonia (0.42 mg/L) and total suspended solid (59.40 mg/L). It also increased the pH (6.8), converted the ammonia into nitrate (29.7 mg/L), and increased the nitrogen uptake by the lettuce (110 mg of nitrogen per plant), resulting in higher growth in lettuce (0.0562 %/day) while maintaining BOD5 level (3.94 mg/L) at acceptable level and 100% of catfish survival rate. Our results demonstrated that microwave pyrolysis biochar can be a promising solution for growing nitrifying bacteria in aquaponic system for simultaneous toxic ammonia remediation and generation of nitrate for growing vegetable in aquaculture industry.
Summary
The authors built a single-fish-tank, four-trough nutrient film technique (NFT) aquaponic system at Universiti Malaysia Terengganu, stocking one 110 L tank with 30 juvenile African catfish (Clarias gariepinus, 10-15 g initial) and running its effluent through four parallel biological filter columns loaded with 0 g (Control), 150 g (Treatment A), 300 g (Treatment B), or 450 g (Treatment C) of a microwave-pyrolysis palm-kernel-shell biochar, each feeding its own NFT trough of 10 lettuce (Lactuca sativa) seedlings, over 70 days. The paper’s primary contribution is materials-science: characterizing the biochar itself (BET surface area 419 m2/g, pore width 1.803 nm, comparing favourably against three literature biochars) and showing that biofilm mass, ammonia removal, TSS removal, and water pH all increased monotonically with biochar dose (0 to 62.7 g biofilm; 0 to 66.7% ammonia removal; 24 to 68% TSS removal across A/B/C). Lettuce growth (height, final biomass, relative growth rate) and nitrate accumulation also increased with biochar dose, and the shared catfish population reached 100% survival with a stated FCR of 1.4 and SGR of 3%/day. The paper positions microwave pyrolysis biochar as a comparably effective, and potentially more scalable, alternative to other biological-carrier media (oyster shell, gravel, LECA) reported in prior aquaponics literature. Because each biochar level was tested in only one physical trough (not independently replicated troughs), the statistical treatment (one-way ANOVA + Tukey on n=3/n=5 subsamples) is likely evaluating variability within a single system rather than between independent systems — a limitation not discussed by the authors. The paper also contains two internal numerical inconsistencies (catfish weight gain; Treatment C nitrate) detailed in the Extraction notes below.
Experiment data
- Location: Universiti Malaysia Terengganu (UMT), Kuala Nerus, Terengganu, Malaysia (Section 2.5)
- Design: Single aquaculture unit (110 L catfish tank) feeding four NFT hydroponic troughs via four biological filter columns loaded with different amounts of microwave pyrolysis biochar (0 g Control, 150 g Treatment A, 300 g Treatment B, 450 g Treatment C); each biochar level run in one physical trough + filter column, no independently replicated systems per level. One-way ANOVA + Tukey’s test for pairwise comparisons.
- Replicates / n: Table 3 footnote states water-quality values are “mean from triplicate data (n=3)”; Table 4 footnote states plant growth values are means of n=5. Both almost certainly refer to repeated samples/subsamples from the single trough per treatment, not independent replicate systems — see Extraction notes on the
quasi-experimentclassification.- Duration: 70 days (Section 2.5)
- Organisms: Lettuce (Lactuca sativa) / African catfish (Clarias gariepinus)
- Statistics: One-way ANOVA (alpha=0.05) + Tukey’s test; software not named
- Ammonia removal: dose-responsive, Control (baseline) -> A 27.78% -> B 43.65% -> C 66.66% (Table 3; Section 3.2.4)
- Biofilm mass formed on biochar (70 d): Control 0 g -> A 18.3 g -> B 32.3 g -> C 62.7 g (Table 2)
- Lettuce final biomass per plant: Control 1.42+/-0.11 g -> A 1.58+/-0.11 g -> B 1.76+/-0.15 g -> C 2.05+/-0.13 g (Table 4; fresh-vs-dry basis unclear, see Extraction notes)
- Catfish: 100% survival, FCR 1.4, SGR 3%/day, final mean weight 93 g (Section 3.3.2); “weight gain” figure internally inconsistent with these, see Extraction notes
Biochar as biological filter media (dose-response)
This paper: The paper’s central finding is a monotonic dose-response: higher microwave-pyrolysis biochar amount in the filter column produced more nitrifying biofilm (0/18.3/32.3/62.7 g dry mass across Control/A/B/C, Table 2), higher water pH (6.66/6.72/6.77/6.84), better ammonia removal (N/A/27.78%/43.65%/66.66%), better TSS removal (N/A/24.35%/45.71%/68.35%), and more nitrate generation from the same ammonia input (Table 3). Only Treatment C’s TSS (59.40 mg/L) fell within the paper’s cited acceptable freshwater range of 25-80 mg/L (Alabaster, 1982); Treatment C also had the highest BOD5 (3.94 mg/L), attributed to more bacterial oxygen demand on the larger biochar surface, but still within the acceptable 30 mg/L ceiling (Boyd, 2003). The biochar itself (produced from palm kernel shell via microwave pyrolysis + steam activation) achieved a BET surface area of 419 m2/g and pore width of 1.803 nm (Table 1), which the authors present as favourable versus three literature biochars (286-404 m2/g).
Compared with:
- #todo Khiari et al. 2020 — bamboo biochar from conventional pyrolysis in a tilapia/greenhouse-plant aquaponic system, but scope limited to turbidity removal only; the explicit gap this paper positions itself against for microwave-pyrolysis biochar specifically (p.3)
- #todo Enduta et al. 2011 — gravel biofilter + water spinach, TAN removal 78.32-85.48% — higher than this paper’s biochar-based removal (66.6% at best), cited in Table 5 comparison (p.8)
- #todo Yen and Chou 2016 — oyster shell biofilter + water spinach, NH3 removal 84.9%, attributed to the shell’s layer-by-layer roughened surface and water spinach’s long roots — higher removal than this paper’s biochar, cited in Table 5 (p.8)
- #todo Graber and Junge 2009 — LECA biofilter + tomato, TN removal 69% — comparable to this paper’s biochar performance, cited in Table 5 (p.8)
- #todo Khalil et al. 2018 — biochar removed up to 69.5% ammonium from fish farming in Egypt (p.2)
- #todo Lan et al. 2016 — rice-husk biochar + fish feed improved striped catfish growth 28% while reducing NH3/NO2/COD 48/31/30% (p.2)
- #todo Li et al. 2019 — pilot-scale aquaponics with hydroponics + immobilized biofilm treatment, cited re: pore/surface-area effects on AOB colonization (p.7)
Lettuce growth and nitrogen uptake
This paper: Lettuce height, final biomass per plant, and relative growth rate (RGR) all increased with biochar dose: height 13.20/14.80/16.32/19.04 cm and RGR 0.0478/0.0493/0.0508/0.0562 %/day across Control/A/B/C (Table 4). Total nitrogen uptake was reported only for Treatment C (110 mg N/plant, Section 3.3.1), with the other three treatments’ values deferred to Supplementary Material S5 not available in this PDF. The authors attribute Control’s smaller weight gain to N and K deficiency. Whether Table 4’s biomass figures are fresh or dry weight is not stated by the paper — see Extraction notes.
Compared with: (no external literature comparison given for lettuce growth/yield specifically; Table 5’s literature comparison is scoped to nutrient-removal efficiency, not plant growth)
Fish growth and water quality (shared system)
This paper: All 30 catfish survived (100%), with FCR 1.4, SGR 3%/day, and final mean weight/length of 93 g / 13 cm from a 10-15 g / unstated initial length start (Section 3.3.2) — these are single, paper-wide figures since one shared fish tank feeds all four troughs. DO (4.75-4.77 mg/L) and water temperature (27.59+/-0.82 C) were effectively constant across treatments, as expected for one shared tank; only pH, ammonia, and nitrate varied meaningfully by treatment, tracking biochar dose as described above. The stated “weight gain of 1.3 g” figure (repeated as “1.33 g” in the Conclusion) is inconsistent with the paper’s own final/initial weight figures — see Extraction notes.
Compared with:
- #todo Joseph, Richard and Daniel 1993; Towers 2014 — cited for the catfish DO (>4 mg/L) and temperature (26-32C) optimum ranges used as this paper’s own literature-cited “acceptable range” benchmark (p.6)
- #todo Maarif 2016 — cited for the catfish pH optimum (6-8) benchmark (p.5)
- #todo Bakhsh and Chopin 2011 — water quality/nutrient aspects in recirculating aquaponic production of freshwater prawn and lettuce, listed in References but not discussed in running text with a specific comparative figure
Citations to chase
- #todo Khiari Z, Alka K, Kelloway S, Mason B, Savidov N (2020) — Integration of Biochar Filtration into Aquaponics: Effects on Particle Size Distribution and Turbidity Removal, Agricultural Water Management 229:1-10
- #todo Enduta A, Jusoh A, Ali N, Wan Nik WB (2011) — Nutrient removal from aquaculture wastewater by vegetable production in aquaponics recirculation system, Desalination and Water Treatment 32:422-430 — already extracted directly in this vault, see
endutaNutrientRemovalAquaculture2011 - #todo Yen HY, Chou JH (2016) — Water purification by oyster shell bio-medium in a recirculating aquaponic system, Ecological Engineering 95:229-236
- #todo Graber A, Junge R (2009) — Aquaponic systems: Nutrient recycling from fish wastewater by vegetable production, Desalination 246:147-156
- #todo Khalil A, Sergeevich N, Borisova V (2018) — Removal of ammonium from fish farms by biochar obtained from rice straw, Adsorption Science & Technology 36:1294-1309
- #todo Lan TT, Preston T, Leng R (2016) — Feeding biochar or charcoal increased the growth rate of striped catfish and improved water quality, Livestock Research for Rural Development 28:1-5
- #todo Li C, Zhang B, Luo P, Shi H, Li L, Gao Y, Lee CT, Zhang Z, Wu WM (2019) — Performance of a pilot-scale aquaponics system using hydroponics and immobilized biofilm treatment for water quality control, Journal of Cleaner Production 208:274-284
- #todo Maarif AF (2016) — System monitoring and controlling water nutrition aquaponics using arduino uno based web server, Kinetik 1:39-46
- #todo Lam SS, Ma NL, Jusoh A, Ambak MA (2015) — Biological nutrient removal by recirculating aquaponic system: Optimization of the dimension ratio between the hydroponic & rearing tank components, International Biodeterioration & Biodegradation 102:107-115 — same author group’s own prior aquaponics work
Extraction notes
Type classification (judgment call): Recorded as quasi-experiment, not experiment. Section 2.1 describes “four identical columns containing different amounts (0 g, 150 g, 300 g, 450 g) of microwave pyrolysis biochar,” i.e. exactly one physical trough + filter column per biochar level — there is no independent physical replication of the treatment unit itself, and no statement that biochar amount was randomly assigned to trough (the four levels are fixed/ordered, not randomised). The “triplicate data (n=3)” in Table 3’s footnote and “n=5” in Table 4’s footnote almost certainly refer to repeated water samples / plant subsamples drawn from within that single trough over time, not independent replicate systems per treatment — classic pseudoreplication if treated as true experimental-unit replication. Per SCHEMA.md Part 1 rule 2 (“Treatments without randomisation or true replication -> quasi-experiment”), this was judged the better fit despite the paper’s own confident ANOVA/Tukey framing. This is a judgment call the paper itself never discusses or names as a limitation.
Trial structure: Four trials.csv rows, one per biochar-amount treatment (T1=Control/0g, T2=Treatment A/150g, T3=Treatment B/300g, T4=Treatment C/450g). Unlike most papers in this vault, there is no hydroponic-only control arm at all — all four rows are aquaponic (fed by the same catfish-tank effluent), varying only in filter media amount. HYD-side columns (Tissue nitrate HYD, FUE HYD, HYD) are recorded NA throughout per SCHEMA.md’s convention for papers with no hydroponic control, and T1 (Control) serves as this paper’s own internal reference point rather than a separate HYD-column value. All fish-related columns (FCR, SGR, survival, feed regime, stocking density, duration) are identical across all four rows since one shared fish tank feeds all four troughs.
⚠️WARN-BLOCK Fish weight gain (p.7-8). Results (3.3.2) states “the catfish have a weight gain of 1.3 g… The mean body weight and body length of catfish achieved at 93 g and 13 cm respectively,” and the Conclusion (Section 4) restates “1.33 g.” Both figures are mutually consistent (MINOR rounding only) but are arithmetically impossible given the same paragraph’s own final weight (93 g) and Methods’ initial stocking weight (10-15 g, Section 2.3): a fish going from >=10 g to 93 g implies a gain of at least ~78-83 g, not 1.3 g. Recomputed only as a check, never entered in a cell: (93-12.5)/70 =~1.15 g/day, close to the stated “1.3 g” — suggesting “1.3 g” may be a mislabeled daily rate rather than the total per-fish gain the Fish weight gain column asks for, but the paper never states this and it is our own reconstruction, not a value the paper gives for that definition. No basis to confidently prefer either reading. Recorded UNCLEAR in all four trials’ Fish weight gain cell (shared fish population). Does not affect Fish size initial/final, SGR, or FCR, all independently and unambiguously stated in the same paragraph.
⚠️WARN-CHECK NO3-N, Treatment C only (p.7, Section 3.2.5 and Abstract). Table 3 (captioned “mean from triplicate data”) gives Treatment C’s nitrate trial mean as 19.69+/-0.06 mg/L, matching Section 3.2.5’s first sentence (“up to 19.7 mg/L”). The very next sentence of the same paragraph instead states “Treatment C (450 g) produced the highest nitrate concentration (29.7 mg/L/week)” — restated identically in the Abstract. Neither passage labels which figure is the trial mean versus a final/peak value; Fig. 5’s stated range (“2.1 to 34.5 mg/L over the experiment period”) is circumstantially consistent with 29.7 being closer to a late-experiment reading, but per SCHEMA.md this was not used to resolve anything (never read a value off a figure). T4’s NO3-N cell uses Table 3’s dispersion-bearing trial mean (19.69+/-0.06); the repeated “29.7 mg/L” figure is preserved in the row’s Experimental Remarks. Readers citing this paper’s headline “29.7 mg/L” nitrate figure should note it is not the same quantity as the trial-mean NO3-N recorded in T4.
⚠️WARN-CHECK Plant fresh vs. dry weight, all four trials (Section 2.5.2 vs. Table 4). Methods states both fresh and dry weight were measured (oven-dried at 80C), but Table 4’s “Initial/Final biomass per plant” values are never labelled fresh or dry. “Total lettuce harvested” (a whole-trough total, conventionally weighed fresh at picking) divides cleanly by 10 plants/trough to reproduce Table 4’s per-plant figures for Control/A/B exactly, favouring a fresh-weight reading; but the paper’s own RGR formula (Eq. 4) is explicitly defined for dry weight, and Table 4’s stated RGR values appear (with an apparent factor-of-100 arithmetic slip, WARN-MINOR, no cell affected since RGR has no dedicated column) to have been computed from these same biomass figures, weakly favouring a dry-weight reading. Recorded under Plant fresh weight (fresh judged the better-supported basis) with this caveat in each trial’s remarks; Plant dry matter left NR.
⚠️WARN-MINOR Treatment C total-harvest arithmetic. 21.5 g total harvested / 10 plants = 2.15 g, versus Table 4’s stated 2.05+/-0.13 g “Final biomass per plant” for Treatment C — a ~5% mismatch, unlike Control/A/B which match exactly when divided by 10. Recorded Table 4’s dispersion-bearing 2.05+/-0.13 as the more structured source; does not change any qualitative ranking (Treatment C remains highest-yielding either way).
⚠️WARN-MINOR RGR arithmetic (paper-internal, no cell affected). Recomputing Eq. 4’s RGR formula from Table 4’s own Control biomass values (0.05 g initial, 1.42 g final, 70 days) gives 4.78 %/day, but Table 4 states RGR = 0.0478 %/day for Control — off by exactly a factor of 100 throughout the RGR row. No dedicated schema column exists for RGR, so no cell is affected; noted for completeness and as supporting context for the fresh-vs-dry-weight WARN-CHECK above.
TAN/NH4-N terminology (judgment call, not a contradiction): Table 3’s column is labelled “Ammonia (mg/L),” measured via the salicylate method (HACH DR2700 kit), which commonly reports total ammonia nitrogen. The paper never uses the term “TAN” itself. Mapped to the schema’s TAN/NH4-N column on this basis; flagged as a terminology judgment call rather than a numeric contradiction.
[not reported] fields, grouped (apply to all four trials unless noted):
- Fish: Fish Category, Total Feed (kg), Fish biomass created (kg), N/P/K feed composition (only crude protein/fibre/fat given) — none stated as figures; computing them from stocking count/ration % would be derivation, not performed.
- Water: Water volume in the system (component volumes given individually — 110 L catfish tank, 30 L mechanical filter — but sump/biofilter/trough fill volumes never stated or summed), Water type, Water classification, EC, NO2-N (nitrite is never measured in this paper — Section 2.5.1 lists only DO, temperature, pH, ammonia, nitrate, BOD5, TSS as measured parameters).
- Plant: Plant Category, Leaf count, SPAD, Plants/m2 (trough dimensions and seedling count given but density never stated as a ratio; computing it would be derivation), Tissue nitrate AP (no tissue nitrate assay performed; only total N content, and only for Treatment C).
- Site: Lat/Long, Average room Temperature — not stated anywhere in the paper.
- Paired presence/detail fields: Iron supplemented, Remineralization, pH Buffers, Climate control, Artificial Lighting, Nutrient supplemented — all silent in the paper (not explicitly stated absent, so
NRrather thanNper CLAUDE.md’s N-vs-NR distinction, even though the overall design implies no external fertiliser is used).
NO COLUMN items (full detail in each trial row’s Experimental Remarks): Table 1 biochar material characterization (BET surface area, pore volume/width, biochar yield, vs. 3 literature biochars); Table 2 biofilm dry-mass formation per trial (0/18.3/32.3/62.7 g); Table 3 BOD5 and TSS trial means and % removal per trial; Table 4 initial biomass per plant, weight gain per plant, RGR per plant, and total lettuce harvested per trial; Table 5’s cross-study biological-carrier comparison (routed to Citations to chase instead, as it is entirely other studies’ data plus this paper’s own headline removal percentages already captured in trials.csv cells).
plant_measurements.csv scope decision: Only one analyte from this paper meets plant_measurements.csv’s tissue-analyte scope: total plant nitrogen content (110 mg N/plant), reported only for Treatment C (Section 3.3.1), with Control/A/B deferred to an unavailable Supplementary Material file. The paper’s Section 2.5.3 describes an ICP-OES elemental-panel methodology (implying P, K, Ca, Mg, etc. were also measured) but never reports any resulting values anywhere in the main text or tables for any treatment — not extracted, since there is nothing to extract. No water-chemistry data was misrouted here; Table 3’s water panel (pH, DO, BOD5, TSS, ammonia, nitrate) is entirely in trials.csv per SCHEMA.md.
Tags judgment call: Tagged Meta/Fish/African-Catfish (reusing the vault’s existing spelling/casing convention, e.g. oladimejiEffectsDifferentGrowth2020) and Meta/Plant/Lettuce (reusing the existing tag, e.g. pantanellaAquaponicsHydroponicsProduction2012). Meta/Region/Southeast-Asia reused from endutaNutrientRemovalAquaculture2011, another Malaysia-based aquaponics paper already in this vault, rather than inventing a new region leaf. Meta/Type/Quasi-experiment reused from existing vault papers (e.g. delaideLettuceLactucaSativa2016, goddekComparisonLactucaSativa2018). No new tag facets introduced.
Quality scoring: 1 ⚠️BLOCK (Fish weight gain) and 0 ⚠️MATERIAL -> caution per SCHEMA.md’s threshold table (“1 BLOCK… = caution”). The two ⚠️CHECK items (NO3-N; fresh/dry weight) and two ⚠️MINOR items do not affect this score per SCHEMA.md.
PDF quality: Clean, fully extractable text layer throughout (10 pages, standard two-column Elsevier typesetting); no OCR issues. All tables (1-5) and both figures with axis labels were machine-readable; no values were read off a chart (Figs. 3-5 are described narratively in the text using the same ranges/trends noted above, not used as a numeric source beyond what running text states).
New wikilink targets introduced: All nine author names (no existing author notes found in the vault for this paper’s author list). [[African catfish (Clarias gariepinus)]] and [[Lettuce (Lactuca sativa)]] reuse the vault’s established organism-naming convention (species + common name in parentheses, matching e.g. pantanellaAquaponicsHydroponicsProduction2012’s [[Nile tilapia (Oreochromis niloticus)]]).
Source: Su et al. - 2020 - Simultaneous removal of toxic ammonia and lettuce .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
suSimultaneousRemovalToxic2020-T1
Fish
| Field | Value |
|---|---|
| Fish | African catfish (Clarias gariepinus) |
| Initial Stock density | 4 |
| FCR | 1.4 |
| SGR | 3 |
| Protein | 33 |
| % of body weight | 3 |
| Fish size initial | 10-15 (range at stocking, not stated as a mean; Methods 2.3, p.3) |
| Fish size final | 93 |
| Feed routine | Twice daily (9:00 a.m. and 7:00 p.m.); uneaten feed removed after 10 minutes to avoid polluting the water (Methods 2.3, p.3) |
| Feed regime | Commercial fish feed pellets (crude protein 33%, fibre 6%, fat 6%), fed at 3 wt% of total body weight (wet weight) daily |
| Fish survival rate | 100 |
| Fish trial duration (days) | 70 |
Water
| Field | Value |
|---|---|
| Water recycle | 9 |
| Daily Water exchange rate | 0 |
| Aq pH | 6.66 +/- 0.15 |
| pHOptimal | 6-8 (accepted pH range for catfish culture, cited from Maarif 2016, p.5; literature-cited acceptable range, not this paper’s own measured value) |
| Dissolved Oxigen | 4.75 +/- 0.36 |
| Water temperature | 27.59 +/- 0.82 |
| TAN / NH4-N | 1.26 +/- 0.47 |
| NO3-N | 8.16 +/- 4.89 |
Plant
| Field | Value |
|---|---|
| Plant | Lettuce (Lactuca sativa) |
| Details | Seeds germinated in seedling-raising plates one week before experiment start; 10 seedlings (~4 cm) transplanted per trough at experiment day 0; grown soilless on lightweight expanded clay aggregate (LECA) in NFT troughs; harvested at day 70 (end of experiment) |
| Days Plant after transplant | 70 |
| Plant height | 13.20 +/- 0.65 |
| Plant fresh weight | 1.42 +/- 0.11 |
System & Setup
| Field | Value |
|---|---|
| System type | Nutrient film technique (NFT) |
| Media Details | LECA (lightweight expanded clay aggregate) growing medium in NFT troughs (shared across all four treatments); biological filter column: 0 g biochar (Control, empty column, no added biological carrier) |
| Biological system already in use | Y (Biological filter column present but empty (0 g biochar) — Control treatment, no added biological carrier media) |
| Air supplement | Y (Air blower aerates the catfish tank to meet catfish oxygen demand (Fig. 1, Section 2.1); flow rate/model not stated) |
| Equipment | YSI 550A dissolved oxygen meter (DO/temperature/pH); HACH DR2700 test kit, salicylate method (ammonia, nitrate); Standard Methods per APHA 1995 (BOD5, TSS); Micromeritics ASAP 2020 automatic sorption analyser (biochar BET surface area/pore width/pore volume); JEOL JSM-6360LA SEM (biochar surface morphology); Sartorius AX224 electronic balance (biofilm dry mass); submersible pump, 2 m3/h capacity (sump-to-biofilter lift); inductively coupled plasma-optical emission spectrometry (ICP-OES; plant nutrient ash digestion method described, Section 2.5.3, but no result values reported in the main text); FlashEA 1112 CHNS elemental analyser (plant nitrogen content, Section 2.5.4); fish measuring board and A3360-LT5001 electronic weighing balance (Smith model; fish length/weight) |
| Control Parameters | Literature-cited acceptable ranges, not experimentally imposed setpoints: DO >4 mg/L and temperature 26-32C for catfish growth (Joseph et al. 1993; Towers 2014); pH 6-8 for catfish culture (Maarif 2016); TSS 25-80 mg/L acceptable for fish freshwater (Alabaster 1982); BOD5 <=30 mg/L acceptable for aquaculture (Boyd 2003) |
| Combination | African catfish (Clarias gariepinus) and lettuce (Lactuca sativa) in a single-fish-tank, four-trough NFT aquaponic system; amount of biological filter media (microwave pyrolysis biochar: 0/150/300/450 g) is the sole experimental treatment variable; no hydroponic-only control arm |
Site
| Field | Value |
|---|---|
| Region | Southeast Asia |
| Country | Malaysia |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | g/plant (fresh-vs-dry ambiguous, see remarks); cm (height); mg/L (water quality); mg N/plant (nitrogen uptake, Treatment C only) |
| Statistic Details | One-way ANOVA (alpha=0.05); Tukey’s test for pairwise treatment comparisons (Methods 2.6, p.4) |
| Statistically analysed | Y |
| Replicates (n) | 3 (water quality, Table 3 footnote); 5 (plant growth, Table 4 footnote); single physical trough/filter column per treatment — no independent replication of the treatment unit itself, see Extraction notes/T1 remarks |
| AP | 1.42 +/- 0.11 |
Experimental Remarks: TRIAL DEFINITION: T1 = Control treatment (0 g of microwave pyrolysis biochar in the biological filter column) — the baseline arm of this paper’s single experimental factor (biochar amount: 0/150/300/450 g), NOT a hydroponic control. All four trial rows in this paper (T1-T4) share one physical fish tank, one catfish population, and the same overall aquaponic system design (Section 2.1); each row differs only in the amount of biochar loaded into its own biological filter column and NFT trough. There is no hydroponic-only arm in this paper at all — ‘Control’ still receives fish-tank effluent, it simply has an empty (0 g) filter column, so it is recorded here as an aquaponic treatment (HYD/HYD-side columns = NA throughout, per SCHEMA.md’s convention for papers with no hydroponic control), with T1 (Control) serving as this paper’s own internal reference point for T2-T4. | DESIGN CAVEAT (affects type classification): each of the four biochar levels was tested in exactly ONE physical trough + filter column (Section 2.1: ‘four identical columns containing different amounts… denoted as control, treatment A, B and C’) — there is no independent physical replication of the treatment unit itself. The n=3 given in Table 3’s footnote (‘mean from triplicate data’) and n=5 in Table 4’s footnote almost certainly refer to repeated water-sample measurements / plant subsamples drawn from within that single trough over time, not independent replicate troughs per treatment. Combined with the absence of any stated randomisation of biochar amount to trough, this paper is recorded as ‘quasi-experiment’ rather than ‘experiment’ per SCHEMA.md Part 1 rule 2 (‘Treatments without randomisation or true replication -> quasi-experiment’). Flagged as a judgment call since the paper never itself discusses this limitation. | WARN-BLOCK Fish weight gain (p.7-8, Section 3.3.2 and Section 4; shared fish population, same underlying facts for all four trials since there is only one fish tank). Results states: ‘The results showed that the catfish have a weight gain of 1.3 g, length gain of 0.2 cm, specific growth rate of 3% per day and feed conversion ratio of 1.4. The mean body weight and body length of catfish achieved at 93 g and 13 cm respectively.’ The Conclusion (Section 4, p.8) restates this as ‘higher survival growth rate of catfish (1.33 g)’. Both figures (1.3 g / 1.33 g) are internally consistent with each other (WARN-MINOR rounding only, between them) but are IMPOSSIBLE given the same paragraph’s own stated final mean body weight (93 g) and Methods’ stated initial stocking weight (10-15 g, Section 2.3): a fish cannot go from >=10 g to 93 g while gaining only 1.3 g total (that arithmetic implies a gain of at least ~78-83 g, not 1.3 g). Recomputed check only, never entered in any cell: (93-12.5)/70 = ~1.15 g/day, close enough to the stated ‘1.3 g’ that ‘1.3 g’ most plausibly represents a DAILY weight-gain rate rather than the TOTAL per-fish weight gain the Fish weight gain column asks for (‘Per fish, as stated’) — but the paper never labels it as a rate, and this reading is our own reconstruction, not a value the paper itself states for the column’s actual definition. No basis to confidently prefer either reading as ‘the’ total weight gain. Cell recorded UNCLEAR. Affects: interpretation of the fish-growth narrative only; does not affect Fish size initial/final, SGR, or FCR cells (all independently and unambiguously stated in the same paragraph). | NOT DERIVED, left NR: Total Feed (kg) (feeding rate given as 3 wt% of body weight/day, twice daily, but no stated total kg consumed over 70 days); Fish biomass created (kg) (only individual mean weights given, no stated total tank biomass initial/final); N/P/K feed composition (only protein 33%/fibre 6%/fat 6% given); Plants/m2 (trough dimensions 10 m x 0.3 m and 10 seedlings/trough given, but density never stated as a ratio); Water volume in the system (component volumes given individually — catfish tank 110 L, mechanical filter/sedimentation tank 30 L, Section 2.1 — but sump tank, biological filter columns, and the four 10 m x 0.3 m x 0.3 m NFT troughs’ fill volumes are never stated or summed; NFT troughs use a thin nutrient film, not a full-depth fill, so trough dimensions cannot be read as a fill volume). | UNIT CONVERSION ONLY: none required in this row (all values as reported are already in the schema’s target units; feed ration and stocking density are given directly in wt%/kg m-3 form). | Fish size initial recorded as the stated range ‘10-15’ (g) — Methods 2.3 states juveniles had ‘an initial weight of 10-15 g’, not a single mean value; recorded as-is rather than averaging (averaging would be derivation). | Water temperature (27.59 +/- 0.82 C) and pHOptimal (6-8, literature-cited acceptable range for catfish per Maarif 2016, not this paper’s own measurement) are identical across all four trials, consistent with a single shared fish tank feeding all four troughs. | Daily Water exchange rate recorded as ‘0’ (not NR): Section 2.5 explicitly states ‘no water was exchanged in the fish tank’ during the 70-day experimental period, only freshwater topped up to replace evaporation/transpiration/sampling losses — an explicit statement of absence, not silence, so ‘0’ applies rather than NR per CLAUDE.md’s N-vs-NR distinction. | NO COLUMN (paper-wide items, recorded here to avoid repeating across all four rows — T2-T4 point back to this row for these): Table 1 biochar characterisation (this study’s palm-kernel-shell microwave-pyrolysis biochar: BET surface area 419 m2/g, total pore volume 0.188 cm3/g, pore width 1.803 nm, 29 wt% biochar yield from palm kernel shell — compared against 3 literature biochars, Table 1, p.4-5). Table 2 biofilm formation (dry biofilm mass gained on the biochar/filter column over the 70-day trial): Control 0 g, Treatment A 18.3 g, Treatment B 32.3 g, Treatment C 62.7 g — this row (Control) = 0 g. Table 3 BOD5 (mg/L, trial mean +/- SD): Control 1.51+/-0.25 — no dedicated schema column for BOD5. Table 3 TSS (mg/L): Control 187.70+/-38.52 — no dedicated column. Table 3 % removal (TSS, Ammonia): Control is the paper’s own 0%/N-A baseline against which A/B/C removal percentages are calculated. Table 4 Initial biomass per plant: 0.05 g; Weight gain per plant: 1.37 g (=Final 1.42 - Initial 0.05, internally consistent, the authors’ own arithmetic); RGR per plant: 0.0478 %/day; Total lettuce harvested (10 plants/trough): 14.2 g (14.2/10 = 1.42, exactly matching Table 4’s ‘Final biomass per plant’ for Control). | WARN-CHECK Plant fresh vs. dry weight, all four trials (Section 2.5.2 vs. Table 4, p.7). Methods 2.5.2 states lettuce growth analysis included ‘determination of the fresh AND dry weight (oven dried at 80C)’ — both were measured — but Table 4’s ‘Initial/Final biomass per plant’ and ‘Weight gain per plant’ rows are not labelled fresh or dry anywhere in the table, caption, or surrounding text. Two considerations pull in different directions: (1) ‘Total lettuce harvested’ (14.2/15.8/17.6/21.5 g for Control/A/B/C) divides cleanly by 10 plants/trough to reproduce Table 4’s ‘Final biomass per plant’ for Control, A, and B exactly (14.2/10=1.42; 15.8/10=1.58; 17.6/10=1.76 — all match), and a harvest total is conventionally weighed fresh, immediately at picking — this favours FRESH weight. (2) Eq. 4’s RGR formula is explicitly defined for DRY weight (‘W2 is the final dry weight of lettuce’), and Table 4’s own RGR row appears to have been computed from these same biomass numbers (recomputed check only: ln(1.42/0.05)/70*100 = 4.78, versus Table 4’s stated 0.0478 — an apparent factor-of-100 arithmetic slip in the paper’s own RGR column, WARN-MINOR, not affecting any trials.csv cell since RGR has no dedicated column) — this weakly favours DRY weight, or at least shows the authors treated these numbers as the RGR input regardless of label. No sentence resolves which weight basis Table 4 actually reports. Recorded in Plant fresh weight (the better-supported reading, given the harvest-total match) with this caveat; Plant dry matter left NR since no clearly distinct dry-weight or %-dry-matter figure exists. Affects: Plant fresh weight and Plant dry matter cells, all four trials; AP column (mirrors Plant fresh weight). Added to REVIEW.md by the batch merge step. | WARN-MINOR Treatment C total-harvest arithmetic (p.7, Table 4). ‘Total lettuce harvested’ for Treatment C = 21.5 g; divided by 10 plants/trough = 2.15 g, which does not exactly match Table 4’s stated ‘Final biomass per plant’ for Treatment C (2.05+/-0.13 g) — a ~0.10 g (~5%) discrepancy, unlike Control/A/B which match exactly (see above). Recorded Table 4’s dispersion-bearing 2.05+/-0.13 value (T4’s own Plant fresh weight cell) as the more structured, SD-carrying source; does not change any qualitative comparison (Treatment C remains the highest-yielding treatment either way). | Plant Category, Fish Category, Water type, Water classification: NR — the paper never applies a categorising term to any of these (per SCHEMA.md’s own-wording rule, not substituted). | Nutrient supplemented recorded NR, not N: the whole system design implies fish effluent is the sole nutrient source (no fertiliser is ever mentioned in Methods), but the paper never explicitly states ‘no additional fertiliser/nutrients were added’ — per CLAUDE.md, silence is NR, not N, even when a design strongly implies absence.
suSimultaneousRemovalToxic2020-T2
Fish
| Field | Value |
|---|---|
| Fish | African catfish (Clarias gariepinus) |
| Initial Stock density | 4 |
| FCR | 1.4 |
| SGR | 3 |
| Protein | 33 |
| % of body weight | 3 |
| Fish size initial | 10-15 (range at stocking, not stated as a mean; Methods 2.3, p.3) |
| Fish size final | 93 |
| Feed routine | Twice daily (9:00 a.m. and 7:00 p.m.); uneaten feed removed after 10 minutes to avoid polluting the water (Methods 2.3, p.3) |
| Feed regime | Commercial fish feed pellets (crude protein 33%, fibre 6%, fat 6%), fed at 3 wt% of total body weight (wet weight) daily |
| Fish survival rate | 100 |
| Fish trial duration (days) | 70 |
Water
| Field | Value |
|---|---|
| Water recycle | 9 |
| Daily Water exchange rate | 0 |
| Aq pH | 6.72 +/- 0.14 |
| pHOptimal | 6-8 (accepted pH range for catfish culture, cited from Maarif 2016, p.5; literature-cited acceptable range, not this paper’s own measured value) |
| Dissolved Oxigen | 4.77 +/- 0.38 |
| Water temperature | 27.59 +/- 0.82 |
| TAN / NH4-N | 0.91 +/- 0.44 |
| NO3-N | 12.49 +/- 7.14 |
Plant
| Field | Value |
|---|---|
| Plant | Lettuce (Lactuca sativa) |
| Details | Seeds germinated in seedling-raising plates one week before experiment start; 10 seedlings (~4 cm) transplanted per trough at experiment day 0; grown soilless on lightweight expanded clay aggregate (LECA) in NFT troughs; harvested at day 70 (end of experiment) |
| Days Plant after transplant | 70 |
| Plant height | 14.80 +/- 0.35 |
| Plant fresh weight | 1.58 +/- 0.11 |
System & Setup
| Field | Value |
|---|---|
| System type | Nutrient film technique (NFT) |
| Media Details | LECA (lightweight expanded clay aggregate) growing medium in NFT troughs (shared across all four treatments); biological filter column: 150 g microwave pyrolysis biochar (~1 cm particle size) |
| Biological system already in use | Y (150 g microwave pyrolysis biochar (~1 cm particle size) as biological carrier in filter column) |
| Air supplement | Y (Air blower aerates the catfish tank to meet catfish oxygen demand (Fig. 1, Section 2.1); flow rate/model not stated) |
| Equipment | YSI 550A dissolved oxygen meter (DO/temperature/pH); HACH DR2700 test kit, salicylate method (ammonia, nitrate); Standard Methods per APHA 1995 (BOD5, TSS); Micromeritics ASAP 2020 automatic sorption analyser (biochar BET surface area/pore width/pore volume); JEOL JSM-6360LA SEM (biochar surface morphology); Sartorius AX224 electronic balance (biofilm dry mass); submersible pump, 2 m3/h capacity (sump-to-biofilter lift); inductively coupled plasma-optical emission spectrometry (ICP-OES; plant nutrient ash digestion method described, Section 2.5.3, but no result values reported in the main text); FlashEA 1112 CHNS elemental analyser (plant nitrogen content, Section 2.5.4); fish measuring board and A3360-LT5001 electronic weighing balance (Smith model; fish length/weight) |
| Control Parameters | Literature-cited acceptable ranges, not experimentally imposed setpoints: DO >4 mg/L and temperature 26-32C for catfish growth (Joseph et al. 1993; Towers 2014); pH 6-8 for catfish culture (Maarif 2016); TSS 25-80 mg/L acceptable for fish freshwater (Alabaster 1982); BOD5 <=30 mg/L acceptable for aquaculture (Boyd 2003) |
| Combination | African catfish (Clarias gariepinus) and lettuce (Lactuca sativa) in a single-fish-tank, four-trough NFT aquaponic system; amount of biological filter media (microwave pyrolysis biochar: 0/150/300/450 g) is the sole experimental treatment variable; no hydroponic-only control arm |
Site
| Field | Value |
|---|---|
| Region | Southeast Asia |
| Country | Malaysia |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | g/plant (fresh-vs-dry ambiguous, see remarks); cm (height); mg/L (water quality); mg N/plant (nitrogen uptake, Treatment C only) |
| Statistic Details | One-way ANOVA (alpha=0.05); Tukey’s test for pairwise treatment comparisons (Methods 2.6, p.4) |
| Statistically analysed | Y |
| Replicates (n) | 3 (water quality, Table 3 footnote); 5 (plant growth, Table 4 footnote); single physical trough/filter column per treatment — no independent replication of the treatment unit itself, see Extraction notes/T1 remarks |
| AP | 1.58 +/- 0.11 |
Experimental Remarks: TRIAL DEFINITION: T2 = Treatment A (150 g of microwave pyrolysis biochar in the biological filter column). Same shared single fish tank/catfish population and overall system design as T1 (Control) — see T1’s Experimental Remarks for the full paper-wide judgment calls (quasi-experiment classification, all NOT DERIVED items, Daily Water exchange rate reasoning) and paper-wide WARN flags (Fish weight gain BLOCK; Plant fresh/dry weight CHECK), which apply identically to this row and are not re-typed here in full. Paired reference = T1 (Control, 0 g biochar), the paper’s own baseline aquaponic arm; no hydroponic control exists in this paper (HYD-side columns = NA). | NO COLUMN (this trial’s own values): Table 2 biofilm formation: 18.3 g dry biofilm mass gained on the biochar over 70 days (vs. Control’s 0 g, Treatment B’s 32.3 g, Treatment C’s 62.7 g). Table 3 BOD5: 2.63+/-0.45 mg/L. Table 3 TSS: 142.00+/-32.84 mg/L; TSS removal 24.35%; Ammonia removal 27.78% (paper states this is NOT significantly different from Control’s ammonia level, Section 3.2.4, p.7). Table 4: Initial biomass per plant 0.05 g; Weight gain per plant 1.53 g (=1.58-0.05, consistent); RGR per plant 0.0493 %/day; Total lettuce harvested 15.8 g (10 plants; 15.8/10=1.58, exactly matches Final biomass per plant, no discrepancy for this trial). | Ammonia (TAN/NH4-N cell) and Nitrate (NO3-N cell) values taken from Table 3’s trial means (0.91+/-0.44 and 12.49+/-7.14 respectively); Results 3.2.4/3.2.5 running text restates figures consistent with these for this specific trial (unlike Treatment C’s Nitrate figure — see T4’s WARN-CHECK).
suSimultaneousRemovalToxic2020-T3
Fish
| Field | Value |
|---|---|
| Fish | African catfish (Clarias gariepinus) |
| Initial Stock density | 4 |
| FCR | 1.4 |
| SGR | 3 |
| Protein | 33 |
| % of body weight | 3 |
| Fish size initial | 10-15 (range at stocking, not stated as a mean; Methods 2.3, p.3) |
| Fish size final | 93 |
| Feed routine | Twice daily (9:00 a.m. and 7:00 p.m.); uneaten feed removed after 10 minutes to avoid polluting the water (Methods 2.3, p.3) |
| Feed regime | Commercial fish feed pellets (crude protein 33%, fibre 6%, fat 6%), fed at 3 wt% of total body weight (wet weight) daily |
| Fish survival rate | 100 |
| Fish trial duration (days) | 70 |
Water
| Field | Value |
|---|---|
| Water recycle | 9 |
| Daily Water exchange rate | 0 |
| Aq pH | 6.77 +/- 0.14 |
| pHOptimal | 6-8 (accepted pH range for catfish culture, cited from Maarif 2016, p.5; literature-cited acceptable range, not this paper’s own measured value) |
| Dissolved Oxigen | 4.76 +/- 0.36 |
| Water temperature | 27.59 +/- 0.82 |
| TAN / NH4-N | 0.71 +/- 0.43 |
| NO3-N | 17.95 +/- 9.04 |
Plant
| Field | Value |
|---|---|
| Plant | Lettuce (Lactuca sativa) |
| Details | Seeds germinated in seedling-raising plates one week before experiment start; 10 seedlings (~4 cm) transplanted per trough at experiment day 0; grown soilless on lightweight expanded clay aggregate (LECA) in NFT troughs; harvested at day 70 (end of experiment) |
| Days Plant after transplant | 70 |
| Plant height | 16.32 +/- 0.72 |
| Plant fresh weight | 1.76 +/- 0.15 |
System & Setup
| Field | Value |
|---|---|
| System type | Nutrient film technique (NFT) |
| Media Details | LECA (lightweight expanded clay aggregate) growing medium in NFT troughs (shared across all four treatments); biological filter column: 300 g microwave pyrolysis biochar (~1 cm particle size) |
| Biological system already in use | Y (300 g microwave pyrolysis biochar (~1 cm particle size) as biological carrier in filter column) |
| Air supplement | Y (Air blower aerates the catfish tank to meet catfish oxygen demand (Fig. 1, Section 2.1); flow rate/model not stated) |
| Equipment | YSI 550A dissolved oxygen meter (DO/temperature/pH); HACH DR2700 test kit, salicylate method (ammonia, nitrate); Standard Methods per APHA 1995 (BOD5, TSS); Micromeritics ASAP 2020 automatic sorption analyser (biochar BET surface area/pore width/pore volume); JEOL JSM-6360LA SEM (biochar surface morphology); Sartorius AX224 electronic balance (biofilm dry mass); submersible pump, 2 m3/h capacity (sump-to-biofilter lift); inductively coupled plasma-optical emission spectrometry (ICP-OES; plant nutrient ash digestion method described, Section 2.5.3, but no result values reported in the main text); FlashEA 1112 CHNS elemental analyser (plant nitrogen content, Section 2.5.4); fish measuring board and A3360-LT5001 electronic weighing balance (Smith model; fish length/weight) |
| Control Parameters | Literature-cited acceptable ranges, not experimentally imposed setpoints: DO >4 mg/L and temperature 26-32C for catfish growth (Joseph et al. 1993; Towers 2014); pH 6-8 for catfish culture (Maarif 2016); TSS 25-80 mg/L acceptable for fish freshwater (Alabaster 1982); BOD5 <=30 mg/L acceptable for aquaculture (Boyd 2003) |
| Combination | African catfish (Clarias gariepinus) and lettuce (Lactuca sativa) in a single-fish-tank, four-trough NFT aquaponic system; amount of biological filter media (microwave pyrolysis biochar: 0/150/300/450 g) is the sole experimental treatment variable; no hydroponic-only control arm |
Site
| Field | Value |
|---|---|
| Region | Southeast Asia |
| Country | Malaysia |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | g/plant (fresh-vs-dry ambiguous, see remarks); cm (height); mg/L (water quality); mg N/plant (nitrogen uptake, Treatment C only) |
| Statistic Details | One-way ANOVA (alpha=0.05); Tukey’s test for pairwise treatment comparisons (Methods 2.6, p.4) |
| Statistically analysed | Y |
| Replicates (n) | 3 (water quality, Table 3 footnote); 5 (plant growth, Table 4 footnote); single physical trough/filter column per treatment — no independent replication of the treatment unit itself, see Extraction notes/T1 remarks |
| AP | 1.76 +/- 0.15 |
Experimental Remarks: TRIAL DEFINITION: T3 = Treatment B (300 g of microwave pyrolysis biochar in the biological filter column). Same shared fish tank/system as T1/T2 — see T1’s Experimental Remarks for the full paper-wide judgment calls and WARN flags (Fish weight gain BLOCK; Plant fresh/dry weight CHECK), which apply identically here. Paired reference = T1 (Control). No hydroponic control in this paper (HYD-side columns = NA). | NO COLUMN (this trial’s own values): Table 2 biofilm formation: 32.3 g dry biofilm mass over 70 days. Table 3 BOD5: 3.41+/-0.54 mg/L. Table 3 TSS: 101.90+/-34.34 mg/L; TSS removal 45.71%; Ammonia removal 43.65% (significantly different from Control, Section 3.2.4, p.7). Table 4: Initial biomass per plant 0.05 g; Weight gain per plant 1.71 g (=1.76-0.05, consistent); RGR per plant 0.0508 %/day; Total lettuce harvested 17.6 g (10 plants; 17.6/10=1.76, exactly matches Final biomass per plant, no discrepancy for this trial). | Nitrate (NO3-N cell, 17.95+/-9.04) is Table 3’s trial mean for Treatment B; the paper’s nitrate-increase narrative (Section 3.2.5, ‘up to 19.7 mg/L’) is stated for Treatment C, not B, so the WARN-CHECK described under T4 does not apply to this row.
suSimultaneousRemovalToxic2020-T4
Fish
| Field | Value |
|---|---|
| Fish | African catfish (Clarias gariepinus) |
| Initial Stock density | 4 |
| FCR | 1.4 |
| SGR | 3 |
| Protein | 33 |
| % of body weight | 3 |
| Fish size initial | 10-15 (range at stocking, not stated as a mean; Methods 2.3, p.3) |
| Fish size final | 93 |
| Feed routine | Twice daily (9:00 a.m. and 7:00 p.m.); uneaten feed removed after 10 minutes to avoid polluting the water (Methods 2.3, p.3) |
| Feed regime | Commercial fish feed pellets (crude protein 33%, fibre 6%, fat 6%), fed at 3 wt% of total body weight (wet weight) daily |
| Fish survival rate | 100 |
| Fish trial duration (days) | 70 |
Water
| Field | Value |
|---|---|
| Water recycle | 9 |
| Daily Water exchange rate | 0 |
| Aq pH | 6.84 +/- 0.12 |
| pHOptimal | 6-8 (accepted pH range for catfish culture, cited from Maarif 2016, p.5; literature-cited acceptable range, not this paper’s own measured value) |
| Dissolved Oxigen | 4.76 +/- 0.35 |
| Water temperature | 27.59 +/- 0.82 |
| TAN / NH4-N | 0.42 +/- 0.31 |
| NO3-N | 19.69 +/- 0.06 |
Plant
| Field | Value |
|---|---|
| Plant | Lettuce (Lactuca sativa) |
| Details | Seeds germinated in seedling-raising plates one week before experiment start; 10 seedlings (~4 cm) transplanted per trough at experiment day 0; grown soilless on lightweight expanded clay aggregate (LECA) in NFT troughs; harvested at day 70 (end of experiment) |
| Days Plant after transplant | 70 |
| Plant height | 19.04 +/- 1.15 |
| Plant fresh weight | 2.05 +/- 0.13 |
System & Setup
| Field | Value |
|---|---|
| System type | Nutrient film technique (NFT) |
| Media Details | LECA (lightweight expanded clay aggregate) growing medium in NFT troughs (shared across all four treatments); biological filter column: 450 g microwave pyrolysis biochar (~1 cm particle size) |
| Biological system already in use | Y (450 g microwave pyrolysis biochar (~1 cm particle size) as biological carrier in filter column) |
| Air supplement | Y (Air blower aerates the catfish tank to meet catfish oxygen demand (Fig. 1, Section 2.1); flow rate/model not stated) |
| Equipment | YSI 550A dissolved oxygen meter (DO/temperature/pH); HACH DR2700 test kit, salicylate method (ammonia, nitrate); Standard Methods per APHA 1995 (BOD5, TSS); Micromeritics ASAP 2020 automatic sorption analyser (biochar BET surface area/pore width/pore volume); JEOL JSM-6360LA SEM (biochar surface morphology); Sartorius AX224 electronic balance (biofilm dry mass); submersible pump, 2 m3/h capacity (sump-to-biofilter lift); inductively coupled plasma-optical emission spectrometry (ICP-OES; plant nutrient ash digestion method described, Section 2.5.3, but no result values reported in the main text); FlashEA 1112 CHNS elemental analyser (plant nitrogen content, Section 2.5.4); fish measuring board and A3360-LT5001 electronic weighing balance (Smith model; fish length/weight) |
| Control Parameters | Literature-cited acceptable ranges, not experimentally imposed setpoints: DO >4 mg/L and temperature 26-32C for catfish growth (Joseph et al. 1993; Towers 2014); pH 6-8 for catfish culture (Maarif 2016); TSS 25-80 mg/L acceptable for fish freshwater (Alabaster 1982); BOD5 <=30 mg/L acceptable for aquaculture (Boyd 2003) |
| Combination | African catfish (Clarias gariepinus) and lettuce (Lactuca sativa) in a single-fish-tank, four-trough NFT aquaponic system; amount of biological filter media (microwave pyrolysis biochar: 0/150/300/450 g) is the sole experimental treatment variable; no hydroponic-only control arm |
Site
| Field | Value |
|---|---|
| Region | Southeast Asia |
| Country | Malaysia |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | g/plant (fresh-vs-dry ambiguous, see remarks); cm (height); mg/L (water quality); mg N/plant (nitrogen uptake, Treatment C only) |
| Statistic Details | One-way ANOVA (alpha=0.05); Tukey’s test for pairwise treatment comparisons (Methods 2.6, p.4) |
| Statistically analysed | Y |
| Replicates (n) | 3 (water quality, Table 3 footnote); 5 (plant growth, Table 4 footnote); single physical trough/filter column per treatment — no independent replication of the treatment unit itself, see Extraction notes/T1 remarks |
| AP | 2.05 +/- 0.13 |
Experimental Remarks: TRIAL DEFINITION: T4 = Treatment C (450 g of microwave pyrolysis biochar in the biological filter column) — the highest-biochar-dose arm, reported by the authors as giving the best overall water-quality and growth performance. Same shared fish tank/system as T1-T3 — see T1’s Experimental Remarks for the full paper-wide judgment calls and WARN flags (Fish weight gain BLOCK; Plant fresh/dry weight CHECK), which apply identically here. Paired reference = T1 (Control). No hydroponic control in this paper (HYD-side columns = NA). | WARN-CHECK NO3-N, Treatment C specifically (p.7, Section 3.2.5, and Abstract p.1-2). Table 3 (captioned ‘Values are mean from triplicate data (n=3)’) gives Treatment C’s trial-mean nitrate concentration as 19.69+/-0.06 mg/L, and the same figure (‘up to 19.7 mg/L’) is restated in the first sentence of Section 3.2.5’s own running-text paragraph. The VERY NEXT sentence of that same paragraph instead states ‘Treatment C (450 g) produced the highest nitrate concentration (29.7 mg/L/week) compared to other treatments’ — a second, different value for the same treatment in the same paragraph. The Abstract independently repeats this second figure (‘converted the ammonia into nitrate (29.7 mg/L)’). Fig. 5’s text description (‘nitrate concentration ranged from 2.1 to 34.5 mg/L over the experiment period’) gives a still-different, unattributed-to-treatment range that is consistent with 29.7 being closer to a late-experiment/final-week reading than the whole-trial mean — but per SCHEMA.md’s rule against ever reading a value off a figure, this was not used to resolve anything, only noted as circumstantial support. No sentence in the paper explicitly labels either 19.69/19.7 or 29.7 as ‘the trial mean’ vs. ‘the final/peak value’ — the two numbers most plausibly measure different things (whole-trial mean vs. end-of-experiment or peak weekly value) that the paper conflates without saying so. Per SCHEMA.md’s explicit instruction to take the trial mean +/- SD for water quality, and given Table 3 is the only one of the two sources that carries dispersion and an explicit ‘mean’ caption, the NO3-N cell for this row uses Table 3’s 19.69+/-0.06 mg/L; the repeated ‘29.7 mg/L’ figure (Results text + Abstract, appearing twice) is preserved here rather than in the cell. Affects: NO3-N cell (T4 only); the Abstract’s headline nitrate figure should not be read as this row’s trial-mean NO3-N. Added to REVIEW.md by the batch merge step. | NO COLUMN (this trial’s own values): Table 2 biofilm formation: 62.7 g dry biofilm mass over 70 days, the highest of the four treatments. Table 3 BOD5: 3.94+/-0.48 mg/L, the highest of the four treatments (Section 3.2.2 attributes this to more bacterial growth/oxygen demand on the larger biochar surface area, not water-quality failure — still within the 30 mg/L acceptable range per Boyd 2003). Table 3 TSS: 59.40+/-21.96 mg/L, the only treatment meeting the 25-80 mg/L acceptable freshwater range (Section 3.2.3); TSS removal 68.35%; Ammonia removal 66.66% (significantly different from Control, Section 3.2.4). Table 4: Initial biomass per plant 0.04 g (vs. 0.05 g for the other three — stated as-is, not treated as an error); Weight gain per plant 2.01 g (=2.05-0.04, consistent); RGR per plant 0.0562 %/day, the highest of the four treatments (also the Abstract’s headline growth-rate figure). Total lettuce harvested 21.5 g — see WARN-MINOR in T1’s remarks for the ~5% mismatch against 10x Final biomass per plant. Nitrogen uptake by Treatment C lettuce specifically reported as 110 mg N/plant (Section 3.3.1, ‘refer to S5, Supplementary Material’ for the other three treatments — not available in this PDF, not extracted for Control/A/B); recorded in plant.csv (Category mineral) for this trial only.
Plant Measurements
| Trial | System | Category | Analyte | Value | Unit | Sig. | Location |
|---|---|---|---|---|---|---|---|
| suSimultaneousRemovalToxic2020-T4 | AP | mineral | Nitrogen (N) uptake | 110 | mg N/plant | NR | Results 3.3.1, p.7 |