Comparative assessment of intensive tomato production in innovative non-circulating aquaponics vs. conventional hydroponics
Metadata
- Cite key: madusankaComparativeAssessmentIntensive2023
- Item type: Journal Article
- Authors: C. Madusanka, D.J. Jayasanka, R.N. Nugara, C. Hewawasam
- Affiliation: Department of Biosystems Technology, Faculty of Technology, University of Sri Jayewardenepura, Pitipana, Homagama, Sri Lanka (Madusanka, Jayasanka, Nugara); Department of Civil and Environmental Technology, same faculty (Hewawasam)
- Journal: Agricultural Engineering International: CIGR Journal 25(3) (2023) 42-55
- Date: 09/2023
- Date added: [not reported]
- DOI: [not reported — no DOI found]
- Funding: Serene International (Pvt) LTD, Sri Lanka (Acknowledgements, p.13)
- URL: https://www.researchgate.net/publication/375338001
- PDF:
Madusanka et al. - 2023 - Comparative assessment of intensive tomato product.pdf
Opinion
A genuinely useful design contribution — a “non-circulating” DWC aquaponic bucket system that avoids adding soluble fertilizer directly to fish water — evaluated with a proper CRD (5 reps) and ANOVA/Tukey. The core yield and FUE numbers are internally consistent and cross-check cleanly (FUE ratios recompute to within rounding). But the paper is loose in a few places that matter: it gives two different, irreconcilable durations for the same sowing-to-harvest interval (86 vs 118 days), and its own Discussion asserts a marketable-yield percentage that doesn’t reconcile with its own Results-section weights. Worth citing for the system design and FUE/RWUE framework, but pull the raw Results-table numbers rather than the Discussion’s secondary percentages.
Abstract
New tendencies in farming techniques which include a composite agricultural production system have evolved as solutions for uninterrupted food supply. Production of high-yielding good-quality tomato (Solanum lycopersicum L.) is one of the leading challenges. This study aimed at evaluating the growth, yield, and fruit quality of hybrid tomato (Umagna), cultivated in non-circulating aquaponics and conventional hydroponics systems. A unique and innovative non-recirculating deep water culture aquaponics system (DWCAS) was developed as a prerequisite for high productivity comparable to current stand-alone fish/plant facilities. Including DWCAS, two other conventional hydroponics systems were compared during the study; the deep water culture hydroponics system (DWCHS) and the open bag system (OBS). The assessment of the production systems was based on the growth behavior, tomato yield, and quality. The maximum yield was observed for the DWCHS (2.81 kg/plant) followed by DWCAS (2.4 kg/plant). The least yield was observed for the OBS (2.34 kg/plant). The results demonstrated the highest average fruit weight (169.44 g/fruit) and marketable yield (2802 g) produced by DWCHS. There was no difference in plant dry matter content among production systems. The fertilizer use efficiency was increased by 11.7% and 85.86% in favor of the DWCAS and DWCHS, respectively. The total rainwater use efficiency was also increased in DWCHS.
Summary
The authors compared three tomato-growing systems in a greenhouse in Welimada, Sri Lanka, over one crop cycle: a conventional coco-peat open bag system (OBS), a deep water culture hydroponic system (DWCHS), and a newly designed non-recirculating deep water culture aquaponic system (DWCAS) stocked with Nile tilapia. All three used the same hybrid tomato cultivar, five replicates each in a completely randomized design, and matched EC/pH management. DWCHS produced the highest total and marketable yield per plant and the largest fruit, while DWCAS produced yields statistically indistinguishable from OBS despite drawing on fish effluent instead of a full mineral nutrient solution. Fertilizer use efficiency (FUE) was higher in both water-culture systems than in OBS, and highest of all in DWCAS, meaning DWCAS produced more tomato per kilogram of mineral fertilizer applied even though its total yield was not the highest. The paper also reports a theoretical rainwater-use extrapolation suggesting that scaling the DWCAS fish tank to 1 m3 could jointly yield several kilograms of tilapia and over 200 kg of tomato fruit, though this figure is explicitly presented as a theoretical projection rather than a directly measured result.
Experiment data
- Location: Serene International (Pvt) LTD greenhouse, Welimada, Sri Lanka (up-country intermediate agro-ecological zone IU3)
- Design: Completely randomized design (CRD), 3 treatments (OBS, DWCHS, DWCAS) x 5 replicates
- Replicates / n: 5 per treatment
- Duration: ⚠️ conflicting — see Extraction notes (86 d after sowing per Methods 2.4, vs 118 d from fish stocking to last harvest per Results 3.3)
- Organisms: Nile tilapia (Oreochromis niloticus) / Tomato (Solanum lycopersicum), cultivar Umagna-F1
- Statistics: ANOVA, Tukey mean-separation test, p<=0.05, Minitab 2019
- Fertilizer Use Efficiency (FUE): DWCAS 38.96 kg/kg vs DWCHS 34.85 kg/kg vs OBS 18.75 kg/kg
- Feed Conversion Rate (FCR): 1.11 (DWCAS)
- Total tomato yield per plant: DWCAS 2400 g vs DWCHS 2817 g vs OBS 2344 g (DWCHS significantly highest, p<0.05)
Yield and growth
This paper: DWCAS (the aquaponic treatment) reached a total yield of 2400 g/plant and a marketable yield of 2239 g/plant, statistically indistinguishable from OBS (2344 / 2250 g/plant) but significantly lower than DWCHS (2817 / 2802 g/plant, p<0.05 and p<0.001 respectively) (Section 3.1, p.7). Average fruit weight was highest in DWCHS (169.44 g/fruit) and lowest in OBS (139.96 g/fruit); DWCAS fell in between (Figure 5). DWCAS had the shortest time to flowering (40 d, p<0.001) but the narrowest stem girth (38.4 mm, significantly less than OBS/DWCHS).
Compared with:
- todo Suhl et al. 2016 — reported marketable-to-total tomato ratios of ~99% in both hydroponics and aquaponics (this paper’s own equivalent ratios, computed from its Results-section weights, are ~93-96%, not the ~30% figure the Discussion quotes when citing Suhl — see Extraction notes)
- todo Wortman 2015 — found marketable yields of several species were reduced in recirculated aquaponics vs hydroponics despite similar vegetative growth rates; cited as the closest analogue to this paper’s DWCAS vs DWCHS gap
- todo Castro et al. 2006 — fish-effluent irrigation increased cherry tomato fruit number/productivity in early harvests despite lower mean fruit weight; cited as a partial parallel
- todo Monsees et al. 2017 — reference FCR of 1.2-1.3 at 40 kg fish/m3 in aquaponics; this paper’s FCR of 1.11 (at a far lower ~7.76 kg/m3 stocking density) is cited as “in agreement”
Fertilizer and water use efficiency
This paper: FUE (kg tomato produced per kg mineral fertilizer applied) was 38.96 (DWCAS), 34.85 (DWCHS), 18.75 (OBS) (Table 3, p.9). Mineral fertilizer addition itself was lowest in DWCAS (0.308 kg) vs DWCHS (0.404 kg) vs OBS (0.625 kg). Total rainwater consumption (plant loop only) was equal in DWCHS and DWCAS (0.0455 m3 each) and highest in OBS (0.0636 m3); the DWCAS fish tank consumed a further 0.065 m3 not counted in that figure. RWUE was computed as 184.2 kg/m3 (OBS) and 309.45 kg/m3 (DWCHS); for DWCAS the paper instead reports a theoretical extrapolation — 6.46 kg tilapia + 263 kg tomato per m3 of rainwater applied to a scaled-up fish tank — explicitly distinct from a measured trial RWUE (Section 3.3, p.9).
Compared with:
- todo Suhl et al. 2016 — reported FUE improved 23.6% in aquaponics vs hydroponics; this paper reports an 11.7% FUE improvement of DWCAS over DWCHS, described in the Discussion as “confirming” Suhl despite the different magnitude
Extraction notes
⚠️BLOCK Fish/plant trial duration. Methods, Section 2.4 (p.6): “Tomatoes were harvested 86 days after the sowing date.” Results, Section 3.3 (p.9): “From initial fish stocking until the last day of tomato harvest (118 days) the total rainwater consumption of 5 plants under OBS was 0.0636 m3.” Section 2.2 (p.5) states fish were stocked “on the same date as the sowing of tomato seeds” — so both sentences describe the identical start-to-harvest interval (sowing = fish stocking), yet give 86 d and 118 d, a 32-day gap. No reconciling statement anywhere in the paper (no mention of two different harvest dates, or of “86” being a sub-interval of “118”). No basis to prefer either value. Recorded UNCLEAR in Fish trial duration (days). Also affects: “Days Plant after transplant” (already NR for the independent reason that no sowing-to-transplant day count is stated, so this column would be NR regardless of which duration is correct, but the underlying total-cycle length used anywhere downstream should be treated as unresolved).
WARN-MINOR (no cell impact) — DWCHS total yield rounding. Abstract (p.42): “DWCHS (2.81 kg/plant).” Results, Section 3.1 (p.7): “The total weight of tomato fruits harvested per plant in DWCHS was 2817 g” (=2.817 kg, which rounds to 2.82, not 2.81). Recorded the more precise 2817 g figure in Plant fresh weight; does not change any cell since 2817 g is used directly.
WARN-MINOR (no cell impact) — Fertilizer-reduction percentage. Table 3 footnote and Discussion (p.9-10) both state fertilizer addition into DWCAS “was reduced by 23.0%” compared to DWCHS. Recomputing from the paper’s own stated kg values (0.404 kg DWCHS, 0.308 kg DWCAS): (0.404-0.308)/0.404 = 23.76%, not 23.0%. The other two reported reduction percentages in the same table (35.36% DWCHS-vs-OBS, 50.72% DWCAS-vs-OBS) both recompute exactly from the stated kg values. Only the DWCAS-vs-DWCHS figure is off by ~0.8 percentage points. Does not affect any cell — Mineral fertilizer addition is recorded as the stated kg values (0.308 / 0.404 / 0.625), not as a derived percentage.
WARN-MINOR (no cell impact) — Marketable-yield-contribution percentage does not reconcile with the paper’s own Results weights. Discussion (p.10): “our study demonstrated a nearly similar contribution toward marketable yield obtained from DWCAS (30.70%) and OBS (30.86%),” offered as an analogue to Suhl et al. (2016)‘s ~99% marketable-to-total ratios. But this paper’s own Results-section weights (Section 3.1, p.7) give a marketable/total ratio of 2239/2400 = 93.3% for DWCAS and 2250/2344 = 96.0% for OBS — nowhere close to 30.70%/30.86%, and the 30% figures do not reconcile with any combination of the paper’s stated absolute weights. Does not affect any extracted cell: Plant fresh weight and the marketable-yield figure quoted above are taken directly from the Results-section gram values, not from this Discussion percentage. Flagged because a reader citing this paper’s “agreement with Suhl” on marketable-yield ratio would be citing an apparently erroneous figure.
WARN-MINOR (no cell impact, low confidence) — Figure 7 pie-chart percentages vs. equal stated volumes. Section 3.3 (p.9) states DWCHS and DWCAS rainwater consumption were both “0.0455 m3” (identical). Figure 7’s pie chart, however, appears to assign them different shares (~30% DWCHS vs ~29% DWCAS) rather than an equal split — reading of a low-resolution chart, so confidence in this specific observation is limited. Does not affect any cell: Water/rainwater figures are recorded from the explicit 0.0455 m3 text statement, not the chart percentages.
WARN-MINOR (no cell impact, low confidence) — possible fruit-width Tukey-letter/value mismatch. Results (p.7-8) states “The highest average fruit width was observed in the OBS (118.4 mm)… DWCHS and DWCAS produced fruits with an average width of 118.0 mm and 109.0 mm respectively” — i.e., OBS is stated as numerically highest. Figure 6’s bar-chart letter groups, as best read from the rendered image, appear to place OBS in a lower significance group alongside DWCAS, with DWCHS alone in the higher group — the opposite of what the stated numeric ranking would suggest under this paper’s own letter convention (highest = “a”, used consistently in Tables 1/2/4). Given the difficulty of reading exact bar-to-letter correspondence from the image, this is flagged with low confidence rather than asserted. No trials.csv column exists for fruit width, so this does not affect any cell; included for the reader’s awareness given the paper’s fruit-quality claims.
Fish biomass created — not recorded, ambiguous basis. Results, Section 3.3 (p.9): “the DWCAS achieved a total fish yield of only 0.42 kg.” This could mean final fish biomass at harvest or biomass gained during the trial (the column’s stated definition is “final − initial biomass”); the paper does not specify which. Rather than guess, left NR in Fish biomass created, with the stated 0.42 kg figure preserved here.
Authors — Zotero export omits an author present in the paper’s own byline. zotero-export.csv lists only 3 authors (Madusanka, Chathura; Jayasanka, Dikkumburage Jasintha; Hewawasam, Choolaka) for this record. The PDF’s own title page (p.42) and the paper’s own suggested citation (p.42: “Madusanka, C., Jayasanka, D. J., Nugara, R. N., and Hewawasam, C. 2023…”) both list four authors, including Ruwani N. Nugara. Used the PDF/paper’s own 4-author byline rather than the Zotero export’s 3-author list, since the paper’s own self-citation is the more authoritative source for its own authorship and the omission looks like a Zotero data-entry gap rather than a deliberate correction.
DOI. Not present in zotero-export.csv (empty DOI field for this record) and not printed anywhere in the PDF. A CrossRef bibliographic search for the title returned no matching record (closest hits were the unrelated Suhl et al. 2016 and other aquaponics papers). Recorded as no DOI found; this CIGR Journal (Agricultural Engineering International, open-access at cigrjournal.org) does not appear to assign DOIs to its articles from this period. URL field uses the ResearchGate mirror instead.
[not reported] fields (grouped): SGR; N, P, K (feed composition, beyond the stated 30% protein); Fish size final; Fish weight gain; Fish survival rate; Total Feed (kg); Water recycle flow rate (treated as NA — system is explicitly non-recirculating); Daily Water exchange rate; Dissolved Oxygen; Water temperature; TAN/NH4-N, NO2-N, NO3-N (no water nutrient analysis reported); Tissue nitrate AP/HYD; SPAD; Plants/m2 (spacing given as 45 cm between plants, not as a density); Days Plant after transplant (no sowing-to-transplant interval stated); pHOptimal; Average room Temperature (only an annual regional climate average of 21.1 degC is cited, p.4, not a measured trial-period greenhouse mean, despite daily greenhouse temperature having been measured per Section 2.3); Iron supplemented; Remineralization.
[unclear] fields: Aq pH and EC are both given only as controlled target ranges/limits (pH 5.5-6.5 via daily HCl/NaOH adjustment; EC maintained below 4.0 mS/cm), not as measured trial means — recorded as range/limit with this caveat rather than as a trial mean.
NO COLUMN items (Experimental Remarks in trials.csv): Marketable fruit count (OBS=14, DWCHS=16, DWCAS=14, ns); vegetative-tissue biomass broken down by organ (Table 4, fresh/dry weight of leaves/stem/root separately — note this is distinct from the fruit-yield figures used for Plant fresh weight/dry matter in this note); days to flowering (OBS 48a, DWCHS 43b, DWCAS 40c, p<0.001); stem girth (OBS 41.0a, DWCHS 40.8a, DWCAS 38.4b mm, p<0.05); fruit length/width/diameter (Figure 6, mm); fruit firmness (kg f-1, ns); fruit juice pH and EC (Table 2 — these are fruit-tissue physicochemical properties, not water chemistry, and do not fit any of the four plant_measurements.csv analyte categories, so excluded from plant.csv as well as trials.csv’s water columns); economic figures (gross/net return, benefit-cost ratio, Rs, p.12-13).
Water panel note: there is no separate water-quality analytical panel (no TAN/NO2/NO3/DO/temperature dataset) to consider for exclusion — the paper simply does not report one beyond the pH/EC setpoints already noted above.
Type classification: experiment — CRD with 5 replicates, three defined treatment arms, ANOVA + Tukey at p<=0.05 throughout (Section 2.9, p.7). Clearly meets the experiment bar, not quasi-experiment or exploratory.
Trial count / definition: Only DWCAS is an aquaponic treatment; OBS and DWCHS are both non-aquaponic controls (fertigated substrate culture and pure hydroponics respectively). Per schema, one row per aquaponic treatment → one trial row (T1 = DWCAS). DWCHS was chosen as the paired HYD control because it is the physically identical bucket/DWC system differing only in nutrient source (fish effluent + mineral top-up vs. mineral solution only) — the cleanest isolation of the “aquaponics effect.” OBS values are recorded in Experimental Remarks as a secondary, structurally different comparison arm (different substrate/media entirely), not as the HYD column.
Linked claims
- Aquaponic tomato yield can approach hydroponic yield while reducing mineral fertilizer use
- Fertilizer use efficiency can be higher in aquaponics than hydroponics even when total yield is lower
- Non-recirculating aquaponic designs can avoid direct fertilizer contact with fish water
Citations to chase
- todo Suhl et al. 2016 — Advanced aquaponics: intensive tomato production in aquaponics vs. conventional hydroponics (already in vault, see
notes/— check for existing note before re-adding) - todo Wortman 2015 — Crop physiological response to nutrient solution EC and pH in ebb-and-flow hydroponics (vegetative growth rate parity, marketable yield reduction in aquaponics)
- todo Castro et al. 2006 — Increasing cherry tomato yield using fish effluent as irrigation water
- todo Monsees et al. 2017 — Decoupled systems on trial: eliminating bottlenecks to improve aquaponics processes (reference FCR 1.2-1.3 at 40 kg fish/m3)
Source: Madusanka et al. - 2023 - Comparative assessment of intensive tomato product.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
madusankaComparativeAssessmentIntensive2023-T1
Fish
| Field | Value |
|---|---|
| Fish | Nile tilapia (Oreochromis niloticus) |
| Initial Stock density | 7.76 |
| FCR | 1.11 |
| Protein | 30 |
| % of body weight | 6 |
| Fish size initial | 29.1 |
| Feed routine | Three times per day, one-third at 08:30, one-third at 13:30, remaining at 16:30 |
| Feed regime | Commercial feed manufactured by Skretting Nutreco company (Norway), average protein content 30%, at 6% of body weight per day |
Water
| Field | Value |
|---|---|
| Water volume in the system | 30 (fish tank only) |
| Water type | Rainwater |
| Aq pH | 5.5-6.5 (target range only, no trial mean reported) |
| FUE AP | 38.96 |
| FUE HYD | 34.85 |
| EC | NR (target <4.0 mS/cm, not a measured trial mean) |
Plant
| Field | Value |
|---|---|
| Plant | Tomato (Solanum lycopersicum L.), hybrid cultivar ‘Umagna-F1’ |
| Details | Sown in nursery 23 June 2020; irrigated with Albert’s solution (1.5 mS/cm) from the 12th nursery day; transplanted at 4-6 true-leaf stage; net pots with coco peat placed on bucket lids 6 days after transplanting; 45 cm inter-plant spacing; harvested at fruit maturity (duration UNCLEAR, see remarks) |
| Plant height | 163.40 |
| Leaf count | 17 |
| Plant fresh weight | 2400 |
| Plant dry matter | 67.0 |
System & Setup
| Field | Value |
|---|---|
| System type | Non-circulating deep water culture aquaponics system (DWCAS) |
| Media Details | Coco peat-filled net pots (10 established pre-transfer) on lids of uniform 4 L paint buckets (top dia 0.192 m, bottom dia 0.168 m, height 0.19 m); water level 0.118 m depth = 2.5 L per bucket; air stone per bucket via air pump, 3.5 L/min; fish tank 30 L rainwater, air stone via air pump, 2.5 L/min; bio-filter 25 L cistern with onion bags, gravel, PVC nets, mechanical filter; nitrifying bacteria added at trial initiation |
| Biological system already in use | N (Nitrifying bacteria added to the biofilter at trial initiation (p.5); no pre-cycling period stated, implying the biological system was not already established before this trial began) |
| Air supplement | Y (Aquarium air stone in fish tank via air pump, airflow rate 2.5 L/min (p.5); air stone in each DWC bucket via air pump, airflow rate 3.5 L/min (shared design with DWCHS, p.5)) |
| pH Buffers | Y (Nutrient solution pH monitored daily and adjusted to 5.5-6.5 using HCl or NaOH, across all treatments including DWCAS (p.6)) |
| Climate control | N (Explicitly non-temperature-controlled polythene-covered greenhouse (Section 2.1, p.4); average annual site temperature 21.1 degC cited from external climate data, not a controlled setpoint) |
| Artificial Lighting | N (Natural lighting used throughout the study period, explicitly stated (p.5)) |
| Nutrient supplemented | Y (DWCAS irrigated with fish effluent PLUS additional mineral fertilizer (10-52-10, 20-20-20, 6-30-30, 13-2-44) during vegetative stage, dosed to match target EC rather than a fixed amount (p.6); Ca supplement increased during fruiting stage across all treatments including DWCAS (p.6)) |
| Equipment | Air pumps (2.5 L/min fish tank; 3.5 L/min DWC buckets); mechanical filter inside bio-filter; bio-filter substrate (25 L cistern, onion bags, gravel, PVC nets); hand-held refractometer (Brix 0-32%, Model MT-032) for TSS; EC/pH meter (Model PCTEST-35) for fruit juice; hand penetrometer (Model FT 327, 200 g-20 kg, 0.7x0.92 mm probe) for firmness; weighing balance (SF-400, 10000 g x 1 g) |
| Control Parameters | EC of nutrient solution maintained below 4.0 mS/cm via mineral fertilizer dosing (p.6); pH maintained 5.5-6.5 via HCl/NaOH (p.6); nutrient solution volume held at 2.5 L per pot, replenished daily for evapotranspiration losses (p.6); fish fed 6% body weight/day, split 3x daily (p.5) |
| Combination | Nile tilapia (Oreochromis niloticus) and hybrid tomato (Solanum lycopersicum cv. Umagna-F1); non-circulating DWC aquaponics (DWCAS) compared against DWC hydroponics (DWCHS, paired HYD control here) and a structurally different coco-peat open bag system (OBS, secondary control, values in remarks) |
Site
| Field | Value |
|---|---|
| Region | South Asia |
| Country | Sri Lanka |
| Lat | 6.90 |
| Long | 80.90 |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | g/plant (fruit yield); cm (plant height); count (leaf number); g (vegetative fresh/dry weight, Table 4) |
| Statistic Details | Analysis of Variance (ANOVA); Tukey mean-separation test, p<=0.05; Minitab version 2019; CRD, 5 replicates |
| Statistically analysed | Y |
| Replicates (n) | 5 |
| AP | 2400 |
| HYD | 2817 |
Experimental Remarks: TRIAL DEFINITION: T1 = DWCAS (non-recirculating deep water culture aquaponics, Nile tilapia + tomato cv. Umagna-F1), the only aquaponic treatment in this paper. Paired HYD control = DWCHS (identical DWC bucket system, mineral nutrient solution only, no fish) - chosen because it isolates the aquaponics effect (same physical system, differs only in nutrient source). A third, structurally different treatment (OBS, coco-peat open bag system, fertigated) was also run as a conventional baseline; its values are not placed in the HYD columns (different substrate/media entirely) but recorded here: total yield 2344 g/plant, marketable yield 2250 g/plant, average fruit weight 139.96 g/fruit, plant height 150.40 cm (b), leaf count 18 (ns), FUE 18.75 kg/kg, mineral fertilizer 0.625 kg, rainwater consumption 0.0636 m3, TSS 3.85%, fruit juice pH 5.02a, EC 10.33b mS/cm, firmness 1.88 kg f-1, days to flowering 48a, stem girth 41.000a mm, Table 4 vegetative fresh weight 365.0b g / dry weight 56.8 g. | WARN-BLOCK Fish/plant trial duration: Methods 2.4 (p.6) ‘Tomatoes were harvested 86 days after the sowing date.’ Results 3.3 (p.9) ‘From initial fish stocking until the last day of tomato harvest (118 days)…’. Section 2.2 (p.5) states fish were stocked on the same date as sowing, so both statements describe the same interval; 86 vs 118, a 32-day gap, no reconciling statement anywhere in the paper. No basis to prefer either value. Recorded UNCLEAR in Fish trial duration (days). | WARN-CHECK Plant fresh weight vs Plant dry matter scope: Plant fresh weight recorded as total FRUIT yield per plant (AP 2400 g / HYD 2817 g, Section 3.1 p.7 - the paper’s headline yield metric, matches the column’s own note ‘distinct from area yield’). Plant dry matter recorded as total VEGETATIVE (leaves+stem+root) dry weight (AP 67.0 g / HYD 69.4 g, Table 4 p.9 - what the paper’s own Abstract and Section 3.5 explicitly call ‘plant dry matter content’). The paper never reports a fruit dry-matter value, so the two cells necessarily describe different plant fractions; no conversion is possible. If vegetative-tissue FRESH weight is preferred instead of fruit yield for scope consistency with the dry-matter cell, the alternative candidates are AP 413.0 g (ab) / HYD 452.0 g (a), Table 4. | UNIT CONVERSION ONLY: coordinates given as degree symbol + ‘.90’ + a double-prime mark (Section 2.1, p.4; e.g. approximately ‘6 deg .90-sec N, 80 deg .90-sec E’) read as decimal 6.90 N / 80.90 E, consistent with Welimada, Sri Lanka - the paper’s own named site. | NOT DERIVED, left NR: Total Feed in kg (only feed type/ration-rate/times-per-day given, no cumulative kg stated); Fish biomass created (paper states ‘total fish yield of only 0.42 kg’, Section 3.3 p.9, but never specifies whether this is final biomass or biomass gained during the trial - not recorded pending clarification of basis, value preserved here); Fish weight gain and Fish size final (only initial weight 29.1 g given, no final weight or gain stated); SGR; feed N/P/K composition beyond the stated 30% average protein; Days Plant after transplant (no stated sowing-to-transplant day count, and independently affected by the 86-vs-118 duration conflict above); Plants/m2 (only a 45 cm inter-plant spacing is given, no bed/pot area). | Aq pH and EC recorded as controlled target ranges/limits (pH adjusted daily to 5.5-6.5 via HCl/NaOH, Section 2.3 p.6; EC maintained below 4.0 mS/cm via fertilizer dosing, Section 2.3 p.6), not as measured trial means - the paper reports control setpoints for the nutrient solution, not a summary statistic of an actually-measured water parameter. | Water recycle recorded NA: system is explicitly described as non-recirculating/non-circulating by design (title, abstract, Section 2.1, p.4); no continuous water-transfer flow rate between fish tank and plant loop is stated anywhere (only air-pump aeration flow rates, 2.5/3.5 L/min, which are air not water flow). | Water volume in the system recorded as 30 L (fish tank only, Section 2.1 p.4); a separate 25 L bio-filter cistern and 2.5 L per plant bucket (x5 buckets) are also stated but not summed here (summing would be derivation). | Fish Category and Water classification left NR - the paper does not categorize beyond naming the species/water source. | Average room Temperature left NR: only the site’s annual regional climate average (21.1 degC, ‘Climate data, 2020’, p.4) is cited; greenhouse temperature was measured daily with a thermometer (Section 2.3, p.6) but no summary/mean of that measurement is reported anywhere in the paper. | AP/HYD columns duplicate Plant fresh weight (2400 g AP / 2817 g HYD, per plant) since this paper reports no separate area-based (per-m2) yield figure to place there instead. | NO COLUMN: marketable fruit count (OBS=14, DWCHS=16, DWCAS=14, ns, p.7); days to flowering (OBS 48a, DWCHS 43b, DWCAS 40c, p<0.001, Table 1); stem girth (OBS 41.000a, DWCHS 40.800a, DWCAS 38.400b mm, p<0.05, Table 1); fruit length/width/diameter (Figure 6, mm - DWCHS highest length 106.2mm and width 118.0mm, OBS lowest length 97.6mm but highest STATED width 118.4mm - see note file for a WARN-MINOR on a possible letter/value mismatch on fruit width); fruit firmness (OBS 1.88, DWCHS 2.02, DWCAS 1.80 kg f-1, ns); fruit juice TSS/pH/EC (Table 2: TSS ns across all treatments, 3.85/3.70/3.60% OBS/DWCHS/DWCAS - also captured in plant_measurements.csv as biochemistry; pH DWCHS 4.86b significantly lower than OBS 5.02a and DWCAS 5.04a; EC DWCHS 10.84a and DWCAS 10.72a both significantly higher than OBS 10.33b mS/cm - fruit-juice pH and EC excluded from plant.csv, no matching analyte category among biochemistry/mineral/microbiology/proximate); economic figures (gross return, net return, benefit-cost ratio in Rs, Section 4 p.12-13, no currency/economics column in schema). | WARN-MINOR (no cell impact): DWCHS total yield rounds to 2.82 kg from the Results-section 2817 g figure, vs 2.81 kg stated in the Abstract; the DWCAS-vs-DWCHS fertilizer-reduction percentage is stated as 23.0% (Table 3, Discussion p.9-10) but recomputes to 23.76% from the paper’s own kg values (the other two reduction percentages in the same table recompute exactly); the Discussion’s marketable-yield-contribution figures (DWCAS 30.70%, OBS 30.86%, p.10, offered as an analogue to Suhl et al. 2016) do not reconcile with the paper’s own Results-section weights (which imply marketable/total ratios of ~93-96%, not ~30%); Figure 7’s pie-chart percentages (DWCHS ~30%, DWCAS ~29%) appear unequal despite the text explicitly stating both consumed an identical 0.0455 m3 of rainwater. None of these affect any extracted cell. | Authors: the PDF byline and the paper’s own suggested citation (p.42) list 4 authors including Ruwani N. Nugara, who is missing from zotero-export.csv’s 3-author list for this record; used the paper’s own 4-author byline instead. | DOI: not present in zotero-export.csv or anywhere in the PDF; a CrossRef bibliographic search on the title returned no matching record. Recorded as no DOI found.
Plant Measurements
| Trial | System | Category | Analyte | Value | Unit | Sig. | Location |
|---|---|---|---|---|---|---|---|
| madusankaComparativeAssessmentIntensive2023-T1 | AP | biochemistry | Total soluble solids (TSS) | 3.6 | Brix % | ns | Table 2 |
| madusankaComparativeAssessmentIntensive2023-T1 | HYD | biochemistry | Total soluble solids (TSS) | 3.70 | Brix % | ns | Table 2 |
| madusankaComparativeAssessmentIntensive2023-T1 | OBS | biochemistry | Total soluble solids (TSS) | 3.85 | Brix % | ns | Table 2 |