Growth and Tissue Elemental Composition Response of Butterhead Lettuce (Lactuca sativa, cv. Flandria) to Hydroponic and Aquaponic Conditions
Metadata
- Cite key: andersonGrowthTissueElemental2017
- Item type: Journal Article
- Authors: T.S. Anderson, D. de Villiers, M.B. Timmons
- Affiliation: Department of Biological and Environmental Engineering, Cornell University, Ithaca, NY (Anderson, Timmons); Department of Horticulture, Cornell University, Ithaca, NY (de Villiers)
- Journal: Horticulturae 3 (2017) 43
- Date: 07/2017 (Received 29 April 2017; Accepted 19 July 2017; Published 26 July 2017)
- Date added: 2026-07-13
- DOI: 10.3390/horticulturae3030043
- Funding: Cornell University Agricultural Experiment Station federal formula funds, Project No. 1237650 and NYC-123421, received from Cooperative State Research, Education, and Extension Service, U.S. Department of Agriculture
- URL: https://doi.org/10.3390/horticulturae3030043
- PDF:
Anderson et al. - 2017 - Growth and Tissue Elemental Composition Response of Butterhead Lettuce (Lactuca sativa, cv. Flandria.pdf
Opinion
A clean, well-instrumented three-arm greenhouse comparison (H5/H7/A7) from a strong RAS-engineering lab, with genuinely useful null results (A7 statistically indistinguishable from best-practice H5 hydroponics across every biomass metric) and a solid ICP-AES tissue panel. The design has one built-in wrinkle the authors are upfront about: the two aquaponic tubs share continuously mixed water, so they are “not completely independent replicates” (p.5) and the statistical model made no adjustment for this — a mild pseudoreplication concern that the authors disclose rather than hide. The paper is also unusually explicit about what it doesn’t measure: no fish growth, survival, or feed-conversion data at all (koi are treated purely as a nutrient source, maintained at a fixed biomass ceiling), and no per-area or per-system yield figure is ever given, only per-head weights. Internal consistency is good — I found only minor, non-blocking numerical slips (an abstract that undercounts differing microelements, a small EC-range/table mismatch, an inconsistent element symbol across tables) and no misattribution. Worth citing for the pH-isolation logic (comparing A7 to both an equivalent-pH and an equivalent-source-chemistry hydroponic control) and for the tissue-element panel, but fish-side data seekers will find nothing here.
Abstract
The primary objective of this research was to compare lettuce performance under conventional hydroponics at pH 5.8 (referred to as H5), hydroponics at pH 7.0 (referred to as H7), and recirculated aquaponic water at pH 7.0 (referred to as A7). Aquaponic nutrients were supplied by continuously recirculating water between a fish rearing system (recirculating aquaculture system or RAS) and the lettuce growing system (with the sole addition being chelated iron). This paper builds upon our previous research where we found that H7 produced 26% less shoot fresh weight (FW) growth than H5 and an 18% reduction in dry weight (DW). In this research, we also evaluated the inorganic hydroponics nutrient solution at pH 7.0 (H7) to provide continuity between experiments and to isolate the pH effect. The A7 plant biomass responses were not different from H5 in all biomass response categories. H7 was different from H5 in shoot FW, DW, and DW/FW, as well as root FW and DW. H7 was different from the A7 in shoot FW, DW/FW, and root DW. There were no tissue elemental differences between H5 and H7 except Cu. The Ca and Na contents differed between H5 and A7, while the microelements Mn, Mo, and Zn differed. Generally, the elemental tissue differences between treatments were proportional to the differences for the same elements in the nutrient solutions. Aquaponic systems are often viewed to be more complicated and more risky because two complex systems are being joined (hydroponics plus RAS). However, the aquaponics system proved to be surprisingly simple to manage in daily operations. Our data suggested that the aquaponics system (A7), which was operated at a higher pH 7.0, was able to offset any negative biomass and elemental effects that occurred in the inorganic hydroponic pH 7.0 treatment (H7) from its increased pH and less optimized nutrient solution elemental concentrations.
Summary
Cornell researchers grew butterhead lettuce (cv. Flandria) in a glass greenhouse under three parallel treatments across three sequential trials: conventional hydroponics at pH 5.8 (H5), the same hydroponic nutrient solution but held at pH 7.0 (H7), and a continuously recirculating aquaponic loop fed by a koi RAS also held at pH 7.0 (A7, supplemented only with chelated iron). The H7 arm exists specifically to isolate whether any A7 effect is due to pH alone versus the nutrient source. Aquaponic lettuce (A7) matched the best-practice hydroponic control (H5) on every biomass measure (fresh weight, dry weight, DW/FW ratio, leaf count, leaf surface area), while the elevated-pH hydroponic treatment (H7) grew smaller shoots but larger roots than both. Tissue elemental analysis (ICP-AES) found only a handful of significant differences — chiefly Ca, Na, Mn, Mo, Zn, Ba and Sr differing between H5 and A7 — and the authors argue these track proportionally with differences already present in the nutrient solutions themselves, staying within normal lettuce tissue ranges. A calcium-phosphate precipitate formed uniquely in the H7 tubs, consistent with the higher pH there. The paper’s headline claim is that whatever chemistry an RAS contributes (mineralized organics, chelators, microflora) evidently offset the otherwise-negative effect of running hydroponics at pH 7.0, and that operating the combined system was in practice straightforward.
Experiment data
- Location: Ithaca, NY, USA (42.4489, -76.4690), glass greenhouse
- Design: Three parallel treatments (H5, H7, A7) across three sequential trials using six identical grow tubs; 3 tubs/trial to H5, 1 tub/trial to H7 (rotated among remaining positions), 2 fixed tubs/trial to A7 (tubs 4 and 5, continuously water-mixed with each other — explicitly not fully independent replicates, p.5)
- Replicates / n: A7 = 2 tubs/trial x 3 trials = 6 tub-instances (not fully independent); H5 = 3 tubs/trial x 3 trials = 9 tub-instances; individual-plant sample sizes given per response variable (e.g. A7 shoot FW N=144, Table 6)
- Duration: Trial 1 = 43 days seed-to-harvest; Trials 2-3 = 35 days seed-to-harvest (target ~150 g head weight); 12-day pre-transplant seedling stage common to all trials
- Organisms: Koi / Butterhead lettuce (Lactuca sativa, cv. Flandria)
- Statistics: Mixed-effects least-squares models (JMP Pro 11), treatment and trial as fixed effects, tub nested within trial as random effect; Tukey HSD, alpha = 0.05
- Plant fresh weight: H5 178 +/- 2 g, H7 137 +/- 3 g, A7 169 +/- 2 g (head FW, Table 6) — A7 not different from H5; H7 different from both
- Tissue elemental composition: A7 vs H5 tissue Ca ~11% lower, Na ~3x higher (2027 vs 720 mg/kg), Mn/Mo/Zn/Ba/Sr(St) also differ (Table 10, p.13)
Biomass response
This paper: The central finding is a null result in the useful direction: A7 (aquaponic, pH 7.0) was not statistically different from H5 (hydroponic, pH 5.8) in any biomass response — shoot FW/DW/DW-FW, root FW/DW/DW-FW, rootball FW/DW/DW-FW, leaf count, leaf surface area, or leaf SA/head-FW (p.11, Tables 6-9). Meanwhile H7 (hydroponic, pH 7.0 — same nutrient chemistry as H5 but elevated pH) produced 23% less shoot FW than H5, consistent with the authors’ own earlier study [ref 14] which found a 26% reduction under the same pH manipulation. H7 also grew significantly larger roots (FW +22%, DW +31-33%) than H5, interpreted as a nutrient-deficiency-driven root-foraging response (p.14). A calcium-phosphate precipitate (Ca:P molar ratio 1.39:1, SD 0.06) formed only in the H7 tubs, tied to daily KOH dosing (Section 4.3, p.16-17).
Compared with:
- todo Anderson et al. 2017 (Horticulturae 3,41) — the authors’ own directly preceding study, hydroponic-only H5 vs H7 pH comparison; this paper explicitly builds on and reproduces its 26%→23% shoot-FW-reduction finding. (p.1, ref 14)
- todo Seawright et al. 1998 — reported continual Ca-phosphate precipitation in an integrated aquaculture-hydroponics system at pH 7.5, cited as a parallel to this paper’s H7 precipitate finding. (p.16, ref 22)
Tissue elemental composition
This paper: Among macronutrients (N, P, K, Ca, Mg, S), only Ca differed, and only between H5 and A7 (H5 1.37%, A7 1.22%, ~11% lower), proportional to the lower Ca in the A7 nutrient solution (p.13-14). Among micronutrients/other elements, the paper’s own Results/Discussion text (p.13) and Table 10 report five elements differing significantly between H5 and A7: Ba, Mn, Mo, Sr (“St” in the table), and Zn — the abstract’s list of “Mn, Mo, and Zn” undercounts this (see Extraction notes). Na differed strongly (H5 720, A7 2027 mg/kg), attributed to the ~10x higher Na in the A7 nutrient solution itself, which in turn is attributed to fish feed and RAS water recycling/evaporation (p.15) — flagged by the authors as a caution point for growers managing Na in fish diets. Cu was the only tissue element differing between the two hydroponic treatments (H5 vs H7), despite equal or lower Cu in the H7/A7 solutions, suggesting a pH effect on Cu uptake or chelator-mediated transport (p.14-15). Fe, B, and Si tissue concentrations were all indistinguishable across treatments despite order-of-magnitude differences in solution concentrations for B and Si, which the authors attribute to passive-uptake equilibrium (B) or possible rockwool-fiber Si contribution (p.15-16).
Compared with:
- todo Resh 2013 — reports 4.5% N in lettuce tissue, lower than this paper’s 5.68-5.84% across all three treatments. (p.14, ref 1)
- todo Pantanella et al. 2012 — reports 2.9% tissue N, well below this paper’s values; authors caution against assuming cross-system applicability of literature N values. (p.14, ref 23)
- todo Liedl et al. 2004 — cited for a proportional Mo tissue-to-solution response pattern, matching this paper’s own Mo finding. (p.13, ref 21)
- todo Cheng & Allen 2001 — reports lower pH should increase lettuce tissue Cu, opposite direction to this paper’s H7/A7 (higher pH, higher Cu) finding — noted by the authors themselves as a discrepancy worth investigating. (p.14-15, ref 24)
Linked claims
- Aquaponic nutrient solutions can match conventional hydroponic biomass yield despite unbalanced nutrient concentrations
- Elevating hydroponic solution pH from 5.8 to 7.0 reduces lettuce shoot growth
- Recirculating aquaculture system water can offset the negative effect of pH 7.0 on lettuce compared to inorganic pH-7.0 hydroponics
- Sodium can accumulate in aquaponic nutrient solutions via fish feed and water recycling
Citations to chase
- todo Anderson, Martini, de Villiers & Timmons (2017), Horticulturae 3,41 — the directly preceding companion study (H5 vs H7 only, no aquaponics arm) that this paper builds on; not yet in this vault.
- todo Seawright, Stickney & Walker (1998) — nutrient dynamics in an integrated aquaculture-hydroponics system, including Ca-phosphate precipitation at pH 7.5.
- todo Pantanella, Cardarelli, Colla, Rea & Marcucci (2012 ISHS proceedings) — aquaponics vs. hydroponics lettuce production and quality comparison.
- todo Love, Jillian, Fry, Li, Hill, Genello, Semmens & Thompson (2015) — commercial aquaponics production/profitability survey, cited for industry context.
Extraction notes
Severity tally: 0 BLOCK, 0 MATERIAL, 4 MINOR → quality: ok (0 BLOCK and <=2 MATERIAL). No misattribution found.
- ⚠️MINOR — EC range mismatch. Section 3.1 (p.10-11) states EC “did not change significantly during the experiment (1300-1500 µS/cm)” as a whole-system summary, but Table 5 (p.11) shows A7 tub means as low as 1161-1166 µS/cm (Trial 1) and starting values as low as 1060-1070 µS/cm, below the stated 1300 floor. Table 5’s own A7-specific values were used directly for the trials.csv EC cell (range 1.16-1.49 dS/m); the prose range appears to fit H5/H7 better than A7. Does not block extraction.
- ⚠️MINOR — Root DW% for H7. Discussion (p.14) states H7 root response “increase of 22% and 33% FW and DW”; Table 7’s own % column gives DW = 131% (a 31% increase), not 33%. ~2-point rounding discrepancy. No dedicated trials.csv column exists for root data regardless, so no cell is affected.
- ⚠️MINOR — Abstract undercounts differing microelements. Abstract lists only “Mn, Mo, and Zn” as differing between H5 and A7; Discussion text (p.13) and Table 10 itself both show five: Ba, Mn, Mo, Sr(“St”), Zn. Table 10’s letter groupings (self-consistent) were used directly for every plant.csv Significance value, so the abstract’s undercount does not affect any recorded cell.
- ⚠️MINOR — Inconsistent element symbol. Table 1 (feed analysis, p.6) labels an element “Sr” (Strontium); Table 2 (nutrient solution, p.7) and Table 10 (tissue, p.13) both instead use “St” for the apparently same element. Recorded verbatim as “St” in plant.csv with a note; no numeric value is in conflict, only the symbol.
- Recomputed check (not a contradiction, remarks only): stated “10 kg/m3 density” ceiling (p.7) doesn’t divide out exactly from either the 1.2 m3 fish-tank volume (→8.3 kg/m3) or the 2.2 m3 stated system volume (→4.5 kg/m3); recorded as the paper’s own stated figure since only one value is ever given.
[not reported]: Fish scientific name (paper says “koi” throughout, never a Latin binomial); Fish Category; Water type; Water classification; Daily Water exchange rate; pHOptimal; FUE AP/HYD; WUE; Dissolved Oxygen (air stones used everywhere but no DO value given); NO2-N (never analyzed or reported); Plant Category; SPAD; Plant height; Average room Temperature (only per-trial day/night means given, Table 4, no combined figure); Fish trial duration, Total Feed, Fish size final, Fish survival rate, Fish weight gain, FCR, SGR (no fish-growth data anywhere — koi treated purely as a nutrient source, held at a fixed biomass ceiling); Tissue nitrate AP/HYD (the ICP-AES tissue panel measures total N, not nitrate ion — nitrate is only tracked in the water/nutrient solution, Table 2); Days Plant after transplant (component figures — 12-day seedling stage, 43/35-day seed-to-harvest durations — are both given, but subtracting them would be derivation, so left NR); AP/HYD yield columns (paper reports only per-head weights, never a per-m2 or per-system yield figure).
[unclear]: The exact calendar year of the three plant trials is never stated; the koi stocking date is given as “February [year not stated]” (p.7), leaving the gap between fish stocking and the plant trials unclear — left NR rather than assumed.
No column home (kept in trials.csv Experimental Remarks only): Root FW/DW (Table 7, all three treatments); rootball FW/DW/DW-FW (Table 8, Trials 1 & 3 only); leaf surface area and leaf-SA/head-FW ratio (Table 9, Trial 3 only); full nutrient-solution start/end panel (Table 2) beyond TAN/NO3-N/Fe/EC/pH; H7 precipitate digestion chemistry (Ca:P molar ratio, % composition).
Water-quality panel note: Table 2’s full nutrient-solution element panel (K, Ca, P, Mg, S, and 12 micro/other elements, each with start/end values for all three treatments) is a genuinely valuable dataset but reduces only partially to trial means in trials.csv’s existing water-chemistry columns (TAN, NO3-N, EC, pH); the remainder has no column and is preserved in Experimental Remarks per Step 7’s water-quality-too-valuable-to-discard provision, rather than being misrouted into plant.csv (it is water chemistry, not plant tissue).
New tags introduced: none — Meta/Type/Experiment, Meta/Region/North-America, Meta/Fish/Koi, and Meta/Plant/Lettuce all reused exactly as spelled in existing vault notes (abbeyBasilOcimumBasilicum2022 for Koi/North-America; singhAquaponicProductionOrnamental2024/tadesseComprehensiveComparisonLettuce2023 for Lettuce).
Source: Anderson et al. - 2017 - Growth and Tissue Elemental Composition Response of Butterhead Lettuce (Lactuca sativa, cv. Flandria.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
andersonGrowthTissueElemental2017-T1
Fish
| Field | Value |
|---|---|
| Fish | Koi (common name only; scientific name not given in this paper) |
| Initial Stock density | 10 (stated as ‘10kg/m3 density’ for a <=10kg culled system biomass ceiling, p.7; see WARN-MINOR in remarks - does not cleanly divide out of either stated volume) |
| Protein | 38 (min crude protein, guaranteed analysis label, Blackwater Creek Farms Max Growth Diet, p.6) |
| N | 6.3 |
| P | 2.2 |
| K | 1.2 |
| % of body weight | 1 (~1% of system biomass/day; 90g/day in 2 feedings on weekdays, 60g/day in 1 feeding on weekends, p.7) |
| Fish size initial | 1 (1g fingerlings, stocked February [year not stated], p.7) |
| Feed routine | Two feedings/day on weekdays, one feeding/day on weekends |
| Feed regime | Commercial floating extruded pellet ‘Max Growth Diet’ (Blackwater Creek Farms); guaranteed analysis min 38% crude protein, min 8% crude fat, max 4% crude fiber, min 1% phosphorus, max 10% moisture (p.6) |
Water
| Field | Value |
|---|---|
| Water recycle | 10 (~10 L/min diverted from bead-filter return flow to the two A7 tubs, p.5,8) |
| Water volume in the system | 2200 (stated as the aquaculture-tank system volume ‘that included the two A7 tubs’, p.7) |
| Aq pH | 6.95 +/- 0.40 (Table 3, tub means/SD) |
| EC | 1.16-1.49 (range across A7 tub/trial means, Table 5; no single overall trial mean given - see WARN-MINOR) |
| Water temperature | 24.8 +/- 1.18 (reported as Root Zone Temperature (RZT) in Table 3, used as water temperature for this floating-raft system) |
| TAN / NH4-N | 0.3-0.6 (A7 start/end values, Table 2; no trial mean reported) |
| NO3-N | 112-122 (A7 start/end values, Table 2; no trial mean reported) |
Plant
| Field | Value |
|---|---|
| Plant | Butterhead lettuce (Lactuca sativa, cv. Flandria) |
| Details | Pelleted seed grown in individual rockwool plugs (Grodan AO25/40, 25mm); 12-day ebb-and-flood seedling stage (H5 solution for all seedlings) then transplanted; grown to an estimated 150g target head fresh weight (Trial 1: 43 days seed-to-harvest; Trials 2-3: 35 days seed-to-harvest), p.7 |
| Plants/m2 | 30 (stated explicitly, p.5) |
| Leaf count | 37.2 (A7, Table 9; Trial 3 only, n=5 tubs; no SD reported in that table) |
| Plant fresh weight | 169 +/- 2 (A7 shoot/head FW, Table 6, N=144) |
| Plant dry matter | 4.0% (converted from A7 DW/FW ratio 0.040 +/- 0.0008 g/g, Table 6; see UNIT CONVERSION ONLY in remarks) |
System & Setup
| Field | Value |
|---|---|
| System type | Deep water culture / floating raft (paper’s own terms ‘deep water hydroponics’ p.3, ‘floating rafts’ p.5) |
| Media Details | HDPE growing tubs 1.82m x 0.91m x 0.3m (0.425 m3 each); rigid Styrofoam rafts 25mm thick with 25mm-dia holes at 200mm spacing; rockwool plugs (Grodan AO25/40) for seedlings; 50 plants/tub (5 rows x 10) at 30 plants/m2 |
| Biological system already in use | Y (Koi stocked as 1g fingerlings in February [year not stated] onto an established RAS; A7 tubs filled with carbon-filtered water and recirculated with the RAS at ~10L/min for one week before Trial 1 transplant to reach equilibrium (p.8)) |
| Air supplement | Y (All six grow tubs (H5, H7, A7 alike) fitted with 2 coarse air stones each (Sweetwater AS-2, 0.1 CFM/stone) for mixing/DO; fish tank also used air stones for oxygenation (p.5)) |
| Iron supplemented | Y (Chelated iron (Sprint 330, Fe-DTPA) added to A7 at 2mg/L equivalent elemental Fe at the start of each trial (the sole nutrient addition to A7 besides K2CO3 for pH); all three treatments (H5,H7,A7) received Fe-DTPA at trial start (p.8)) |
| pH Buffers | Y (1M K2CO3 used for daily A7 pH adjustment to 7.0 (no acid required); chosen because roughly half its ions raise pH while the remainder adds alkalinity, reducing pH swings from CO2/feeding/nitrification (p.8)) |
| Climate control | Y (Argus Control System regulated day/night greenhouse air temperature setpoints (day heat 24degC/cool 25degC; night heat 19degC/cool 20degC; 2.5h ramp) and daily light integral; evaporative pads and vent fans for cooling, water-to-air heat exchangers for heating (p.3,5)) |
| Artificial Lighting | Y (20x 400W high-pressure sodium lights supplementing natural light to a consistent daily light integral (DLI ~14.2-14.4 mol/m2/day across trials, Table 4), p.3) |
| Nutrient supplemented | Y (H5/H7: full inorganic modified half-strength Sonneveld & Straver nutrient solution (Table 2); A7: RAS water only, supplemented solely with chelated Fe-DTPA (2mg/L per trial) plus K2CO3 for pH (p.7-8)) |
| Equipment | Argus monitoring/control system; 20x 400W HPS lights; LI-COR LI-190R quantum sensor; per-tub recirculating pumps (24 Lpm, 18min HRT); Sweetwater AS-2 air stones; commercial bead filter (Aquaculture Systems Technology); activated-charcoal filter (50kg ProLine AC55 media); pin-style EC tester (Oakton EC Testr11+); automated ICP-AES with Vulcan 84 digestion unit; LI-3100C leaf area meter; 70degC drying ovens |
| Control Parameters | pH (5.8 target H5 via HNO3/KOH; 7.0 target H7 via HNO3/KOH; 7.0 target A7 via K2CO3 only); greenhouse day/night air temperature and DLI (Argus system); EC monitored but not controlled |
| Combination | Koi and butterhead lettuce cv. Flandria; three-arm comparison of hydroponic pH5.8 (H5), hydroponic pH7.0 (H7), and aquaponic pH7.0 (A7), replicated across three sequential trials pooled in one mixed-model analysis |
Site
| Field | Value |
|---|---|
| Region | North America |
| Country | USA |
| Lat | 42.4489 |
| Long | -76.4690 |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | g (head/root FW,DW per plant); g/g (DW/FW ratio); mg/kg dry wt, % dry wt (tissue elemental, Table 10); cm2 (leaf SA); count (leaf count) |
| Statistic Details | Mixed-effects least-squares models, JMP Pro 11; treatment and trial as fixed effects, tub nested within trial as random effect; Tukey HSD post-hoc, alpha=0.05 (p.10) |
| Statistically analysed | Y |
| Replicates (n) | 2 tubs/trial x 3 trials = 6 tub-instances for A7 (tubs 4,5); explicitly stated as not fully independent replicates due to continuous water mixing between the two tubs (p.5); H5 = 3 tubs/trial x 3 trials = 9 tub-instances. Individual-plant sample sizes per response variable given directly in Tables 6-9 (e.g. A7 shoot FW N=144) |
Experimental Remarks: TRIAL DEFINITION: T1 = the sole aquaponic treatment (A7, pH 7.0), continuously recirculated between the koi RAS and two of six grow tubs (tubs 4 and 5, connected by a passive equalizing pipe; paper states p.5 ‘data from these two tubs are not completely independent replicates’), pooled across three sequential trials (30 Jun-12 Aug; 14 Aug-18 Sep; 20 Oct-24 Nov; no year stated for these three date ranges). Paired control = conventional hydroponic pH 5.8 (H5), recorded in the HYD columns, per the paper’s own stated primary objective (‘our primary objective was to compare A7 to H5’, p.8). A second treatment, hydroponic pH 7.0 (H7), was run in parallel as a secondary reference to isolate the pH effect from nutrient source, but H7 is NOT an aquaponic treatment, so per SCHEMA.md (‘one row per aquaponic treatment’) it gets no row of its own; its values are given below rather than in the HYD columns. The paper pools all three trials into one mixed-effects analysis (treatment and trial as fixed effects, tub-within-trial as random effect, p.10); this single row represents that pooled dataset, not any one individual trial. H7 secondary-treatment values (NOT used in any AP/HYD cell above): Shoot Table 6 - FW 137+/-3g (77% of H5), DW 6.3+/-0.11g (90%), DW/FW 0.044+/-0.0006 (114%); Root Table 7 - FW 9.6+/-0.30g (122%), DW 0.42+/-0.016g (131%), DW/FW 0.045+/-0.0006 (111%); Rootball Table 8 (Trials 1,3 only) - FW 3.2g (105%), DW 0.21g (103%), DW/FW 0.067 (99%); Leaf Table 9 (Trial 3 only) - count 38.4 (97%), leaf SA 2731cm2 (100%), leaf SA/head FW 17.0 cm2/g (117%); tissue elements per Table 10 (full column, p.13); pH 6.96+/-0.22, RZT 25.6+/-1.44 degC (Table 3); EC tub means by trial (Table 5) 1430/1485/1382 microS/cm (T1/T2/T3); nutrient solution start/end per Table 2. H7 tubs also showed a granular Ca-P precipitate unique to that treatment (Section 4.3, Fig.5): 97-98% of precipitate mass as Ca+P (as PO4 3-), Ca:P molar ratio 1.39:1 (SD 0.06), attributed to daily KOH pH adjustment; not observed in H5 or A7. NO COLUMN (no dedicated trials.csv column exists for root or leaf-surface-area data): Root FW/DW Table 7 - H5 7.8+/-0.11g / 0.32+/-0.006g; H7 9.6+/-0.30g / 0.42+/-0.016g; A7 8.2+/-0.16g / 0.32+/-0.011g. Rootball (rockwool-contained root, mostly taproot) Table 8, Trials 1+3 only - H5 FW3.0/DW0.21/DW-FW0.068; H7 FW3.2/DW0.21/DW-FW0.067; A7 FW2.7/DW0.18/DW-FW0.070. Leaf surface area (SA) and leaf SA/head-FW ratio, Table 9, Trial 3 only - H5 2721cm2 / 14.5 cm2 g-1; H7 2731cm2 / 17.0; A7 2363cm2 / 14.1. Average head FW for the leaf-SA subsample (Trial 3, smaller n) differed from the full-dataset Table 6 means: 188/161/168g (H5/H7/A7, leaf-SA subsample) vs 178/137/169g (Table 6, full dataset) - paper itself attributes this to different sample subsets (p.11), not a contradiction. Nutrient-solution start/end concentrations (Table 2) for K, Ca, P, Mg, S and micronutrients B, Cu, Zn, Mo and other elements Si, Pb, Al, St(Sr), Ba - no water-nutrient-panel column exists beyond TAN/NO2-N/NO3-N/EC/pH; A7 nutrient solution Na was 61 mg/L (start and end) vs 6-7 mg/L in H5/H7, attributed in Discussion (p.15) to fish feed Na and RAS water recycling - this is the stated explanation for the large A7 tissue Na difference recorded in plant.csv. Precipitate digestion chemistry (H7 only, water/system finding, not plant tissue) given above. NOT DERIVED, left NR: Days Plant after transplant (paper states seed-to-harvest duration only - Trial 1 = 43 days, Trials 2-3 = 35 days, target ~150g head weight - and separately a 12-day pre-transplant seedling/ebb-and-flood phase common to all trials and treatments (p.7); subtracting, e.g. 43-12=31, would be derivation, so left NR). FCR, SGR, Fish size final, Fish biomass created (kg), Fish survival rate, Fish weight gain, Total Feed (kg), Fish trial duration (days) - the koi were stocked once as 1g fingerlings in February [year not stated for the 2017-published trials, raising an unresolved timeline gap - see below] and maintained continuously at a culled ceiling of <=10kg system biomass across all three plant trials; no endpoint fish weight, growth, survival, or feed total tied to a defined fish ‘trial’ window is given anywhere in the paper. AP / HYD (per-m2 or per-system yield) - the paper reports only per-head (per-plant) FW/DW (used in Plant fresh weight / Plant dry matter columns above) and never a per-area or per-system total yield figure; multiplying per-head weight by the stated planting density (30 plants/m2) to get a yield/m2 would be derivation. WARN-MINOR EC: Section 3.1 text (p.10-11) states EC ‘did not change significantly during the experiment (1300-1500 microS/cm)’ as a whole-system summary, but Table 5 (p.11) shows A7 tub means as low as 1161-1166 microS/cm (Trial 1) and A7 starting values as low as 1060-1070 microS/cm (Trial 1, tubs 4-5), below the stated 1300 floor; A7 Trial 2-3 means (1466-1490) do fall inside the stated range. Table 5’s own A7-specific tub/trial means used directly for the EC cell (range 1.16-1.49 dS/m, this trial’s actual data); the prose summary appears to describe the H5/H7 tubs more accurately than A7. Does not block extraction since Table 5 values, not the prose range, are used. WARN-MINOR Root DW% for H7: Discussion (p.14) states the root response for H7 vs H5 ‘showed an increase of 22% and 33% FW and DW for H7’; Table 7’s own % column gives H7 FW=122% (a 22% increase, matches) but DW=131% (a 31% increase, not 33% as stated in the running text). ~2-point discrepancy, likely a rounding artifact from computing % off raw means vs. SE-weighted values. Does not affect any extracted cell (root data has NO COLUMN home here regardless; Table 7’s 0.42+/-0.016 g DW value, not the ‘33%’ text figure, is the one quoted above). WARN-MINOR abstract vs. body text on differing microelements: Abstract states only ‘the microelements Mn, Mo, and Zn differed’ (between H5 and A7); Discussion Section 3.3 (p.13) and Table 10 itself both show FIVE elements differing between H5 and A7 with distinct A/B letter groups: Ba, Mn, Mo, Sr(‘St’), and Zn. The abstract’s list of three appears to be an incomplete summary; Table 10’s letter groupings (used directly for every plant.csv Significance value) are internally self-consistent and are what this extraction relies on, so the abstract’s undercount does not affect any recorded cell. WARN-MINOR element-symbol inconsistency: Table 1 (feed elemental analysis, p.6) labels one element ‘Sr’ (Strontium, 269 mg/kg); Table 2 (nutrient solution, p.7) and Table 10 (shoot tissue, p.13) both instead label the apparently-same element ‘St’ (e.g. Table 10: H5 94, H7 98, A7 75 mg/kg). Recorded in plant.csv exactly as printed in Table 10 (‘St’), with a note that it is presumed to be the same element Table 1 calls ‘Sr’ (Strontium) based on table position/context; the paper’s own labeling is internally inconsistent across its three tables, not resolved here since no numeric value is actually in conflict, only the two-letter vs. one-letter element symbol. WARN-MINOR Initial Stock density arithmetic: paper states (p.7) fish were ‘continually culled to maintain a system biomass of <=10kg (10kg/m3 density)’. Recomputed check (remarks only, not a cell value): 10 kg / 1.2 m3 (fish tank volume alone) = 8.3 kg/m3; 10 kg / 2.2 m3 (stated ‘system’ volume including the two A7 tubs) = 4.5 kg/m3 - neither denominator reproduces the stated 10 kg/m3 exactly. This looks like the authors’ own rounded/approximate figure rather than two conflicting numbers in the paper (only one density value is ever stated), so it is not elevated to MATERIAL/BLOCK; the paper’s stated ‘10 kg/m3’ is recorded as-is in the Initial Stock density cell. UNIT CONVERSION ONLY: coordinates ‘42 deg 26’ 56.2” N 76 deg 28’ 08.3” W’ (p.3) converted to decimal degrees: 42 + 26/60 + 56.2/3600 = 42.4489 N; 76 + 28/60 + 8.3/3600 = 76.4690, recorded as -76.4690 (West negative per SCHEMA.md). No DMS component exceeds 59, so this is a normal conversion, not a recovered-error case. | UNIT CONVERSION ONLY: Plant dry matter recorded as percent (4.0%), converted directly from the paper’s own DW/FW ratio for A7 in Table 6 (0.040 +/- 0.0008 g/g, i.e. x100); HYD (H5) DW/FW = 0.039 +/- 0.0003 g/g = 3.9%. | UNIT CONVERSION ONLY: TAN and NO3-N cells record the paper’s stated start/end values (Table 2) rather than a computed mean, per SCHEMA.md’s ‘range only, no trial mean reported’ rule - no averaging was performed. [not reported] fields grouped: Fish scientific name (paper calls them ‘koi’ throughout, e.g. p.7, but never gives Cyprinus carpio or any Latin binomial); Fish Category; Water type; Water classification; Daily Water exchange rate; pHOptimal (background pH targets for hydroponics/RAS are discussed in the Introduction, p.2, but never stated as a distinct ‘optimal pH’ finding of this study, separate from the treatment set points already captured in Aq pH); FUE AP; FUE HYD; WUE; Dissolved Oxigen (air stones used to maintain DO in every tub, p.5, but no DO value is ever reported); Water temperature is reported only as ‘Root Zone Temperature’ (RZT) in Table 3 (A7: 24.8 +/- 1.18 degC) - used directly for the Water temperature cell since RZT is the recirculating tub-water temperature in this floating-raft design; NO2-N (never analyzed or reported anywhere in the paper); Plant Category; SPAD; Plant height; Average room Temperature (Table 4 gives day/night air temperature means separately, per trial, with no single combined value; computing one would be derivation); Fish trial duration; Total Feed (kg); Fish size final; Fish survival rate; Fish weight gain; FCR; SGR; Tissue nitrate AP and HYD (the paper’s ICP-AES panel measures total elemental N, not nitrate ion, in tissue - see Table 10; nitrate is only tracked in the nutrient SOLUTION, Table 2, which is water chemistry not tissue). [unclear]: exact calendar year of the three plant trials (30 Jun-12 Aug / 14 Aug-18 Sep / 20 Oct-24 Nov) is never stated in this 2017 paper, while the fish stocking date is explicitly ‘February of 2014’ (p.7) - a 3+ year gap if trials occurred in the same year as the companion papers from this lab (e.g. ref [14], Anderson et al. 2017, Horticulturae 3,41), or the trials could have occurred in 2014 itself with fish stocked shortly before; the paper gives no basis to resolve this, so Fish trial duration and any age-since-stocking figure are left NR rather than guessed.
Plant Measurements
| Trial | System | Category | Analyte | Value | Unit | Sig. | Location |
|---|---|---|---|---|---|---|---|
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Carbon Content | 33.5 ± 0.27 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Carbon Content | 33.5 ± 0.46 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Carbon Content | 33.7 ± 0.32 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | N | 5.82 ± 0.05 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | N | 5.68 ± 0.09 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | N | 5.84 ± 0.06 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | P | 1.03 ± 0.02 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | P | 1.11 ± 0.04 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | P | 1.04 ± 0.02 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | K | 3.79 ± 0.06 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | K | 3.87 ± 0.1 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | K | 3.9 ± 0.07 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Ca | 1.37 ± 0.03 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Ca | 1.33 ± 0.05 | % | A,B | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Ca | 1.22 ± 0.04 | % | B | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Mg | 0.32 ± 0.001 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Mg | 0.36 ± 0.002 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Mg | 0.36 ± 0.001 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | S | 0.25 ± 0.003 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | S | 0.24 ± 0.004 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | S | 0.25 ± 0.003 | % | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Fe | 61 ± 1 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Fe | 60 ± 2 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Fe | 60 ± 1 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Mn | 77 ± 3.2 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Mn | 65 ± 5.6 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Mn | 45 ± 3.9 | mg/kg | B | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | B | 28.0 ± 1.7 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | B | 38.3 ± 2.9 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | B | 30.6 ± 2.1 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Cu | 7.0 ± 0.6 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Cu | 9.9 ± 1.0 | mg/kg | B | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Cu | 10.7 ± 0.7 | mg/kg | B | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Zn | 34 ± 2.2 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Zn | 31 ± 3.9 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Zn | 64 ± 2.7 | mg/kg | B | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Mo | 0.86 ± 0.09 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Mo | 0.83 ± 0.16 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Mo | 0.2 ± 0.12 | mg/kg | B | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Na | 720 ± 128 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Na | 717 ± 222 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Na | 2027 ± 157 | mg/kg | B | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Al | 15 ± 1 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Al | 13 ± 1 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Al | 14 ± 1 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Ni | 0.06 ± 0.03 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Ni | 0.03 ± 0.05 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Ni | 0.09 ± 0.04 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Si | 28.3 ± 0.4 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Si | 27.3 ± 0.7 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Si | 28.2 ± 0.5 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Pb | 1.4 ± 0.5 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Pb | 0.7 ± 0.9 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Pb | 0.8 ± 0.6 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | St (Sr, presumed Strontium) | 94 ± 2 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | St (Sr, presumed Strontium) | 98 ± 3 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | St (Sr, presumed Strontium) | 75 ± 2 | mg/kg | B | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | As | 0.36 ± 0.02 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | As | 0.35 ± 0.03 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | As | 0.35 ± 0.02 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Ba | 1.2 ± 0.1 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Ba | 1.0 ± 0.2 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Ba | 3.0 ± 0.2 | mg/kg | B | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Cd | 0.13 ± 0.0 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Cd | 0.12 ± 0.01 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Cd | 0.13 ± 0.0 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Co | 11 ± 1 | ug/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Co | 10 ± 2 | ug/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Co | 12 ± 1 | ug/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | HYD | mineral | Cr | 0.26 ± 0.03 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | H7 | mineral | Cr | 0.31 ± 0.05 | mg/kg | A | Table 10, p.13 |
| andersonGrowthTissueElemental2017-T1 | AP | mineral | Cr | 0.31 ± 0.04 | mg/kg | A | Table 10, p.13 |