Growth, yield, plant quality and nutrition of basil (Ocimum basilicum L.) under soilless agricultural systems
Metadata
- Cite key: sahaGrowthYieldPlant2016
- Item type: Journal Article
- Authors: S. Saha, A. Monroe, M. Day
- Affiliation: Department of Biology, Georgia Southern University, 4324 Old Register Road, Statesboro, GA 30460, USA
- Journal: Annals of Agricultural Sciences 61(2) (2016) 181-186
- Date: 12/2016
- Date added: 2019-01-18
- DOI: 10.1016/j.aoas.2016.10.001
- Funding: Georgia Southern University Sustainability Fee Grant
- URL: https://doi.org/10.1016/j.aoas.2016.10.001
- PDF:
Saha et al. - 2016 - Growth, yield, plant quality and nutrition of basi.pdf
Opinion
A small, tidy undergraduate-adjacent greenhouse comparison (same lab and system as
monroeEffectsCrayfishAquaponics2015, a companion lettuce poster) that is the first to test crayfish specifically as the aquatic species for basil. The headline result — aquaponic basil beating hydroponic basil on height/fresh weight/dry weight by 14/56/65%, with no difference in leaf nutrients or chlorophyll — is internally consistent and the authors’ own percentage claims all recompute correctly from the reported means. The main soft spot is the replication structure: only one physical tank per treatment per location (4 locations total), with the “five replicates” in the ANOVA actually being five basil plants sharing that one tank — true independent-system replication is n=4/treatment, not n=20. This is disclosed by the authors, not hidden, but it means the significance tests should be read as somewhat optimistic. No fish/crayfish performance data (growth, survival, FCR) is reported at all — crayfish are purely a nutrient source here, same pattern as the lab’s lettuce poster. Worth citing as a second basil-aquaponics data point with a novel aquatic species, but the small n and pseudoreplication caveat should travel with any citation of the significance results.
Abstract
Traditional agricultural systems are challenged by globally declining resources resulting from climate change and growing population. Alternative agricultural practices such as aquaponics (includes crop plant and aquatic species) and hydroponics (includes crop plant only) have the potential to generate high yield per unit area using limited land, water, and no soil. A soilless agricultural study was conducted at the Georgia Southern University, Statesboro, GA, USA from August to November, 2015. The growth, yield, quality, and nutrition of basil (Ocimum basilicum L.) cultivar Aroma2, were compared between aquaponic and hydroponic systems using crayfish (Procambarus spp.) as the aquatic species. Non-circulating floating raft systems were designed using 95 L polyethylene tanks. Equal amounts of start-up fertilizer dose were applied to both systems. The objective was to understand how the additional nutritional dynamics associated with crayfish influence the basil crop. Both fresh and dry basil plant weights were collected after harvest, followed by leaf nutrient analysis. Leaf chlorophyll content, water pH, nitrogen and temperature were measured periodically. Aquaponic basil (AqB) showed 14%, 56%, and 65% more height, fresh weight, and dry weight, respectively, compared to hydroponic basil (HyB). It is logical to assume that crayfish waste (excreta and unconsumed feed) has supplied the additional nutrients to AqB, resulting in greater growth and yield. The chlorophyll content (plant quality) or leaf nutrients, however, did not differ between AqB and HyB. Further research is needed to investigate aquaponic crayfish yield, overall nutritional dynamics, cost-benefit ratio, and other plant characteristics under soilless systems.
Summary
Researchers at Georgia Southern University ran a greenhouse comparison (Aug-Nov 2015) of basil (cv. Aroma 2) grown in a non-circulating floating-raft aquaponic system (95 L tanks stocked with 4 mature red crayfish/White River crayfish each, Procambarus clarkii / P. zonangulus) against an identical hydroponic control with no crayfish. Both treatments received the same one-time start-up dose of a commercial liquid fertilizer (Floranova Grow 7:4:10); the aquaponic tanks additionally received crayfish waste and a small weekly ration of algal wafers/Spirulina flakes fed to the crayfish. The design was a randomized block layout across 4 greenhouse locations, with one aquaponic and one hydroponic tank per location (4 tanks/treatment total) and 5 basil plants per tank, analyzed by GLM/ANOVA (SAS). At harvest, aquaponic basil was significantly taller (89.9 vs 78.7 cm, +14%), heavier fresh (150.2 vs 96.6 g/plant, +56%) and heavier dry (15.9 vs 9.6 g/plant, +65%) than hydroponic basil, all p<0.05. Leaf chlorophyll (SPAD) and a full leaf macro/micronutrient panel (N, P, K, Mg, Ca, S, B, Cu, Fe, Mn, Na, Zn) showed no significant differences between the two systems, despite aquaponic tanks carrying substantially higher measured water nitrate/nitrite/ammonia over the final 8 weeks. The authors attribute the growth/yield advantage to additional nutrients supplied by crayfish excreta and uneaten feed, and note that this extra nutrient supply changed how much the plant grew without changing the nutrient concentration of its tissue. No crayfish growth, survival, or feed-conversion data are reported anywhere in the paper.
Experiment data
- Location: Climate-controlled greenhouse, Department of Biology, Georgia Southern University, Statesboro, GA, USA (32.4453, -81.7792)
- Design: Randomized block design; 2 treatments (aquaponic, hydroponic) x 4 locations/blocks within the greenhouse; one 95 L non-circulating floating-raft tank per treatment per location (4 tanks/treatment total), 5 basil plants per tank; GLM with treatment x location (+ interaction) terms
- Replicates / n: Paper states “five replicates at location j receiving treatment i” (i.e. 5 plants/tank x 4 locations = 20 plants/treatment analyzed); true independent physical-system replication is n=4 tanks/treatment — see Extraction notes on pseudoreplication
- Duration: Not explicitly stated as a number of days/weeks anywhere in the text. Seedlings (3-week-old) transplanted 2015-08-14; harvested 2015-11-19 (~97 days between these two stated dates, not itself recorded as a duration by the authors — left NR per the no-derivation rule)
- Organisms: Basil (Ocimum basilicum) cv. Aroma 2 / Crayfish (red crayfish Procambarus clarkii, White River crayfish Procambarus zonangulus)
- Statistics: General linear model (PROC GLM), factors = treatment x location/block; ANOVA; SAS 9.3 (SAS Institute, 2012); alpha = 0.05
- Basil yield: Height 89.9+/-4.5 cm (AqB) vs 78.7+/-3.9 cm (HyB), +14%, p<0.05; fresh weight 150.2+/-18 g/plant vs 96.6+/-10.4 g/plant, +56%, p<0.05; dry weight 15.9+/-2 g/plant vs 9.6+/-1 g/plant, +65%, p<0.05 (all SE, not SD)
- Leaf chlorophyll (SPAD) and leaf nutrients: No significant difference AqB vs HyB for SPAD or any of 12 leaf macro/micronutrients measured (N, P, K, Mg, Ca, S, B, Cu, Fe, Mn, Na, Zn)
- Water nitrogen species (last 8 weeks before harvest): Aquaponic tanks markedly higher than hydroponic for nitrate (95+/-19.2 vs 80+/-12.5 ppm), nitrite (1.26+/-0.4 vs 0.3+/-0.3 ppm) and ammonium (0.34+/-0.1 vs 0.03+/-0.01 ppm) — not itself statistically tested in the paper
Yield and growth
This paper: Aquaponic basil (AqB) significantly outperformed hydroponic basil (HyB) on all three growth/yield measures taken at harvest: plant height (89.9+/-4.5 vs 78.7+/-3.9 cm, +14%, p<0.05, Fig. 1), fresh weight (150.2+/-18 vs 96.6+/-10.4 g/plant, +56%, p<0.05, Fig. 3), and dry weight (15.9+/-2 vs 9.6+/-1 g/plant, +65%, p<0.05, Fig. 4). Weekly height tracking (Fig. 2) shows the AqB advantage was consistent across the growing period rather than a late-season effect. Scaled to a hectare basis using the 5-plant-per-lid dry weight (79.5 g AqB, 48 g HyB per 506.25 cm2), the authors report a dry-matter yield of 15,900 kg/ha (AqB) vs 9,600 kg/ha (HyB) — see Extraction notes for a minor rounding discrepancy in this specific area-scaling step. The authors attribute the advantage to crayfish-derived nutrients (excreta + uneaten feed) supplementing the equal start-up fertilizer dose both treatments received, consistent with the markedly higher measured water nitrate/nitrite/ammonium in the aquaponic tanks (see Water quality section).
Compared with:
- todo Rakocy et al. 2004 — aquaponic basil yield of 1.8 kg/m2 vs 0.6 kg/m2 for field-grown basil, cited in the Introduction as evidence basil generally does better under soilless/aquaponic systems (p.182); different comparison (soilless vs field, not aquaponic vs hydroponic) and no hydroponic control in that study.
- todo Savidov 2005 and Lennard and Leonard 2006 — cited as reporting greater production in tomato, cucumber and lettuce under aquaponics vs hydroponics, offered as corroborating the direction of this paper’s basil result (p.184).
- todo Effendi et al. 2015 — spinach-aquaponic systems gave 5% higher crayfish survival than crayfish monoculture, cited in the Introduction as evidence plant biofilters (incl. spinach) benefit crayfish; this paper does not itself report any crayfish survival data of its own (p.182).
- todo Gallardo-Colli et al. 2014 — tilapia cohabited with crayfish (Procambarus acanthophorus) in an aquaponic system produced 9.4% higher green corn fodder (Zea mays) yield than hydroponics; cited as prior crayfish-aquaponic precedent (p.182).
Plant quality and leaf nutrition
This paper: Leaf chlorophyll content (SPAD, mean of one upper + one lower leaf/plant) did not differ significantly between AqB (29.3+/-0.3) and HyB (28.7+/-0.3) at harvest (Fig. 5), nor across the weekly time series (Fig. 6, no numeric values given). A full leaf macro/micronutrient panel — N, P, K, Mg, Ca, S (Table 1, % dry weight) and B, Cu, Fe, Mn, Na, Zn (Table 2, mg/kg dry weight) — likewise showed no significant AqB-vs-HyB difference for any of the 12 analytes (no letter superscripts in either table; confirmed “None of the macronutrients differ…” / “None of the micronutrient contents were shown to be different…” in Results, p.185). The authors interpret this as showing that the extra crayfish-derived nutrients drove more growth without changing the concentration of nutrients accumulated in leaf tissue — i.e. the plants grew bigger, not “richer.”
Compared with:
- todo Rakocy and Hargreaves 1993 — also found no chlorophyll-content difference between aquaponic and hydroponic lettuce, cited as a directly corroborating precedent for this paper’s null SPAD result (p.185).
Water quality
This paper: Mean water pH did not differ significantly between aquaponic (6.9+/-0.24) and hydroponic (7+/-0.15) tanks; both are described as within the desirable 6-9 range for freshwater aquatic organisms (citing Boyd and Tucker, 1998, a general literature range, not this paper’s own target). Mean daytime water temperature was 23.4+/-0.41degC and greenhouse mean daytime air temperature 25.1+/-0.72degC, both reported as single values not split by treatment. Over the final 8 weeks before harvest, aquaponic tanks averaged higher nitrate (95+/-19.2 vs 80+/-12.5 ppm), nitrite (1.26+/-0.4 vs 0.3+/-0.3 ppm) and ammonium (0.34+/-0.1 vs 0.03+/-0.01 ppm) than hydroponic tanks — consistent with the crayfish-nutrient-supply explanation for the yield difference, though this comparison is not itself run through a significance test in the paper. See Extraction notes for an unresolved question about whether the nitrate figures are reported as NO3 or NO3-N.
Compared with: (no external literature comparison given for the paper’s own water-chemistry figures specifically)
Linked claims
- Crayfish waste can supply additional plant-available nutrients in an aquaponic system
- Aquaponics and hydroponics can produce statistically equivalent leaf nutrient content despite yield differences
- Crayfish can serve as the aquatic species in a floating-raft aquaponics system
Citations to chase
- todo Rakocy, J.E., Shultz, R.C., Bailey, D.S., Thoman, E.S. (2004) — Aquaponic production of tilapia and basil: comparing a batch and staggered cropping system, Acta Hort. 648, 63-69 — 1.8 kg/m2 aquaponic basil vs 0.6 kg/m2 field basil
- todo Savidov, N. (2005) — Evaluation of aquaponics technology in Alberta, Canada, Aquaponics J. 37, 20-25 — higher aquaponic yields for tomato/cucumber
- todo Lennard, W.A., Leonard, B.V. (2006) — A comparison of three different hydroponic sub-systems (gravel bed, floating and nutrient film technique) in an Aquaponic test system, Aquacult. Int. 14, 539-550
- todo Effendi, H., Utomo, B.A., Darmawangsa, G.M. (2015) — Phytoremediation of freshwater crayfish (Cherax quadricarinatus) culture wastewater with spinach (Ipomoea aquatica) in aquaponic system, AACL Bioflux 8(3), 421-430 — 5% higher crayfish survival in spinach-aquaponic vs monoculture
- todo Gallardo-Colli, A., Hernandez-Vergara, M.P., Perez-Rostro, C.I., Ramirez-Gutierrez, S.C. (2014) — Biculture tilapia/crayfish in aquaponic system, Glob. Adv. Res. J. Agric. Sci. 3(8), 233-244 — 9.4% higher green corn fodder yield vs hydroponics
- todo Rakocy, J.E., Hargreaves, J.A. (1993) — Integration of vegetable hydroponics with fish culture: a review — no chlorophyll difference in aquaponic vs hydroponic lettuce
Extraction notes
TRIAL DEFINITION: T1 = the sole aquaponic treatment (crayfish + basil, non-circulating floating raft) vs its paired hydroponic control, recorded in the HYD-labelled cells. Only one aquaponic treatment in this paper (no density/species sub-treatments), so one trials.csv row, per the alcarrazQualityLettuceLactuca2018 single-treatment convention.
WARN-CHECK NO3-N basis unstated (p.183). Results states: “The mean content of nitrate, nitrite, and ammonium during last 8 weeks before harvest were (in ppm), 95±19.2 and 80±12.5 […] for aquaponic and hydroponic tanks, respectively.” Methods states these were “measured weekly using freshwater aquarium master test kit” (p.183) — no kit brand named. Consumer/hobbyist freshwater test kits of this type conventionally report a “nitrate (NO3)” colorimetric reading in ppm, not NO3-N, but the paper never states which basis its 95/80 ppm figures use. Recorded the literal value (95±19.2 ppm, AP) in the trials.csv NO3-N cell, since that is the schema’s only nitrate-species water column, with this basis caveat attached; if the kit in fact reports NO3 rather than NO3-N, the true NO3-N figure would be roughly 4.43x smaller (~21.4 ppm). Affects: NO3-N cell only. Added to REVIEW.md by the batch merge step.
WARN-MINOR dry-matter area-yield rounding (p.184). Results states: “Five plants were grown in each container that resulted total mean dry plant weight of 79.5g (AqB) and 48g (HyB) per lid surface area (506.25cm2). This translates to a dry basil yield of 15,900kg ha-1 in aquaponics and 9600kg ha-1 in hydroponics.” Recomputing directly from the paper’s own stated inputs gives 79.5g/0.050625m2 = 1,570.4 g/m2 = 15,704 kg/ha, and 48g/0.050625m2 = 948.1 g/m2 = 9,481 kg/ha — about 1.2% below the paper’s stated 15,900/9,600. Both stated figures match exactly if a rounded lid area of 500 cm2 is used instead (79.5g/500cm2 = 15,900 kg/ha exactly; 48g/500cm2 = 9,600 kg/ha exactly), suggesting the authors used a rounded 500 cm2 for this specific extrapolation rather than the precise 506.25 cm2 given elsewhere in Methods. No cell affected: trials.csv AP/HYD columns hold the paper’s own literally-stated 15,900/9,600 kg/ha, not this recomputation.
Pseudoreplication design note (not a contradiction). Methods states “Two treatments had been randomly assigned between two plots [per location]… Hence there were five replicates at location j receiving treatment i” (p.183), and separately “Four tanks each for aquaponics and hydroponics were allotted” across 4 locations (p.182). So true independent physical-system replication is n=4 tanks/treatment (one AP tank + one HYD tank per location x 4 locations); the “5 replicates” the paper’s own ANOVA treats as independent are the 5 basil plants sharing that one tank — plant-level subsamples, not independent systems. Replicates (n) recorded as 20 (the paper’s own stated per-location “replicate” count of 5, summed across 4 locations); the tank-level n=4 is noted here for methodological caution when interpreting the significance results.
UNIT CONVERSION ONLY: coordinates (p.182). The PDF text layer renders the site coordinate as the corrupted string “USA(32”2604300N,81”4604500W)”. Character-level inspection (pdfplumber glyph positions) shows this decomposes as digits “32”, a baseline-shifted quote-glyph (matching a degree-symbol substitution), digits “26”, one further baseline-shifted “0” (matching a single prime/minute-mark substitution), digits “43”, two more baseline-shifted “0”s (matching a double-prime/second-mark substitution), then “N” — i.e. the source PDF’s embedded font substituted degree/prime/double-prime glyphs with digit-shaped glyphs, rather than the authors having mis-written a DMS coordinate. Decoded as 32°26’43”N, and by the identical pattern the longitude string as 81°46’45”W. Converted: Lat = 32 + 26/60 + 43/3600 = 32.4453; Long = -(81 + 46/60 + 45/3600) = -81.7792. Both minutes/seconds are valid (<60). Confirmed geographically consistent with the paper’s own stated site, “the biology department of Georgia Southern University, Statesboro, GA” (actual campus coordinates are approximately 32.42-32.45N, 81.78-81.79W). Original corrupted string preserved here per SCHEMA.md’s coordinate-recovery guidance.
NOT DERIVED, left NR:
- Fish trial duration (days) / Days Plant after transplant — transplant date (2015-08-14) and harvest date (2015-11-19) are both explicitly stated (a ~97-day span is directly computable), but the paper never itself states a duration in days/weeks anywhere in the text. Per the no-derivation rule, and the precedent set in
pantanellaAquaponicsHydroponicsProduction2012(crop-2 duration left NR despite a stated start date), left NR rather than computed. Crayfish stocking date relative to transplant is not stated either, adding further uncertainty to any fish-side duration specifically. - Initial Stock density — “Four mature crayfishes were released in each aquaponic tank” (83 L water) is given, but no crayfish weight (individual or aggregate) is stated anywhere, so no kg/m3 figure is computable even by derivation.
- Total Feed (kg) — weekly per-tank feed rate is stated (3g algal wafers + 1.5g Spirulina flakes = 4.5 g/tank/week), and duration is approximately inferable from the transplant/harvest dates, but since duration itself is not stated by the paper (see above) and total feed across the trial is never itself stated as a figure, left NR rather than back-calculated.
- FCR, SGR, Fish size initial/final, Fish weight gain, Fish biomass created (kg), Fish survival rate — the crayfish are never weighed, tracked for growth, or assessed for survival anywhere in this paper; purely a nutrient source for the basil crop (same pattern noted in
pantanellaAquaponicsHydroponicsProduction2012andmonroeEffectsCrayfishAquaponics2015, the latter the same lab/system). - Protein (feed) — two distinct feed products with two different crude-protein percentages (algal wafers 30% CP, Spirulina flakes 55% CP, p.183) were fed together with no single blended composition figure stated; recording either percentage alone in the single Protein cell would misrepresent a two-product diet, so left NR with both percentages captured in Feed regime instead.
PDF-quality / [unclear]: container dimensions (p.182). Methods states (as extracted): “Dark polyethylene containers (16.88 18.75 27cm3 – h w d) of 95L (25Gal) capacity were used as study tanks.” The three given figures (16.88, 18.75, 27 cm) do not reconcile to the stated 95 L capacity under a simple length x width x height product (=8,551 cm3 = 8.6 L), nor under an inches-misprint reading (~140 L). The lid dimensions given two sentences later (“18.75cm x 27cm”, surface area 506.25 cm2) are internally consistent with two of the three h/w/d figures, suggesting the multiplication signs and/or a units qualifier were lost in the source PDF’s text layer around this specific parenthetical, rather than the authors having mis-stated the tank capacity. Recorded the container capacity/fill volume as literally stated (95 L/25 Gal nominal, 83 L/22 Gal fill) in Water volume in the system; the un-reconciled h/w/d figures are preserved verbatim in Media Details rather than corrected or guessed.
SE, not SD, throughout. Every dispersion value in this paper (Tables 1-2, Figures 1 and 3-6, and all in-text water/temperature figures) is explicitly stated by the authors to be standard error: “The numbers presented with ± are standard error (SE) values” (footnote, p.183) and “±SE represents standard error” (every figure caption, p.184). All ”±” values in trials.csv and plant.csv for this paper are SE, not SD — flagged so downstream analysis does not mistake them for SD.
Water recycle = NA, not NR. Methods explicitly describes the design as a “non-circulating floating raft system” (p.183) — the paper states there is no recycle/flow loop at all, rather than staying silent about one, so Water recycle is recorded NA (not applicable to this design) rather than NR.
Type classification. Recorded as experiment. Methods explicitly states “the experiment was designed as randomized block design” with treatments “randomly assigned” to plots within each of 4 locations/blocks (p.183), true replication (4 independent tanks/treatment), and a formal statistical model (GLM/ANOVA, SAS 9.3, alpha=0.05) — meeting SCHEMA.md Part 2 decision rule 2 cleanly (explicit “randomized” language is present, unlike the judgment call required for pantanellaAquaponicsHydroponicsProduction2012).
[not reported] fields, grouped:
- Fish: Fish Category (beyond the paper’s own “commercial aquatic species” phrase, used instead), Initial Stock density, FCR, SGR, feed N/P/K %, % of body weight, Fish size initial/final, Total Feed (kg), Fish biomass created (kg), Fish survival rate, Fish weight gain, Fish trial duration (days) — see NOT DERIVED block above.
- Water: Water classification, Daily Water exchange rate, pHOptimal, FUE AP, FUE HYD, WUE, Dissolved Oxygen, EC — none measured or stated anywhere in the paper.
- Plant: Days Plant after transplant (see NOT DERIVED above), Leaf count (only height, fresh weight, dry weight and SPAD were measured), Tissue nitrate AP/HYD (leaf tissue was analyzed for N/P/K/Mg/Ca/S and B/Cu/Fe/Mn/Na/Zn, never for nitrate itself as a distinct tissue analyte).
Tags judgment call: Meta/Fish/Crayfish reuses the existing facet from monroeEffectsCrayfishAquaponics2015 (an overlapping-author companion poster from the same Georgia Southern lab/system); Meta/Plant/Basil reuses the existing facet from mourantianBasilFunctionalGrowth2023 and others. Meta/Region/North-America per Georgia, USA.
Author wikilink judgment call: Reused the vault’s existing initialed spellings S. Saha, A. Monroe, M. Day, established in monroeEffectsCrayfishAquaponics2015 (same three authors, an overlapping poster/paper from the same lab), in preference to this paper’s own fuller byline form “Martin R. Day”, per CLAUDE.md’s instruction to check for an existing spelling first.
Metadata sourcing: zotero-export.csv contains a matching row (Key 4955X868) confirming title, authors (Saha, Subhrajit; Monroe, Amber; Day, Martin R.), year, DOI, journal, volume/issue/pages and Date Added (2019-01-18); independently cross-checked against Crossref (https://api.crossref.org/works/10.1016/j.aoas.2016.10.001), which agrees on all fields. Minor cosmetic note: the PDF’s own running header renders the journal name as “Annals of Agricultural Science” (singular), while both Zotero and Crossref give “Annals of Agricultural Sciences” (plural) — the latter used here as authoritative per CLAUDE.md. Zotero’s own Manual Tags field for this item (“review; ✔️; yield; comparisontable”) is the user’s personal Zotero tagging, not a type classification, and was not used to determine type: here (paper is clearly experiment, see above).
PDF quality: Clean native text layer throughout (6 pages, standard two-column Elsevier layout), fully extractable via pdfplumber; no OCR needed, not routed to NEEDS_OCR.md. Aside from the coordinate and container-dimension glyph-substitution issues noted above (both isolated to their specific parenthetical spans), the rest of the text layer extracted cleanly, including the bar-chart data-label numbers embedded in Figs. 1/3/4/5 (all of which are also independently stated in the running Results text, so none of this note’s numbers were read from a chart with no corroborating text/table value).
Source: Saha et al. - 2016 - Growth, yield, plant quality and nutrition of basi.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
sahaGrowthYieldPlant2016-T1
Fish
| Field | Value |
|---|---|
| Fish | Crayfish (Red crayfish Procambarus clarkii; White River crayfish Procambarus zonangulus) |
| Fish Category | Commercial aquatic species (p.182) |
| Feed routine | Weekly |
| Feed regime | Algal wafers (Tetra Holding Inc., 30% crude protein) 3 g/tank + Spirulina flakes (Ocean Nutrition, 55% crude protein) 1.5 g/tank, applied weekly to aquaponic tanks only (p.182-183) |
Water
| Field | Value |
|---|---|
| Water volume in the system | 83 L (95 L/25 Gal polyethylene container capacity, filled with 83 L/22 Gal tap water, p.182; single combined tank per replicate — non-circulating floating raft, one tank = the whole system, not a separate fish-tank/grow-bed split) |
| Water type | Tap water, dechlorinated (AquaSafe, 10 mL/37.85L rate, 22 mL/tank), neutral starting pH (p.182) |
| Aq pH | 6.9 +/- 0.24 (SE) |
| Water temperature | 23.4 +/- 0.41 (SE) — mean daytime water temperature, not stated separately for AP vs HYD (p.183) |
| TAN / NH4-N | 0.34 +/- 0.1 (SE) — aquaponic tanks, mean of last 8 weeks before harvest (p.183) |
| NO2-N | 1.26 +/- 0.4 (SE) — aquaponic tanks, mean of last 8 weeks before harvest (p.183) |
| NO3-N | WARN-CHECK 95 +/- 19.2 (SE) ppm — aquaponic tanks, mean of last 8 weeks before harvest; NO3 vs NO3-N basis not stated (freshwater aquarium master test kit). See Experimental Remarks. |
Plant
| Field | Value |
|---|---|
| Plant | Basil (Ocimum basilicum L.) cv. Aroma 2 |
| Details | Seedlings from a local nursery (Newington, GA), 3-week-old at transplant; transplanted into slotted net pots (coconut coir + vermiculite, 60:40) on 2015-08-14; 5 plants per 506.25 cm2 lid area; harvested by cutting at the base on 2015-11-19 |
| Plant Category | Annual herb (p.182) |
| Plants/m2 | 100 |
| SPAD (aquaponics) | 29.3 +/- 0.3 (SE) |
| Plant height | 89.9 +/- 4.5 (SE) |
| Plant fresh weight | 150.2 +/- 18 (SE) |
| Plant dry matter | 15.9 +/- 2 (SE) g/plant (dry weight, not percent) |
System & Setup
| Field | Value |
|---|---|
| System type | Non-circulating floating raft system (p.183) |
| Media Details | Slotted net pots, 12.7 cm (5 in) diameter, coconut coir lining + vermiculite (60:40); dark polyethylene container lids with 5 circular cuts (one per corner + center); lid surface area 506.25 cm2 (18.75 cm x 27 cm); container h/w/d figures as printed do not reconcile with stated 95 L capacity, likely a PDF text-layer artifact — see Experimental Remarks |
| Biological system already in use | Y (Zym-Bac (Home Grown Ponics), a nitrifying-bacteria source, applied once at study start to all tanks (both AP and HYD) to promote nitrification (p.182-183); no separate/dedicated biofilter compartment — combined single-tank floating raft system) |
| Air supplement | Y (Two 30 mm round air stones + a double air pump (75-225 L capacity) placed in all tanks, both AP and HYD (p.182)) |
| Iron supplemented | Y (0.1% Fe included in the one-time Floranova Grow (7:4:10) start-up fertilizer dose, applied equally to AP and HYD tanks (1.25 mL/L, 104 mL/tank) (p.183); not a targeted/isolated Fe supplement, part of the base formulation) |
| Climate control | Y (Climate-controlled greenhouse, Georgia Southern University, Statesboro, GA (p.185); no specific setpoints given) |
| Nutrient supplemented | Y (Floranova Grow (7:4:10, General Hydroponics Inc.) one-time start-up dose: 7% N, 4% P2O5, 10% K2O, 4% Ca, 2% S, 1.5% Mg, 0.1% Fe, <0.1% each B/Cl/Co/Mn/Mo/Zn; 1.25 mL/L, 104 mL/tank; applied once to both AP and HYD tanks at study start (p.182-183)) |
| Equipment | SPAD 502P chlorophyll meter (Spectrum Technologies Inc., Aurora, IL); freshwater aquarium master test kit (nitrate/nitrite/ammonium); standard ruler (height); Fisher Scientific Isotemp Standard Lab Oven (60C/96h drying); ICP-OES Spectro Arcos FHS16 (Germany) for leaf digestion/analysis (EPA Method 3052 digestion, EPA Method 200.8 analysis); double air pump + air stones; SAS 9.3 |
| Control Parameters | Water pH & temperature measured 2x/week; greenhouse daytime air temperature measured daily; water nitrate/nitrite/ammonium measured weekly (freshwater test kit); SPAD measured weekly; plant height measured weekly |
| Combination | Crayfish (Procambarus clarkii, Procambarus zonangulus) and basil (Ocimum basilicum cv. Aroma 2) in a non-circulating floating-raft aquaponic system vs. an equivalent hydroponic control |
Site
| Field | Value |
|---|---|
| Region | North America |
| Country | USA |
| Lat | 32.4453 |
| Long | -81.7792 |
| Average room Temperature | 25.1 +/- 0.72 (SE) — greenhouse mean daytime air temperature (p.183) |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | cm (height); g/plant (fresh & dry weight); SPAD units (chlorophyll); kg/ha (dry-matter area yield); % or mg/kg (leaf macro/micronutrients, see plant.csv); ppm (water N species) |
| Statistic Details | General linear model (PROC GLM), factors = treatment (aquaponic/hydroponic) x location/block (4 greenhouse locations); SAS 9.3 (SAS Institute 2012); alpha=0.05; lowercase-letter significance grouping on Figs 1, 3, 4 |
| Statistically analysed | Y |
| Replicates (n) | 20 |
| AP | 15900 |
| HYD | 9600 |
Experimental Remarks: TRIAL DEFINITION: T1 = the sole aquaponic treatment (Procambarus clarkii / P. zonangulus crayfish + basil, non-circulating floating raft) vs its paired hydroponic control (basil only, no crayfish, same fertilizer start-up dose), recorded in the HYD-labelled cells. Only one aquaponic treatment in this paper (no density/species sub-treatments), so one row, per the alcarrazQualityLettuceLactuca2018 single-treatment convention. | WARN-CHECK NO3-N basis unstated, p.183. Results states: ‘The mean content of nitrate, nitrite, and ammonium during last 8weeks before harvest were (in ppm), 95±19.2 and 80±12.5 […] for aquaponic and hydroponic tanks, respectively.’ Methods states these were ‘measured weekly using freshwater aquarium master test kit’ (p.183). Consumer/hobbyist freshwater test kits of this kind (e.g. API Freshwater Master Test Kit) conventionally report a ‘nitrate (NO3)’ colorimetric reading in ppm, not NO3-N; the paper never states which basis its 95/80 ppm figures use, nor does it name the kit brand. Recorded the literal value (95±19.2 ppm, AP) in the NO3-N cell since that is the schema’s only nitrate-species water column, with this basis caveat attached; if the kit in fact reports NO3 rather than NO3-N, the true NO3-N figure would be roughly 4.43x smaller (~21.4 ppm). Affects: NO3-N cell only (no tissue-nitrate or plant.csv cell depends on this water reading). Added to REVIEW.md by the batch merge step. | WARN-MINOR dry-matter area-yield rounding, p.184. Results states: ‘Five plants were grown in each container that resulted total mean dry plant weight of 79.5g (AqB) and 48g (HyB) per lid surface area (506.25cm2). This translates to a dry basil yield of 15,900kg ha-1 in aquaponics and 9600kg ha-1 in hydroponics.’ Recomputing directly from the paper’s own stated inputs (79.5 g and 48 g per 506.25 cm2) gives 79.5/0.050625m2=1570.4 g/m2=15,704 kg/ha and 48/0.050625m2=948.1 g/m2=9,481 kg/ha — roughly 1.2% below the paper’s stated 15,900/9,600. Both stated figures instead match EXACTLY if a rounded lid area of 500 cm2 is used: 79.5g/500cm2=1590 g/m2=15,900 kg/ha; 48g/500cm2=960 g/m2=9,600 kg/ha. The authors most likely used a rounded 500 cm2 for the hectare-extrapolation rather than the precise 506.25 cm2 stated elsewhere in Methods. No cell affected: the AP/HYD columns hold the paper’s own literally-stated 15,900/9,600 kg/ha, not this recomputation; shown here only as the arithmetic cross-check required for derived-adjacent values. | DESIGN NOTE (not a contradiction): Methods states ‘Two treatments had been randomly assigned between two plots [per location]… Hence there were five replicates at location j receiving treatment i’ (p.183), and separately ‘Four tanks each for aquaponics and hydroponics were allotted’ across 4 locations (p.182). So the true independent physical-system replication is n=4 tanks/treatment (one AP tank + one HYD tank per location x 4 locations); the ‘5 replicates’ the paper’s own ANOVA treats as independent are the 5 basil plants sharing that one tank, i.e. plant-level subsamples rather than independent systems. Replicates (n) recorded as 20 (4 locations x 5 plants = total plants/treatment analyzed), the paper’s own stated per-location ‘replicate’ count summed across locations; the tank-level n=4 is noted here for methodological caution, not entered as the cell value. | UNIT CONVERSION ONLY: coordinates, p.182. The PDF text layer renders the site coordinate as the corrupted string ‘USA(32”2604300N,81”4604500W)’ — character-position inspection (pdfplumber char-level extraction) shows this decomposes as digits ‘32’, a raised quote-glyph (baseline-shifted, matching a degree-symbol substitution), digits ‘26’, a single baseline-shifted ‘0’ (matching a single prime/minute-mark glyph substitution), digits ‘43’, two baseline-shifted ‘0’s (matching a double-prime/second-mark glyph substitution), then ‘N’ — i.e. the font’s degree/prime/double-prime glyphs were substituted with digit-shaped glyphs by a broken embedded-font cmap, not a DMS-writing error by the authors. Decoded as 32°26’43”N, and by the identical pattern the longitude string as 81°46’45”W. Converted: Lat 32 + 26/60 + 43/3600 = 32.4453; Long -(81 + 46/60 + 45/3600) = -81.7792. Both minutes/seconds are valid (<60). Confirmed geographically consistent with the paper’s stated site, ‘the biology department of Georgia Southern University, Statesboro, GA’ (actual campus coordinates ~32.42-32.45N, 81.78-81.79W). Recorded as decimal degrees; original corrupted string preserved here per SCHEMA.md coordinate-recovery guidance. | NOT DERIVED, left NR: Fish trial duration (days) and Days Plant after transplant — transplant date (2015-08-14) and harvest date (2015-11-19) are both explicitly stated (a ~97-day span is directly computable), but the paper never itself states a duration in days/weeks anywhere in the text; per the no-derivation rule and the precedent set in pantanellaAquaponicsHydroponicsProduction2012 (crop-2 duration left NR despite a stated start date), left NR rather than computed. Crayfish stocking date relative to transplant is not stated either (tanks were set up and crayfish released before seedlings were transplanted, per Methods paragraph order, but no date is given for crayfish release), adding further uncertainty to any fish-side duration figure specifically. | Initial Stock density — ‘Four mature crayfishes were released in each aquaponic tank’ (83 L water) is given, but no crayfish weight (individual or aggregate) is stated anywhere, so no kg/m3 figure is computable even by derivation. | Total Feed (kg) — weekly per-tank feed rate is stated (3 g algal wafers + 1.5 g Spirulina flakes = 4.5 g/tank/week), and duration is approximately known from the transplant/harvest dates above, but since duration itself is not stated by the paper (see above) and total feed across the trial is never itself stated as a figure, left NR rather than back-calculated. | FCR, SGR, Fish size initial, Fish size final, Fish weight gain, Fish biomass created (kg), Fish survival rate — the crayfish are never weighed, tracked for growth, or assessed for survival anywhere in this paper; they are treated purely as a nutrient source for the basil crop (mirrors the fish-as-nutrient-source pattern noted in pantanellaAquaponicsHydroponicsProduction2012 and monroeEffectsCrayfishAquaponics2015, same lab/system). | Protein (feed) — two distinct feed products with two different crude-protein percentages (algal wafers 30% CP, Spirulina flakes 55% CP, p.183) were fed together with no single blended composition figure stated; recording either percentage alone in the single Protein cell would misrepresent a two-product diet, so left NR with both percentages captured in Feed regime instead. | PDF-QUALITY / [unclear]: container dimensions, p.182. Methods states (as extracted): ‘Dark polyethylene containers (16.88 18.75 27cm3 – h w d) of 95L (25Gal) capacity were used as study tanks.’ The three given figures (16.88, 18.75, 27 cm) do not reconcile to the stated 95 L capacity under a simple length x width x height product (16.88x18.75x27=8,551 cm3=8.6 L, not 95 L), nor under an inches misprint reading (~140 L). The lid dimensions given two sentences later (‘18.75cm x 27cm’, surface area 506.25 cm2) are internally consistent with two of the three h/w/d figures, suggesting the multiplication signs and/or a units qualifier were lost in the source PDF’s text layer around the h/w/d parenthetical specifically, rather than the authors having mis-stated the tank capacity. Recorded the container capacity/fill volume as literally stated (95 L / 25 Gal nominal capacity, 83 L / 22 Gal tap-water fill) in Water volume in the system; the un-reconciled h/w/d figures are preserved verbatim in Media Details rather than corrected or guessed. | SE, NOT SD, THROUGHOUT: every dispersion value in this paper (Tables 1-2, Figures 1, 3-6, and all in-text water/temperature figures) is explicitly stated by the authors to be standard error: ‘The numbers presented with ± are standard error (SE) values’ (footnote, p.183) and ‘±SE represents standard error’ (every figure caption, p.184). All ’±’ values in this row and in the companion plant.csv rows are SE, not SD — flagged here so downstream analysis does not mistake them for SD. | Table 1/Table 2 significance, p.185. Neither table carries any letter superscript on any value; the shared caption states ‘Means that are not followed by different letters were not significantly different at P<0.05 level,’ and Results text independently confirms both: ‘None of the macronutrients differ between aquaponics and hydroponics’ and ‘None of the micronutrient contents were shown to be different between aquaponic and hydroponic basil’ (p.185). All 12 leaf macro/micronutrients (N,P,K,Mg,Ca,S,B,Cu,Fe,Mn,Na,Zn) are therefore ns; carried into plant.csv Significance column as ‘ns’ throughout. | Water-chemistry measurement window, p.183. The TAN/NH4-N, NO2-N and NO3-N cells reflect the paper’s own stated mean ‘during last 8 weeks before harvest,’ not necessarily the full ~14-week study period (weekly monitoring is stated to have run the whole trial, but only this final-8-week summary mean is given in text); recorded as the paper’s own trial-mean regardless, since no full-period mean is stated anywhere. | Aq pH / HYD pH, p.183. Paper states both: ‘The mean water pH in aquaponic (6.9±0.24) and hydroponic (7±0.15) systems were not significantly different.’ AP value (6.9±0.24 SE) recorded in Aq pH per the vault convention that this column is aquaponic-loop-only (established in pantanellaAquaponicsHydroponicsProduction2012/levizouCircularTriTrophicSystem2025); HYD value (7±0.15 SE) has no dedicated column and is given here for reference. | NO COLUMN items: HYD-side mean water pH, 7±0.15 (SE), p.183 (see note above). HYD-side water-nitrogen species, last-8-weeks-before-harvest mean, p.183: TAN/NH4-N 0.03±0.01 ppm; NO2-N 0.3±0.3 ppm; NO3-N 80±12.5 ppm (same NO3-vs-NO3-N basis caveat as the AP value above). AquaSafe dechlorinator applied once to all tanks at study start, 10 mL/37.85L(10Gal) rate, 22 mL/tank (p.182-183) — captured qualitatively under Water type. | [not reported] fields, grouped: Fish — Fish Category (beyond the paper’s own ‘commercial aquatic species’ phrase, used instead), Initial Stock density, FCR, SGR, feed N/P/K %, % of body weight, Fish size initial/final, Total Feed (kg), Fish biomass created (kg), Fish survival rate, Fish weight gain, Fish trial duration (days) — see NOT DERIVED block above for why each is NR rather than computed from partial inputs. | Water — Water classification, Daily Water exchange rate, pHOptimal, FUE AP, FUE HYD, WUE, Dissolved Oxygen, EC — none of these are measured or stated anywhere in the paper (the system is non-circulating with no recycle loop, so several of these concepts do not clearly apply; see Water recycle = NA below). | Plant — Days Plant after transplant (see NOT DERIVED above), Leaf count (only height, fresh weight, dry weight and SPAD were measured; no leaf count anywhere), Tissue nitrate AP/HYD (leaf tissue was analyzed for N/P/K/Mg/Ca/S and B/Cu/Fe/Mn/Na/Zn, never for nitrate itself as a distinct tissue analyte). | Site — Lat/Long required a corrected reading, see UNIT CONVERSION ONLY above (not itself NR, but flagged as recovered-from-corruption). | Water recycle = NA (not NR): Methods explicitly describes the design as a ‘non-circulating floating raft system’ (p.183) — i.e. the paper states there is no recycle/flow loop at all, rather than staying silent about one, so NA (not applicable to this design) rather than NR (schema’s N/NR distinction extended here to the recycle-rate field by the same logic). | Type classification: recorded as experiment. Methods explicitly states ‘the experiment was designed as randomized block design’ with treatments ‘randomly assigned’ to plots within each of 4 locations/blocks (p.183), true replication (4 independent tanks/treatment), and a formal statistical model (GLM/ANOVA, SAS 9.3, alpha=0.05) — meeting SCHEMA.md Part 2 decision rule 2 cleanly (explicit ‘randomized’ language, unlike the judgment call required for pantanellaAquaponicsHydroponicsProduction2012). | Tags judgment call: Meta/Fish/Crayfish (reusing the existing facet from monroeEffectsCrayfishAquaponics2015, an overlapping-author companion poster from the same Georgia Southern lab/system) and Meta/Plant/Basil (reusing the existing facet from mourantianBasilFunctionalGrowth2023 and others). Meta/Region/North-America per Georgia, USA. | Author wikilink judgment call: reused the vault’s existing initialed spellings S. Saha, A. Monroe, M. Day, established in monroeEffectsCrayfishAquaponics2015 (same three authors, overlapping poster/paper from the same lab), in preference to this paper’s own fuller byline form ‘Martin R. Day’, per CLAUDE.md’s instruction to check for an existing spelling first.
Plant Measurements
| Trial | System | Category | Analyte | Value | Unit | Sig. | Location |
|---|---|---|---|---|---|---|---|
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Nitrogen (N) | 5.45 ± 0.08 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Nitrogen (N) | 5.21 ± 0.11 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Phosphorus (P) | 1.71 ± 0.05 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Phosphorus (P) | 1.63 ± 0.05 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Potassium (K) | 0.78 ± 0.02 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Potassium (K) | 0.69 ± 0.02 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Magnesium (Mg) | 0.54 ± 0.01 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Magnesium (Mg) | 0.45 ± 0.02 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Calcium (Ca) | 2.93 ± 0.11 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Calcium (Ca) | 2.92 ± 0.13 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Sulfur (S) | 0.32 ± 0.01 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Sulfur (S) | 0.3 ± 0.01 | % DW | ns | Table 1, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Boron (B) | 42.5 ± 1.7 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Boron (B) | 37.6 ± 1.6 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Copper (Cu) | 14.1 ± 0.86 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Copper (Cu) | 15.9 ± 0.74 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Iron (Fe) | 96.1 ± 4.1 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Iron (Fe) | 99.1 ± 3.8 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Manganese (Mn) | 100.3 ± 10.5 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Manganese (Mn) | 92.7 ± 3.6 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Sodium (Na) | 89.7 ± 13.5 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Sodium (Na) | 86.0 ± 8.2 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | mineral | Zinc (Zn) | 62.2 ± 5.3 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | HYD | mineral | Zinc (Zn) | 65.4 ± 2.2 | mg/kg DW | ns | Table 2, p.185 |
| sahaGrowthYieldPlant2016-T1 | AP | biochemistry | Chlorophyll (SPAD) | 29.3 ± 0.3 | SPAD units | ns | Fig. 5, p.184 |
| sahaGrowthYieldPlant2016-T1 | HYD | biochemistry | Chlorophyll (SPAD) | 28.7 ± 0.3 | SPAD units | ns | Fig. 5, p.184 |