Implementation of an experimental nutrient film technique-type aquaponic system
Metadata
- Cite key: castillocastellanosImplementationExperimentalNutrient2016
- Item type: Journal Article
- Authors: D. Castillo-Castellanos, I. Zavala-Leal, J. M. J. Ruiz-Velazco, A. Radilla-García, J. T. Nieto-Navarro, C. A. Romero-Bañuelos, J. González-Hernández
- Affiliation: Escuela Nacional de Ingeniería Pesquera, Universidad Autónoma de Nayarit (UAN), Carretera a Los Cocos Km. 12, Bahía de Matanchén, San Blas, Nayarit, Mexico
- Journal: Aquaculture International 24(2) (2016) 637–646
- Date: 04/2016
- Date added: 2019-12-20
- DOI: 10.1007/s10499-015-9954-z
- Funding: Programa para el Mejoramiento al Profesorado (PROMEP-SEP) and Consejo Nacional de Ciencia y Tecnología (CONACYT); specifically the Teacher Improvement Program (PROMEP) of the Ministry of Public Education (SEP), project “Propuesta y evaluación de tres sistemas de recirculación agroacuícolas ecoeficientes protegidos para el cultivo de tilapia, camarón y lechuga,” Grant Number SIP12-178
- URL: https://doi.org/10.1007/s10499-015-9954-z
- PDF:
Castillo-Castellanos et al. - 2016 - Implementation of an experimental nutrient film te.pdf
Opinion
A clean, well-instrumented engineering-implementation paper: the core claim (an NFT aquaponic loop can drop the sump entirely with no operational problems over a full 7-week cycle) is well supported and clearly described, with real replication (9 aquaponic vs. 3 hydroponic modules) and proper ANOVA/Tukey testing throughout. The AP-vs-HYD yield comparisons for lettuce and cucumber are usable as-is. Two things temper trust a little: the reported specific-growth-rate value (4.95) does not reconcile with the paper’s own stated formula and inputs when recomputed (see Extraction notes), and Table 3’s “n=3” footnote sits awkwardly against the Methods’ stated 9-vs-3 system design — worth a second look before citing either number precisely. Water chemistry reporting is thin (only temperature/DO/pH/EC/ammonia as ranges, no nitrite or nitrate at all, which is a real gap for an aquaponics paper), and no tissue-quality or biochemical analysis was done on the plants.
Abstract
Aquaponics has been considered as having great potential as an organic production method of aquatic organisms and vegetables, because the nutrient-rich water from aquatic organisms is utilized for plant growth. The essential elements of an aquaponic system consist of the following five: a tank to maintain aquatic organisms; a clarifier or sedimentation; a biofilter; a hydroponic component; and a sump pump. In this paper, we propose the design and implementation of a nutrient film technique-type aquaponic system, which does not include the sump pump. Aquaponic systems were tested during one production cycle of the Carolina cucumber (Cucumis sativus) and Parris Island lettuce (Lactuca sativa). The aquatic organism employed was tilapia (Oreochromis niloticus). Nine systems were utilized with the aquaponic technique, and the remaining three were utilized with only the hydroponic technique as controls in plant production. The proposed aquaponic system worked efficiently during the entire crop cycle without any problems or deficiencies from lack sump. No changes were made in the flow of water in the aquaponic system during the growing season. Tilapia survival was 97.2 ± 2.4 %. Specific growth and food conversion ratios were 4.95 and 0.99, respectively. In plant survival, there were no significant differences (p>0.05) between both production techniques. In lettuce, foliar wet weight, foliar dry weight, and yield were higher (p<0.05) in hydroponics. In cucumber, fruit number and yield were higher (p<0.05) in the hydroponic system. Fruit length, width, and weight exhibited no differences (p>0.05).
Summary
The authors designed, built, and ran 12 identical greenhouse recirculating modules to test whether a nutrient film technique (NFT) aquaponic system could be run without the conventional fifth component — a sump pump used to return water to the fish tank — while still functioning normally. Nine modules ran as aquaponic systems (tilapia + NFT-grown lettuce and cucumber sharing the same water loop) and three ran as hydroponic controls (same physical design, but fed Steiner nutrient solution instead of fish effluent, with an empty clarifier/biofilter). Over a single 7-week cycle, the no-sump aquaponic design ran without water-flow problems, tilapia survived at 97.2% with an FCR of 0.99, and lettuce/cucumber plant survival did not differ between techniques. However, lettuce grew substantially better in hydroponics (higher wet/dry weight, height, and yield), and cucumber yield and fruit number were also higher in hydroponics, while cucumber fruit dimensions (length, width, weight) did not differ — the authors attribute this to a fixed >=12 cm harvest cutoff masking any size difference, and to comparatively low fish biomass loading limiting the nutrient supply reaching the aquaponic plants. The paper’s main contribution is demonstrating the no-sump NFT design works structurally, not that aquaponic yields matched hydroponic ones.
Experiment data
- Location: Greenhouse at Escuela Nacional de Ingeniería Pesquera, Universidad Autónoma de Nayarit, San Blas, Nayarit, Mexico (130 m² greenhouse, 8.10 × 16 m).
- Design: 12 identical NFT-type recirculating modules; 9 run as aquaponic systems (tilapia + lettuce + cucumber sharing one water loop, no sump), 3 run as hydroponic controls (same design, Steiner nutrient solution, empty clarifier/biofilter). Systems randomized in the greenhouse. One-way ANOVA with Tukey’s a posteriori test, p<0.05.
- Replicates / n: Methods state 9 aquaponic vs. 3 hydroponic modules (p.5); Table 3’s own statistical footnote states “n=3” for both arms — see Extraction notes, ⚠️MATERIAL.
- Duration: 7 weeks (49 days) for both fish and plant components (p.5).
- Organisms: Tilapia (Oreochromis niloticus) var. Spring / Lettuce (Lactuca sativa), Parris Island (romaine) / Cucumber (Cucumis sativus), Carolina.
- Statistics: Shapiro–Wilk (normality), Cochran’s outlying variance C test (homogeneity of variance), one-way ANOVA, Tukey’s a posteriori test, p<0.05; survival data arcsine-transformed.
- Feed Conversion Rate (FCR): 0.99 (tilapia, whole trial).
- Specific Growth Rate (SGR): 4.95 (Table 2, labelled “g day⁻¹”) — ⚠️MATERIAL, see Extraction notes.
- Tilapia survival: 97.2 ± 2.4 % over 49 days.
- Lettuce yield: 47.9 g m⁻² (AP) vs. 726.8 g m⁻² (HYD), p<0.05.
- Cucumber yield: 78.7 g m⁻² (AP) vs. 421.6 g m⁻² (HYD), p<0.05.
System design (no-sump NFT)
This paper: The proposed design uses only 4 of the 5 conventional aquaponic elements — fish tank (1 m³, HMW/HD PET), clarifier/sedimentation (119.5 L), biofilter (105 L, corrugated plastic hose substrate), and NFT hydroponic channels (six 4” PVC channels/module, 3 m long, 10 plants/channel, 60 plants/module) — omitting the sump reservoir entirely, with water pumped directly from the biofilter to the channels (Quiet One model 3000, 0.12 L/s) and returned to the fish tank by gravity. The system ran the full 49-day cycle “without any problems or deficiencies from lack of sump,” with no changes to water flow (p.6–7).
Compared with: todo Rakocy et al. 2007 — standard aquaponic system description including sump reservoir, contrasted against this no-sump design (p.7).
Fish (tilapia) performance
This paper: Stocked at 90 fish m⁻³ (≈7 kg m⁻³, 78.1 ± 1.9 g fish⁻¹ mean initial weight) across the 9 aquaponic tanks (810 fish total, 63.7 kg total initial biomass). Harvested at 242.9 ± 8.9 g fish⁻¹ (191.2 kg total), 97.2% survival, FCR 0.99, SG 4.95 (Table 2), over 49 days. Fed Nutripec (Purina) 3.5 mm pellets, 38% protein / 8% lipids, at 3.5% of total biomass/day, 4×/day.
Compared with:
- todo Rakocy et al. 2004 — similar tilapia survival (>95%) under comparable handling; SGR “similar”; FCR 1.7 vs. this paper’s 0.99, attributed to feeding management (p.8).
- todo Al-Hafedh et al. 2008 — tilapia at 160 fish m⁻³, initial weights 42.5–248 g, FCR 1.4, SGR 1.5 g day⁻¹, in floating raft aquaponics, Saudi Arabia (p.8).
- todo Danaher et al. 2011 — tilapia survival >95% in raft aquaponics (p.8).
- todo Danaher et al. 2013 — tilapia survival reported below 85% in some raft aquaponic systems (p.8).
Lettuce yield and growth
This paper: Aquaponic lettuce: survival 54.6%, wet weight 18.8 g plant⁻¹, dry weight 1.5 g plant⁻¹, height 18.7 cm, yield 47.9 g m⁻². Hydroponic lettuce: survival 59.2% (ns vs. AP), wet weight 123.2 g plant⁻¹, dry weight 9.1 g plant⁻¹, height 42.2 cm, yield 726.8 g m⁻² — all significantly higher than aquaponic (p<0.05, Table 3). Attributed to insufficient nitrogen supply from the comparatively low fish biomass loading (~7 kg m⁻³) relative to plant demand.
Compared with:
- todo Quillere et al. 1995 — lettuce weight 166 ± 58 g plant⁻¹ in an artificial fish/bacteria/plant ecosystem, higher than either arm of this study (p.8).
- todo Al-Hafedh et al. 2008 — leaf lettuce weights of 157/212/289 g plant⁻¹ and yields up to 6060 g m⁻² in floating aquaponic systems at higher fish stocking density (160 fish m⁻³) (p.8).
- todo Bugarín-Montoya et al. 2011 — leafy vegetables such as lettuce show higher, steadier nitrogen absorption through their cycle than fruiting plants, offered as an explanation for lettuce showing the largest AP-vs-HYD gap in this study (p.9).
Cucumber yield and fruit quality
This paper: Aquaponic cucumber: survival 100% (ns vs. HYD), fruit length 11.8 cm, width 4.1 cm, weight 127.6 g, fruit number 1.76 m⁻², yield 78.7 g m⁻². Hydroponic cucumber: survival 100%, fruit length 12.8 cm, width 4.2 cm, weight 150.4 g (all ns vs. AP), fruit number 8.5 m⁻² and yield 421.6 g m⁻² (both p<0.05 vs. AP). Fruit dimensions did not differ because only fruit reaching the ≥12 cm commercial-size cutoff were harvested/measured in either arm; fruit number and yield did differ, attributed to the aquaponic system’s lower nutrient loading relative to the higher demand of fruiting crops.
Compared with:
- todo Ulloa et al. 2005 — 112.5 g fruit plant⁻¹ in an experimental aquaponic system, vs. 33.7 g fruit plant⁻¹ (derived, not this paper’s stated unit) in this study; difference attributed to a much higher fish biomass in Ulloa et al. (54 kg m⁻³, 300 fish × 90 g in a 500 L tank) vs. this study’s 7 kg m⁻³ (p.8–9).
- todo Diver 2006 — fruiting crops (tomato, pepper, cucumber) have higher nutrient demand than leafy crops, cited as context for the larger AP-vs-HYD cucumber yield gap (p.9).
Water quality
This paper: Reported only as ranges over the 7-week aquaponic culture period (Table 1, no trial mean given): temperature 26.6–31.6 °C, dissolved oxygen 3.1–6.1 mg L⁻¹, pH 7.5–8.5, electrical conductivity 0.5–0.9 mS cm⁻¹, ammonia 0.5–4.8 mg L⁻¹. No nitrite or nitrate monitoring is reported anywhere in the paper. Water recirculation was ~800% of tank volume per day; ~30 L of make-up water was added every third day to offset evaporation and plant uptake.
Linked claims
- Removing the sump does not impair aquaponic NFT system function
- Aquaponic systems underperform hydroponics on lettuce yield under low fish stocking density
- Fruiting crops have higher nutrient demand than leafy crops in aquaponics
Citations to chase
- todo Rakocy et al. (2007) — standard 5-element aquaponic system description including sump, contrasted with this paper’s no-sump design.
- todo Rakocy et al. (2004) — tilapia FCR (1.7) and SGR benchmark for UVI aquaponic system.
- todo Al-Hafedh et al. (2008) — floating-raft aquaponic tilapia/lettuce performance at higher stocking density, Saudi Arabia.
- todo Danaher et al. (2011) — tilapia survival benchmark in raft aquaponics.
- todo Danaher et al. (2013) — lower tilapia survival (<85%) reported in some raft aquaponic systems.
- todo Quillere et al. (1995) — lettuce weight benchmark in an artificial fish/bacteria/plant ecosystem.
- todo Bugarín-Montoya et al. (2011) — nitrogen absorption patterns of leafy vs. fruiting vegetables.
- todo Ulloa et al. (2005) — experimental aquaponic cucumber fruit weight and fish biomass loading benchmark.
- todo Diver (2006) — nutrient demand of fruiting vs. leafy aquaponic crops.
Extraction notes
⚠️MATERIAL — Replicates (n). Materials and methods (p.5): “Nine of the 12 systems were utilized with the aquaponic technique, while the remaining three … employed only the hydroponic technique,” i.e. 9 AP vs. 3 HYD experimental units. Table 3’s own footnote (p.7), attached directly to the ANOVA/Tukey significance letters in that table, instead states “(p<0.05; n=3)” for both columns, with no stated sub-blocking of the 9 AP systems into groups of 3. No basis in the text to fully reconcile the two statements. Recorded Replicates (n) = 3 in both trial rows, since it is the value directly tied to the reported statistical test in the very table these rows summarize; the Methods-stated 9 AP / 3 HYD system count is preserved in the remarks for the record. Affects confidence in the effective aquaponic-side replication behind every significance letter in Table 3.
⚠️MATERIAL — SGR (Specific growth). Table 2 / Abstract / Results all consistently give “Specific growth (g day⁻¹)” = 4.95 (three independent restatements, no competing value in the paper). Recomputing the paper’s own stated formula (Methods, p.6: SG = (ln wₜ − ln w₀)/t) with its own stated inputs (w₀ = 78.1 g, wₜ = 242.9 g, t = 49 days, all from Table 2) gives ≈0.0231 (2.3%/day) — not 4.95. The Table 2 unit label “g day⁻¹” is also dimensionally inconsistent with a log-ratio specific growth rate, which is inherently a proportional (%/day) quantity rather than a mass/day quantity. Since there is no second value in the paper to weigh against 4.95, this is a recompute-fails-to-check case rather than a two-source contradiction; the printed value (4.95) was recorded in the SGR cell as the only value the authors state, with the arithmetic mismatch flagged here for anyone comparing this figure against other tilapia growth-rate studies.
⚠️CHECK — Days Plant after transplant. The only stated duration for the plant trial is the whole-experiment “7 weeks (49 days) for both treatments” (p.5). Fig. 1 labels a “Seedling production room” as part of the facility, implying seedlings were raised elsewhere and then moved into the channels, but Methods (p.6) instead says lettuce and cucumber “were seeded in hydroponic module in both techniques” — i.e. direct seeding, not transplanting. No explicit transplant date or seedling age is given anywhere. Recorded 49 days (the only figure available) in both trial rows, flagged UNCLEAR as to whether it counts from seed, an unstated transplant date, or is simply the fish-trial duration applied to the plants by extension.
⚠️CHECK — Daily Water exchange rate. Two different water-turnover figures are given with no explicit mapping onto a single “exchange rate” concept: “water recirculation in each tank was about 800% daily” (p.5, a closed-loop turnover rate) vs. “~30 L every third day” supplied to replace evaporation/plant-uptake losses (p.5, a net make-up-water figure). Recorded 800%/day as the closer textual match to the schema’s ”% of system volume per day” definition; the make-up-water figure is preserved as the alternative candidate in the trials.csv remarks.
NO COLUMN (routed to Experimental Remarks in trials.csv, not extractable into a dedicated column): plant survival % (lettuce AP 54.6%/HYD 59.2%, ns; cucumber AP 100%/HYD 100%, ns); cucumber fruit length (AP 11.8/HYD 12.8 cm, ns), width (AP 4.1/HYD 4.2 cm, ns), weight (AP 127.6/HYD 150.4 g, ns), and fruit number (AP 1.76/HYD 8.5 m⁻², p<0.05); fish aggregate stocking/harvest totals (810 fish, 63.7 kg → 191.2 kg, yield 21.2 kg m⁻³ crop⁻¹, Table 2).
[not reported] fields, grouped: Fish Category; N, P, K (feed composition beyond protein/lipid); Total Feed (kg, cumulative); Fish biomass created (kg) and Fish weight gain (g) — inputs given (initial/final mean weight, total stocking/harvest weight) but no gain figure is stated as such, so left NR rather than derived; Water type; Water classification; pHOptimal; FUE AP/HYD; WUE; NO2-N; NO3-N (no nitrite/nitrate monitoring reported anywhere in the paper); SPAD; Leaf count; Tissue nitrate AP/HYD (no tissue analysis performed); Plant Category (both species); Plant height, Plant fresh weight, Plant dry matter for cucumber (not measured); Average room Temperature (Table 1’s temperature range is explicitly water temperature via the YSI pro2030 probe, not greenhouse air); Lat/Long (no coordinates given, only a facility address); Media Details; Control Parameters; Biological system already in use; Iron supplemented; Remineralization; pH Buffers; Climate control; Artificial Lighting.
No plant biochemistry/mineral/microbiology/proximate analytes were performed — only growth/agronomic metrics (survival, weight, height, fruit dimensions, yield), all of which are covered in trials.csv. plant_measurements.csv is header-only (0 rows) for this paper.
Water panel note (not excluded, but thin): Table 1’s water-quality panel (temperature, DO, pH, EC, ammonia) is explicitly titled “during aquaponic experimental culture” — no separate hydroponic-arm water-quality data is given, and no summary trial mean is given for any parameter, only ranges (recorded with “range only, no trial mean reported” per SCHEMA.md).
Publication year: the PDF’s running header/copyright line shows 2015 (published online 26 Oct 2015, ahead of print); the Zotero export and the formal journal citation (Aquaculture International 24(2):637–646) give 2016. Year 2016 was used per CLAUDE.md’s zotero-priority rule — this is a publication-lag artifact, not an internal contradiction within the paper.
Wikilink candidates (not yet confirmed against vault canonical forms): Specific Growth Rate (SGR) vs. Specific growth rate vs. SGR — check vault before this fragments; used “Specific Growth Rate (SGR)” here to match the existing “Feed Conversion Rate (FCR)” convention referenced in CLAUDE.md.
Source: Castillo-Castellanos et al. - 2016 - Implementation of an experimental nutrient film te.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
castillocastellanosImplementationExperimentalNutrient2016-T1
Fish
| Field | Value |
|---|---|
| Fish | Tilapia (Oreochromis niloticus var. Spring) |
| Initial Stock density | 7 kg/m3 (as stated, Discussion p.9: ‘7 kg m-3 (90 fishes weighing 78 g each in 1-m-3 tank)‘) |
| FCR | 0.99 |
| SGR | ⚠️MATERIAL 4.95 |
| Protein | 38 |
| % of body weight | 3.5 |
| Fish size initial | 78.1 +/- 1.9 |
| Fish size final | 242.9 +/- 8.9 |
| Feed routine | Four times a day |
| Feed regime | Nutripec (Purina, México) commercial fish feed, 3.5-mm pellets, 38% protein and 8% lipids, fed at 3.5% of total biomass (TB) per tank |
| Fish survival rate | 97.2 +/- 2.4 |
| Fish trial duration (days) | 49 |
Water
| Field | Value |
|---|---|
| Water recycle | 7.2 L/min |
| Water volume in the system | 1000 L (fish tank only; 1-m3 tank per Fig. 2 / Discussion p.9) |
| Daily Water exchange rate | ⚠️CHECK 800 (%/day, described as water recirculation, Materials and methods p.5) |
| Aq pH | 7.5-8.5 (range only, no trial mean reported, Table 1) |
| Dissolved Oxigen | 3.1-6.1 (range only, no trial mean reported, Table 1) |
| EC | 0.5-0.9 (range only, no trial mean reported, Table 1) |
| Water temperature | 26.6-31.6 (range only, no trial mean reported, Table 1) |
| TAN / NH4-N | 0.5-4.8 (range only, no trial mean reported, Table 1; paper reports as ‘ammonia’, not specified as TAN vs NH4-N) |
Plant
| Field | Value |
|---|---|
| Plant | Lettuce (Lactuca sativa), Parris Island (romaine) |
| Details | Parris Island romaine lettuce; grown in NFT hydroponic channels of the same aquaponic/hydroponic modules; harvested at the end of the 7-week culture cycle |
| Days Plant after transplant | ⚠️CHECK 49 |
| Plants/m2 | 10 |
| Plant height | 18.7 |
| Plant fresh weight | 18.8 |
| Plant dry matter | 1.5 (g/plant dry weight, not %, as reported) |
System & Setup
| Field | Value |
|---|---|
| System type | Nutrient film technique (NFT)-type aquaponic system, no sump (aquaponic arm); paired hydroponic control operated identically except the clarifier/biofilter contained nothing (p.5) |
| Air supplement | Y (1-Hp blower aeration (Pioneer RS-0750) with 1.5-inch PVC pipe circuit (threaded valves at 3/8” and 1/8” outputs), silicone hoses, and cylindrical air stones (0.5” diameter x 1” length) producing medium-sized bubbles, installed in the aquaculture module (p.4)) |
| Nutrient supplemented | N (Aquaponic arm relies solely on fish-derived nutrients from tilapia effluent, no synthetic solution added. Paired hydroponic control received Steiner (1984) universal nutrient solution: 50% concentration for the first 4 weeks, 75% for the remaining 3 weeks, replaced weekly (p.5)) |
| Equipment | Quiet One model 3000 external pump (0.12 L/s); 1-Hp blower aeration (Pioneer RS-0750); YSI pro2030 (temperature, dissolved oxygen, EC); HANNA HI98129 (pH); HANNA Multiparameter HI83200 (ammonia); ictiometer (fish length); OHAUS SP2001 balance (fish weight) |
| Control Parameters | NR (only monitored ranges given, Table 1; no automated setpoints/control system described) |
| Combination | Tilapia (O. niloticus var. Spring) + lettuce (L. sativa, Parris Island) and cucumber (C. sativus, Carolina) co-cultivated in the same NFT modules; aquaponic (9 systems) vs. hydroponic (3 systems) technique compared |
Site
| Field | Value |
|---|---|
| Region | North-America |
| Country | Mexico |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | g m-2 (yield); g plant-1 (wet/dry weight); cm plant-1 (height) |
| Statistic Details | One-way ANOVA (normality: Shapiro-Wilk; homogeneity: Cochran’s outlying variance C test), Tukey’s a posteriori test, significance p<0.05; survival data arcsine-transformed prior to analysis (p.6) |
| Statistically analysed | Y |
| Replicates (n) | ⚠️MATERIAL 3 |
| AP | 47.9 |
| HYD | 726.8 |
Experimental Remarks: TRIAL DEFINITION: T1 = lettuce grown in the NFT-type aquaponic system (no sump), one of 9 replicate aquaponic modules (Materials and methods p.5). Paired control = lettuce grown in the paired hydroponic module (3 replicate modules, same physical design, Steiner nutrient solution instead of fish effluent), recorded in the HYD columns. Only one aquaponic system design in this paper; T2 records cucumber, co-cultivated in the same modules/treatment. | UNIT CONVERSION ONLY: pump flow 0.12 L/s -> 7.2 L/min (Water recycle); tank volume ‘1 m-3’ -> 1000 L (Water volume in the system). | WARN-MATERIAL Replicates (n): Materials and methods (p.5) states ‘Nine of the 12 systems were utilized with the aquaponic technique, while the remaining three … employed only the hydroponic technique,’ implying 9 AP vs 3 HYD experimental units. Table 3’s own footnote (p.7), attached directly to the ANOVA/Tukey letters reported in that table, states ‘(p<0.05; n=3)’ for both columns. No sub-blocking of the 9 AP systems into groups of 3 is described, so it is unclear whether n=3 means only 3 of the 9 AP systems entered this particular statistical comparison, or the AP systems were pooled into 3 blocks. Recorded n=3, the value directly tied to the reported statistical test in Table 3 (the actual comparison this row reports); the Methods-stated system count (9 AP / 3 HYD) is preserved here for the record. Affects: confidence in the AP arm’s effective replication for every significance letter in Table 3. | WARN-MATERIAL SGR: Table 2(c)/Abstract/Results all consistently state ‘Specific growth (g day-1)’ = 4.95, repeated 3x with no competing value anywhere in the paper. Recomputing the paper’s own stated formula (Methods p.6: SG = (ln wt - ln w0)/t) with its own stated inputs (w0=78.1 g, wt=242.9 g, t=49 days, all Table 2) yields ~0.0231 (2.3%/day), not 4.95. The Table 2 unit label ‘g day-1’ is also dimensionally inconsistent with a log-ratio specific growth rate (a proportional %/day quantity, not grams/day). No second value exists in the paper to prefer over 4.95 (a check-fails-not-a-second-candidate case, not a two-source contradiction), so the printed value 4.95 is recorded in the SGR cell as the only value the authors give, flagged for this arithmetic mismatch. Affects: comparability of this paper’s SGR against other tilapia growth-rate studies. | WARN-CHECK Days Plant after transplant: Materials and methods (p.5) states ‘The experiment lasted 7 weeks for both treatments’ = 49 days, the only stated duration for the plant trial. Fig. 1 labels a ‘Seedling production room’ as part of the facility, implying seedlings were raised elsewhere before being moved into the hydroponic channels, but the Methods text says lettuce and cucumber ‘were seeded in hydroponic module in both techniques’ (p.6) — seeding, not transplanting. No explicit transplant date or seedling age at transplant is stated anywhere. Recorded 49 (the only duration given), but UNCLEAR whether this is counted from seed, an unstated transplant date, or is the fish-trial duration applied to plants by extension. | WARN-CHECK Daily Water exchange rate: Materials and methods (p.5) states ‘Water recirculation in each tank was about 800% daily,’ a closed-loop turnover rate, distinct from the separately-stated water replenishment ‘they supplied about 30 L every third day’ to replace evaporation/plant-uptake losses — two different senses of ‘exchange’ (loop turnover vs. net make-up water), neither mapped explicitly by the paper onto a single ‘daily exchange rate’ concept. Recorded 800%/day (recirculation) as the closer textual match to the column’s ’% of system volume per day’ definition; the ~30 L/3 days make-up-water figure is the alternative candidate, preserved here. | NO COLUMN: Lettuce plant survival AP 54.6% vs HYD 59.2% (ns, p>0.05, Table 3) — no dedicated plant-survival column in trials.csv. Fish stocking/harvest totals: 810 fish total, 63.7 kg total stocking weight, 191.2 kg total harvest weight, yield 21.2 kg m-3 crop-1 (Table 2) — aggregate figures with no matching column once per-fish mean weight and % survival are already captured. | NOT DERIVED, left NR: Fish biomass created (kg) — Table 2 gives total stocking (63.7 kg) and harvest (191.2 kg) weight, from which a gain is computable but not stated as such; Fish weight gain (g) — likewise computable from 78.1/242.9 g means but not stated as a ‘gain’; Total Feed (kg) — only a daily ration rate (3.5% TB/day) given, no cumulative total; N/P/K (feed) — only protein (38%) and lipid (8%) composition given; Fish Category, Water type, Water classification, Plant Category — paper does not categorise; pHOptimal, FUE AP/HYD, WUE, SPAD, Leaf count, Tissue nitrate AP/HYD, NO2-N, NO3-N — not measured/reported (Table 1’s water panel covers only temperature, oxygen, pH, EC and ammonia; no nitrite/nitrate monitoring reported for the aquaponic loop); Average room Temperature — Table 1’s temperature range is explicitly water temperature (measured with the YSI pro2030 water-quality probe), not greenhouse air temperature; Lat/Long — no coordinates given for the Nayarit, Mexico site, only a facility address. | Publication year note: PDF running header/copyright shows 2015 (published online 26 Oct 2015); Zotero export and the formal journal citation (Aquaculture International 24(2):637-646) give 2016. Year 2016 used per CLAUDE.md’s zotero-priority rule — not an internal contradiction of the paper itself.
castillocastellanosImplementationExperimentalNutrient2016-T2
Fish
| Field | Value |
|---|---|
| Fish | Tilapia (Oreochromis niloticus var. Spring) |
| Initial Stock density | 7 kg/m3 (as stated, Discussion p.9: ‘7 kg m-3 (90 fishes weighing 78 g each in 1-m-3 tank)‘) |
| FCR | 0.99 |
| SGR | ⚠️MATERIAL 4.95 |
| Protein | 38 |
| % of body weight | 3.5 |
| Fish size initial | 78.1 +/- 1.9 |
| Fish size final | 242.9 +/- 8.9 |
| Feed routine | Four times a day |
| Feed regime | Nutripec (Purina, México) commercial fish feed, 3.5-mm pellets, 38% protein and 8% lipids, fed at 3.5% of total biomass (TB) per tank |
| Fish survival rate | 97.2 +/- 2.4 |
| Fish trial duration (days) | 49 |
Water
| Field | Value |
|---|---|
| Water recycle | 7.2 L/min |
| Water volume in the system | 1000 L (fish tank only; 1-m3 tank per Fig. 2 / Discussion p.9) |
| Daily Water exchange rate | ⚠️CHECK 800 (%/day, described as water recirculation, Materials and methods p.5) |
| Aq pH | 7.5-8.5 (range only, no trial mean reported, Table 1) |
| Dissolved Oxigen | 3.1-6.1 (range only, no trial mean reported, Table 1) |
| EC | 0.5-0.9 (range only, no trial mean reported, Table 1) |
| Water temperature | 26.6-31.6 (range only, no trial mean reported, Table 1) |
| TAN / NH4-N | 0.5-4.8 (range only, no trial mean reported, Table 1; paper reports as ‘ammonia’, not specified as TAN vs NH4-N) |
Plant
| Field | Value |
|---|---|
| Plant | Cucumber (Cucumis sativus), Carolina |
| Details | Carolina cucumber; grown in NFT hydroponic channels of the same aquaponic/hydroponic modules; fruits harvested once reaching >=12 cm length (commercial size for this variety) |
| Days Plant after transplant | ⚠️CHECK 49 |
| Plants/m2 | 6.6 |
System & Setup
| Field | Value |
|---|---|
| System type | Nutrient film technique (NFT)-type aquaponic system, no sump (aquaponic arm); paired hydroponic control operated identically except the clarifier/biofilter contained nothing (p.5) |
| Air supplement | Y (1-Hp blower aeration (Pioneer RS-0750) with 1.5-inch PVC pipe circuit (threaded valves at 3/8” and 1/8” outputs), silicone hoses, and cylindrical air stones (0.5” diameter x 1” length) producing medium-sized bubbles, installed in the aquaculture module (p.4)) |
| Nutrient supplemented | N (Aquaponic arm relies solely on fish-derived nutrients from tilapia effluent, no synthetic solution added. Paired hydroponic control received Steiner (1984) universal nutrient solution: 50% concentration for the first 4 weeks, 75% for the remaining 3 weeks, replaced weekly (p.5)) |
| Equipment | Quiet One model 3000 external pump (0.12 L/s); 1-Hp blower aeration (Pioneer RS-0750); YSI pro2030 (temperature, dissolved oxygen, EC); HANNA HI98129 (pH); HANNA Multiparameter HI83200 (ammonia); ictiometer (fish length); OHAUS SP2001 balance (fish weight) |
| Control Parameters | NR (only monitored ranges given, Table 1; no automated setpoints/control system described) |
| Combination | Tilapia (O. niloticus var. Spring) + lettuce (L. sativa, Parris Island) and cucumber (C. sativus, Carolina) co-cultivated in the same NFT modules; aquaponic (9 systems) vs. hydroponic (3 systems) technique compared |
Site
| Field | Value |
|---|---|
| Region | North-America |
| Country | Mexico |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | g m-2 (yield) |
| Statistic Details | One-way ANOVA (normality: Shapiro-Wilk; homogeneity: Cochran’s outlying variance C test), Tukey’s a posteriori test, significance p<0.05; survival data arcsine-transformed prior to analysis (p.6) |
| Statistically analysed | Y |
| Replicates (n) | ⚠️MATERIAL 3 |
| AP | 78.7 |
| HYD | 421.6 |
Experimental Remarks: TRIAL DEFINITION: T2 = cucumber grown in the same NFT-type aquaponic system (no sump) as T1, one of 9 replicate aquaponic modules; co-cultivated with lettuce (T1) in the same physical modules (Materials and methods p.5-6: ‘Thirty-six lettuces and 24 cucumbers per hydroponic module were seeded’). Paired control = cucumber grown in the paired hydroponic module (3 replicate modules), recorded in the HYD columns. Fish/water/system-level fields below are identical to T1 (shared system) — see T1’s Experimental Remarks for the full evidence trail on the WARN flags reproduced here. | WARN-MATERIAL Replicates (n): see T1 remarks — Table 3’s cucumber section carries the same ‘(p<0.05; n=3)’ footnote as the lettuce section (one footnote covering the whole table), against Methods’ stated 9 AP / 3 HYD system count. Recorded n=3 for the same reason as T1. | WARN-MATERIAL SGR: see T1 remarks (fish-level figure, shared across T1/T2 since both plant trials ran in the same aquaponic systems using the same tilapia population). | WARN-CHECK Days Plant after transplant: see T1 remarks — same 49-day whole-experiment duration applies to cucumber, seeded/transplanted under the same unclear basis. | WARN-CHECK Daily Water exchange rate: see T1 remarks — 800%/day recirculation vs. ~30 L/3 days make-up water, shared system-level figure. | NO COLUMN: Cucumber plant survival AP 100% vs HYD 100% (ns, Table 3) — no dedicated plant-survival column. Fruit length AP 11.8 cm vs HYD 12.8 cm (ns, p>0.05); Fruit width AP 4.1 cm vs HYD 4.2 cm (ns); Fruit weight AP 127.6 g vs HYD 150.4 g (ns); Fruit number AP 1.76 m-2 vs HYD 8.5 m-2 (p<0.05, Table 3) — trials.csv has no dedicated fruit-dimension/fruit-count columns; only the area yield (g m-2) maps onto the AP/HYD columns used here. Commercial harvest cutoff: cucumbers cut only once reaching >=12 cm length (Materials and methods p.6), which the Discussion (p.8) offers as the likely reason fruit length/width/weight show no AP-vs-HYD difference despite yield and fruit number differing sharply. | NOT DERIVED, left NR: Plant fresh weight and Plant dry matter — the paper reports per-fruit weight (127.6/150.4 g, routed to NO COLUMN above) and area yield (routed to AP/HYD) for cucumber, but never a whole-plant fresh or dry weight analogous to lettuce’s; Plant height, Leaf count — not measured/reported for cucumber; Plant Category — paper does not categorise; other NR fields as in T1 remarks (shared fish/water-level fields). | Publication year note: see T1 remarks.