Scaling aquaponic systems: Balancing plant uptake with fish output
Metadata
- Cite key: buzbyScalingAquaponicSystems2014
- Item type: Journal Article
- Authors: K. Buzby, L. Lin
- Affiliation: Department of Civil and Environmental Engineering, West Virginia University, P.O. Box 6103, Morgantown, WV 26506, United States
- Journal: Aquacultural Engineering 63 (2014) 39-44
- Date: 10/2014 (Received 10 December 2013; Accepted 24 September 2014; Available online 14 October 2014 — no separate “published”/issue date stated)
- Date added: 2019-02-02
- DOI: 10.1016/j.aquaeng.2014.09.002
- Funding: US Department of Agriculture Special Grant (2010-343386-21745)
- URL: https://doi.org/10.1016/j.aquaeng.2014.09.002
- PDF:
Buzby and Lin - 2014 - Scaling aquaponic systems Balancing plant uptake .pdf
Opinion
A tidy, honestly-scoped methods demonstration rather than a definitive removal-rate dataset: the authors are explicit that both experiments exist “to explore the capabilities of the method,” and the design matches that framing — one channel per crop/age, no independent replicate channels, no between-group significance test (only regression r2 for the concentration-vs-time curves). That’s a real limitation for anyone wanting to cite these removal rates as generalizable numbers, but it doesn’t undermine the paper’s actual contribution, which is the batch-isolation method itself (stop channel inflow, spike TAN, sample hourly, back out a rate) as a fast alternative to multi-month whole-system trials. The lettuce-vs-nasturtium contrast (nasturtium: both TAN and nitrate removed, 80% DIN; lettuce: TAN only, 48% DIN, no nitrate removal) is a genuinely useful negative result given nitrate removal by lettuce is well documented in warmer aquaponic systems — the authors’ own explanation (low DIN concentration, cold trout-facility water, ammonium preference at low temperature/low N) is plausible and citable. Internally clean: no numeric contradictions found between abstract, body text, and tables. Two definitional gaps are worth a user’s attention: phosphate is measured by a phosphorus-based standard method but reported under a “PO4” label with no clarification of basis, and “plant age” is given as weeks-from-sowing with no stated transplant event.
Abstract
To be effective at nutrient removal, aquaponics systems should be sized correctly to balance nutrient production from fish culture and nutrient uptake by plants. We describe a method where the plant component was isolated from the fish rearing operation so that nutrient removal could be evaluated independently. Two crops, lettuce (Latuca sativa) and nasturtium (Tropaeolum majus), were evaluated. Nasturtium had higher removal rates and removed both total ammonia nitrogen (TAN) and nitrate resulting in 80% DIN removal while lettuce removed only 48%. Lettuce removed only TAN and was ineffective at nitrate removal. Older plants were more effective at DIN removal while younger plants were more effective at PO4 removal. When normalized for biomass, younger plants had much higher removal rates. These results demonstrated that both crop and cropping method have considerable impacts on nutrient removal. This method will allow operators to adjust their system quickly and easily to meet remediation goals.
Summary
Buzby and Lin developed a rapid batch method for measuring nutrient removal by an aquaponic plant component in isolation from the fish-rearing side, using trout raceway effluent at a West Virginia University research facility. In the first experiment they compared TAN, nitrate and phosphate removal by lettuce (‘Red Sails’) and nasturtium (‘Whirlibird Mix’) grown in a spring-fed, flow-through effluent channel over a 4-hour static (inflow-stopped) trial: nasturtium removed both TAN and nitrate, reaching 80% dissolved inorganic nitrogen (DIN) removal, while lettuce removed only TAN and actually saw nitrate rise slightly (48% DIN removal); both crops were slower at phosphate removal than nitrogen removal. In a second experiment they repeated the same batch removal assay on lettuce at 6, 10 and 13 weeks from sowing to see how plant age affects uptake, finding that phosphate removal (especially normalized per unit biomass) declined steadily as plants aged, while nitrogen removal was more variable across the three ages. The paper argues the source water’s low N:P ratio (6.6) explains the comparatively slow, N-limited phosphate removal in this system. The authors frame the isolated-channel method as a practical tool for growers to estimate the retention time or channel length needed to hit a water-quality target for a specific crop, without running a multi-month whole-system trial.
Experiment data
- Location: West Virginia University Aquaculture Facility, Morgantown, WV, USA (fish-rearing building + adjacent unheated 7.9 x 14.6 m double-poly/polycarbonate greenhouse); coordinates not stated in the paper
- Design: Two separate batch nutrient-removal experiments using effluent temporarily diverted from trout raceways into isolated plant channels (inflow stopped during measurement, TAN-spiked ~30 min pre-trial, hourly water sampling for up to 4 h). Experiment 1 = crop comparison, one channel of lettuce vs. one channel of nasturtium. Experiment 2 = plant-age comparison, lettuce measured at 6, 10 and 13 weeks from sowing.
- Replicates / n: One channel per treatment (crop or age) — no independent replicate channels; 3 trays/channel; 3 plants destructively sampled per channel for biomass (sub-sample, not an independent replicate)
- Duration: Each removal trial = 4 h; Experiment 2 plants assessed at 6, 10 and 13 weeks from sowing
- Organisms: Lettuce (Lactuca sativa) cv. ‘Red Sails’; Nasturtium (Tropaeolum majus) cv. ‘Whirlibird Mix’; effluent source only — Rainbow trout (Oncorhynchus mykiss) (not co-located with plants during the trial itself)
- Statistics: Linear and exponential regression of nutrient concentration vs. time (r2 reported, Tables 1-2); no significance test (ANOVA/t-test) comparing crops or ages is reported
- Total Ammonia Nitrogen (TAN): Lettuce 81% reduction to a final 0.11 mg/L; Nasturtium 89% reduction to a final 0.06 mg/L (Exp. 1, 4 h)
- Nitrate (NO3): Nasturtium reduced from 0.30 to 0.11 mg/L; Lettuce did not reduce nitrate — concentration rose over the 4 h, attributed to nitrification, no numeric value given (Exp. 1)
- Phosphate (PO4): initial 0.14 mg/L (shared source water); Nasturtium 63% reduction to 0.05 mg/L; Lettuce 37% reduction to 0.09 mg/L (Exp. 1, 4 h; see Extraction notes on the PO4 vs. PO4-P unit ambiguity)
- DIN (TAN+NO3) removal, % of initial: Lettuce 48%, Nasturtium 80% (Exp. 1)
Crop comparison (Experiment 1): lettuce vs. nasturtium
This paper: Nasturtium out-performed lettuce on every removal metric measured. Both crops removed TAN quickly (lettuce 81% -> 0.11 mg/L, nasturtium 89% -> 0.06 mg/L over 4 h, plateauing after ~3 h), but only nasturtium also removed nitrate (0.30 -> 0.11 mg/L), giving it 80% overall DIN removal against lettuce’s 48%. Neither crop was as effective at phosphate removal as at DIN removal; the authors attribute this to a source-water N:P ratio of 6.6, which by Koerselman and Meuleman’s (1996) threshold (N:P < 14) indicates N-limitation, and note that preliminary experiments found ammonia additions increased PO4 removal rates. Regression-derived retention times to reach 50% removal were substantially shorter for nasturtium than lettuce for both DIN (1.6 h vs. 4.2 h, exponential model) and PO4 (2.7 h vs. 5.3 h) (Table 1). The authors flag their own finding — that lettuce did not remove nitrate at all — as surprising given nitrate removal by lettuce is documented in other aquaponic and hydroponic systems, and offer several candidate explanations: their system’s much lower baseline DIN concentration (0.56 mg/L vs. >11-27 mg/L in Lennard & Leonard 2004 and Rakocy et al. 1993), a preference for ammonium over nitrate at low N concentrations (citing Xu et al. 1992), and colder water temperature (12.5 degC trout facility average vs. 22-28 degC in warm-water tilapia/catfish aquaponic systems) increasing reliance on ammonium as an N source (citing von Wiren et al. 1997).
Compared with:
- todo Rakocy et al. 2006 — recommended 60-100 g feed/m2 of plant growing area for a tilapia aquaponic system; this paper’s method is offered as an alternative, faster way to derive an analogous sizing figure for other crops/fish. (p.39-40, Introduction)
- todo Al-Hafedh et al. 2008 — determined an optimum ratio of 56 g feed/m2 for tilapia in Saudi Arabia. (p.39-40)
- todo Endut et al. 2010 — established 15-42 g fish feed/m2 for African catfish/water spinach in Malaysia. (p.40)
- todo Rakocy et al. 1993 — reported aquaponic nitrate concentrations of 22-26.7 mg/L, much higher than this paper’s 0.56 mg/L combined inorganic N, offered as a possible reason lettuce here did not remove nitrate. (p.41)
- todo Lennard and Leonard 2004 — reported average nitrate concentrations >11 mg/L in a gravel-bed aquaponic system. (p.41)
- todo Lennard and Leonard 2006 — nitrate removal by lettuce documented in a comparison of three hydroponic sub-systems within an aquaponic system. (p.41)
- todo Savvas et al. 2006 — nitrate removal by lettuce documented in a closed hydroponic (non-aquaponic) system. (p.41)
- todo Xu et al. 1992 — ammonium preferred over nitrate as N source when concentrations are low (maize). (p.41)
- todo von Wiren et al. 1997 — review: low temperature increases plant reliance on ammonium as mineral N source. (p.41)
Plant-age comparison (Experiment 2): lettuce at 6, 10 and 13 weeks
This paper: DIN removal rate was not monotonic with age — lowest at 10 weeks (0.143 mg/L/h) and highest at 13 weeks (0.248 mg/L/h), versus 0.215 mg/L/h at 6 weeks (Table 2, exponential model). Phosphate removal rate, by contrast, decreased steadily with age (0.241 -> 0.120 -> 0.116 mg/L/h). When both are normalized for channel biomass (which grew from 9.97 to 20.70 to 56.65 kg fresh weight over the three ages), both DIN and PO4 normalized removal rates decreased considerably with age (DIN: 0.0216 -> 0.0069 -> 0.0044 mg/L/kg FW/h; PO4: 0.0242 -> 0.0058 -> 0.0020 mg/L/kg FW/h). The authors tie the early, high per-biomass PO4 removal to fast relative growth rate early on (0.10, declining to 0.01 by 10 weeks and unchanged thereafter — units not stated in the paper) and to phosphate’s role in root development and possible luxury uptake in young plants. The unexpected dip-then-rise in DIN removal at 10-13 weeks is explicitly flagged by the authors as “difficult to interpret,” speculating an unmeasured environmental stressor (e.g. high air temperature) rather than a nutrient-supply effect, since nutrient concentrations were reportedly consistent across the experiment.
Compared with:
- todo Clarkson 1985 — nutrient uptake regulated by growth rate and nutrient demand, cited to interpret the high early-stage phosphate removal. (p.42)
- todo Adler et al. 2003 — proposes a “conveyor” cropping system (staggered cohorts by age along the channel) as a way to keep nutrient removal more consistent over time than a single-age batch planting; cited as a design implication of this paper’s age-effect finding. (p.42)
Linked claims
- Nasturtium removes both TAN and nitrate from aquaponic effluent while lettuce removes only TAN
- Younger plants remove phosphate faster per unit biomass than older plants in aquaponic systems
- Low water temperature and low dissolved inorganic nitrogen concentration favor ammonium over nitrate uptake in aquaponic lettuce
- Isolating the plant component from the fish component allows rapid measurement of aquaponic nutrient removal rates
Citations to chase
- todo Rakocy, Hargreaves & Bailey (1993) — nutrient accumulation in a recirculating aquaculture system integrated with hydroponic vegetable production; source of the 22-26.7 mg/L nitrate comparison figure.
- todo Lennard and Leonard (2004) — reciprocating vs. constant flow gravel-bed aquaponic system comparison; source of the >11 mg/L nitrate comparison figure.
- todo Koerselman and Meuleman (1996) — vegetation N:P ratio as a nutrient-limitation diagnostic; basis for this paper’s N-limitation argument.
- todo Adler, Summerfelt, Glenn and Takeda (2003) — mechanistic approach to phytoremediation of water; source of the “conveyor” cropping-system concept.
- todo Schneider, Sereti, Eding and Verreth (2005) — cited in the Introduction for the 30-65% feed N / up to 40% feed P excretion estimate; not this paper’s own measurement.
Extraction notes
Severity tally: 0 BLOCK, 0 MATERIAL, 2 CHECK, 0 MINOR -> quality: ok (0 BLOCK and <=2 MATERIAL; CHECK tier does not count toward the BLOCK/MATERIAL threshold per SCHEMA.md). No misattribution found. No numeric contradictions were found between the abstract, body text, figures and tables — all cross-checked values (DIN %, retention times, Table 2 rate rankings vs. body-text description) agree.
- ⚠️CHECK — Phosphate unit basis. Methods (2.3, p.40) states phosphate was analyzed as “phosphate-P (4500-P E. Ascorbic acid method),” a Standard Methods procedure that reports orthophosphate as mg P/L. Results text and Fig. 1/2 axes instead label the same values as “PO4” (e.g. “The initial concentration was 0.14 mg/L,” Fig. 1 y-axis “mg/L PO4”). The paper never states whether the reported numbers are mg PO4/L or mg PO4-P/L — a ~3.07x difference (PO4/P molar mass ratio). Not reconcilable from the text. No dedicated phosphate column exists in trials.csv’s schema (only TAN/NH4-N, NO2-N, NO3-N have columns), so this affects only the NO COLUMN phosphate figures recorded in Experimental Remarks, which are reported exactly as printed by the paper (unconverted), with this ambiguity flagged there per trial.
- ⚠️CHECK — Plant-age basis (“weeks from sowing” vs. “days after transplant”). The trials.csv schema’s “Days Plant after transplant” column implies a transplant event, but Experiment 2 reports plant age only as “6, 10 and 13 weeks” from sowing (Section 3.2, p.42; Table 2 header “Time since sowing (weeks)”). Methods 2.2 describes seeds sown directly into trays that were placed into the channel “immediately after sowing” — no separate transplanting step into a different growing medium/location is described. Sowing and channel-placement are therefore effectively simultaneous, but the paper never uses the word “transplant.” Recorded under the clearest available basis (weeks-from-sowing, unit-converted to days: 42/70/91) with this basis stated in each Experiment-2 trial row’s Experimental Remarks.
- Not flagged as a contradiction, noted only: the Discussion (p.41) states “our system had a combined inorganic N (nitrate + TAN) concentration of 0.56 mg/L,” used to contrast with warmer aquaponic systems in the literature. This describes the system’s typical/baseline operating concentration, not the deliberately TAN-spiked starting concentration of the Experiment 1 removal trial (“channels were spiked with TAN to maximize the range of conditions,” Methods 2.2) — these are two different quantities describing two different conditions, not a conflict, so no cell was affected and no candidate values were read off Fig. 1’s DIN panel.
- Site-description specific conductance is printed as “161 S/cm2” (p.40) with no micro-sign, almost certainly a dropped “µ” in PDF text extraction (161 µS/cm2 is the physically sensible reading for a freshwater spring and is the value carried into trials.csv’s EC column via UNIT CONVERSION ONLY: 0.161 dS/m). Not treated as a data contradiction since only one value is printed; flagged here for transparency only.
[not reported] fields (grouped): Fish Category; Initial Stock density; FCR; SGR; Protein/N/P/K % of feed; % of body weight (feed rate); Fish size initial/final; Feed routine; Feed regime; Total Feed (kg); Fish biomass created (kg); Fish survival rate; Fish weight gain; Fish trial duration (days) — the trout facility is described only as an effluent source; no fish growth-performance data of any kind is reported anywhere in the paper. Also NR: Water recycle (channel-specific L/min); Water volume in the system; Daily Water exchange rate; pHOptimal; Dissolved Oxygen; NO2-N (never analyzed — only TAN, NO3, PO4 were measured, per Methods 2.3); Plants/m2; SPAD; Plant height; Leaf count; Plant fresh weight (per-plant; only channel-total biomass in kg is given, Table 2); Plant dry matter (%); Tissue nitrate AP; Lat/Long (no coordinates given in the paper — not looked up externally per the prime directive); Average room Temperature; Biological system already in use; Air supplement; Iron supplemented; Remineralization; pH Buffers; Artificial Lighting; Days Plant after transplant for T1/T2 (Experiment 1 gives no plant-age figure at all, only that germination/growth were “visually monitored to confirm plants were growing well” beforehand).
[unclear] fields: none beyond the two ⚠️CHECK items above.
No column home (kept in trials.csv Experimental Remarks only): Phosphate (PO4/PO4-P, see CHECK above) initial/final concentrations and both exponential and linear removal-rate/normalized-rate/retention-time/r2 values (Table 1); DIN exponential and linear removal-rate/normalized-rate/retention-time/r2 values and overall %DIN-removed (Table 1, body text); Table 2’s per-age DIN and PO4 removal rates and biomass-normalized rates plus channel biomass (kg FW) for Experiment 2; source-water N:P ratio (6.6); relative growth rate time course (0.10 -> 0.01, units not stated); facility-wide (non-trial-specific) spring/effluent baseline chemistry (flow ~1325 L/min, pH 6.9, background nitrate 0.28 mg/L, background phosphate 0.15 mg/L, no detectable TAN, raceway-elevated TAN range 0.08-0.49 mg/L, nitrate/phosphate increase <=0.05 mg/L); greenhouse and channel construction dimensions; raceway/facility fish-capacity figures (8 sections, 3629 kg capacity); tray cell counts (128-cell lettuce, 32-cell nasturtium, 3 trays/channel).
Plant-analyte (plant.csv) note — 0 rows produced: No biochemistry, tissue-mineral, microbiology, or proximate analyte was measured on plant tissue anywhere in this paper. The only plant measurements are fresh/dry biomass (destructive weighing, used only to normalize water nutrient-removal rates) and the water nutrient concentrations themselves (TAN, nitrate, phosphate) — the latter are explicitly water chemistry and route to trials.csv per SCHEMA.md, not to plant.csv. plant.csv for this paper is therefore header-only.
Water panel excluded from plant.csv: none additional beyond the above — all water-chemistry values (TAN, NO3, PO4, spring baseline chemistry, EC, pH, temperature) were routed to trials.csv (as trial-mean/range values in the schema’s existing columns) or to trials.csv’s Experimental Remarks (as NO COLUMN items where no dedicated column exists, e.g. phosphate). None were misrouted into plant.csv.
New tags introduced: Meta/Type/Exploratory (new leaf — no existing note in this vault currently uses this type; follows the Title-Case leaf convention of Meta/Type/Experiment, Meta/Type/Policy, Meta/Type/Meta-analysis). Meta/Fish/Trout (new — checked against existing Meta/Fish/Goldfish, Meta/Fish/Koi, Meta/Fish/Koi-Carp, Meta/Fish/Tilapia, Meta/Fish/Yellow-Perch; no existing Trout leaf). Meta/Plant/Nasturtium (new — checked against existing Meta/Plant/Basil, Meta/Plant/Chicory, Meta/Plant/Coriander, Meta/Plant/Indian-Spinach, Meta/Plant/Lettuce, Meta/Plant/Rocket; no existing Nasturtium leaf). Meta/Region/North-America and Meta/Plant/Lettuce reused as spelled in existing vault notes (e.g. andersonGrowthTissueElemental2017, abbeyBasilOcimumBasilicum2022).
New wikilink targets introduced: [[Total Ammonia Nitrogen (TAN)]], [[Phosphate (PO4)]] — no existing note of either name was found in the vault (checked via full-text search); [[Nitrate (NO3)]] reused from existing vault usage (e.g. atiqueAquaponicsBeneficialTerms2022, aslanidouNutrientsUseEfficiency2023).
Type classification, justified: exploratory, not experiment. The authors collected new empirical data from a real physical system they operated (trout raceway effluent diverted into isolated plant channels, real water sampling and lab analysis) — this rules out modelling (no simulation or computational model is built; the paper’s “scaling”/“balancing” framing in the title and Introduction describes the motivating industry problem, not a model the authors construct) and rules out review/narrative-review (substantial original data collected). However, per SCHEMA.md’s decision rule 3, exploratory vs. experiment turns on replication and formal hypothesis testing: each treatment (crop or plant age) was run in a single, unreplicated channel, and the only statistical procedure applied is linear/exponential regression of concentration against time (r2 reported) — there is no ANOVA, t-test, or other between-group significance test anywhere in the paper. The authors themselves describe the work as designed “to explore the capabilities of the method” (Introduction, p.40), which is the paper’s own framing of intent, consistent with an unreplicated, hypothesis-generating pilot rather than a fully replicated comparative experiment.
Source: Buzby and Lin - 2014 - Scaling aquaponic systems Balancing plant uptake .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
buzbyScalingAquaponicSystems2014-T1
Fish
| Field | Value |
|---|---|
| Fish | Rainbow trout (Oncorhynchus mykiss) |
Water
| Field | Value |
|---|---|
| Water type | Aquaculture effluent (spring-fed, flow-through trout raceway effluent) |
| Water classification | Flow-through |
| Aq pH | 6.9 |
| EC | UNIT CONVERSION ONLY: 0.161 dS/m (161 uS/cm2 as printed, p.40) |
| Water temperature | 12.5 |
| TAN / NH4-N | range only: initial NR (81% reduction stated, Fig.1) - final 0.11 mg/L |
| NO2-N | NR (never analyzed; only TAN, NO3, PO4 measured per Methods 2.3) |
| NO3-N | NR (nitrate rose over the 4-h trial, attributed to nitrification; no numeric value reported, Results 3.1) |
Plant
| Field | Value |
|---|---|
| Plant | Lettuce (Lactuca sativa ‘Red Sails’) |
| Details | Sown directly into 128-cell Styrofoam trays w/ vermiculite; 3 trays placed in channel immediately after sowing; nutrient removal measured in Experiment 1 (crop comparison), single 4-h batch trial, hourly sampling after TAN spike |
| Plant Category | Leafy vegetable |
| Days Plant after transplant | NR (Experiment 1 gives no plant-age figure; plants only ‘visually monitored to confirm growing well’ before measurement) |
System & Setup
| Field | Value |
|---|---|
| System type | Flow-through plywood/EPDM-lined trough channel (38 cm W x 2.44 m L x 15 cm water depth); plants in floated Styrofoam trays; inflow temporarily stopped and submersible pump used for mixing during the batch removal trial |
| Media Details | Vermiculite (Therm-O-Rock East Inc., Grade 3A) in 128-cell Styrofoam seed-starter trays, 3 trays/channel |
| Climate control | Y (Thermostatically-driven louvered shutter (east wall) + exhaust fan (west wall); roll-up sides venting lower 1.2 m; unheated double-poly/polycarbonate greenhouse, no supplemental heating) |
| Nutrient supplemented | Y (Channels spiked with TAN (ammonia) ~30 min before each removal trial ‘to maximize the range of conditions the channels might operate under’ (Methods 2.2) - an experimental TAN spike for the removal-rate assay, not a routine fertigation/nutrient-dosing regime) |
| Equipment | Submersible magnetic-drive statuary pump (Model SP 330, Tetra) for in-channel mixing; 0.5 hp submersible pump (raceway tail box to distribution manifold); 2.5 cm ball valve (channel inlet); 5 cm drain standpipe (channel outlet, maintains 15 cm depth) |
| Control Parameters | Channel inflow temporarily stopped for the duration of each removal trial (batch/static-water mode); TAN spike applied ~30 min pre-trial; water sampled at t=0 then hourly to 4 h (plus t=0.5 h when biomass was substantial) |
| Combination | Trout raceway effluent (flow-through, spring-fed) directed to a temporarily isolated plant channel for a batch nutrient-removal assay; fish and plant components physically separate during measurement |
Site
| Field | Value |
|---|---|
| Region | North America |
| Country | United States |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | mg/L (TAN, NO3-N, PO4 removal); g fresh/dry weight (destructive biomass sampling, oven-dried 102 degC) |
| Statistic Details | Linear and exponential regression of nutrient concentration vs. time (r2 reported, Tables 1-2); no significance test (ANOVA/t-test) comparing crops or plant ages is reported |
| Statistically analysed | Y |
| Replicates (n) | 1 |
Experimental Remarks: TRIAL DEFINITION: T1 = Experiment 1, lettuce channel, single 4-h batch nutrient-removal trial (TAN-spiked, inflow stopped, hourly sampling). Paired comparison crop = nasturtium (T2), same experiment/conditions; no hydroponic control in this paper (AP vs HP left empty; FUE/WUE = NA). Replication = 1 channel (unreplicated); n=3 plants destructively sampled from this channel for biomass (sub-sample, not an independent replicate). | WARN-CHECK Phosphate unit basis: Methods 2.3 states phosphate analyzed as ‘phosphate-P (4500-P E. Ascorbic acid method)’ (implies mg P/L), but Results text and Fig. 1/2 y-axes label the same values ‘PO4’ (e.g. ‘initial concentration was 0.14 mg/L’, axis ‘mg/L PO4’). Paper never states which basis the numbers use (~3.07x apart: PO4 vs PO4-P). No dedicated phosphate column exists in this schema, so this affects only the NO COLUMN PO4 figures below, recorded exactly as printed by the paper (unconverted). | NO COLUMN: PO4 (as printed, see unit caveat above) initial 0.14 mg/L, final 0.09 mg/L (37% removal); exponential removal rate 0.12 mg/L/h, normalized 0.019 mg/L/kg FW/h, retention time to 50% removal 5.3 h, r2=0.99 (Table 1); linear removal rate 0.01 mg/L/h, normalized 0.002 mg/L/kg FW/h, retention 5.9 h, r2=0.99. DIN (TAN+NO3, paper’s own definition) exponential removal rate 0.17 mg/L/h, normalized 0.027 mg/L/kg FW/h, retention 4.2 h, r2=0.98; linear removal rate 0.11 mg/L/h, normalized 0.017 mg/L/kg FW/h, retention 4.1 h, r2=0.96; overall %DIN removed over 4 h = 48%. Source/system water N:P ratio = 6.6 (cited via Koerselman & Meuleman 1996 as indicative of N-limitation, offered as explanation for slower PO4 vs DIN removal). Facility-wide (not trial-specific) source spring: flow ~1325 L/min, pH 6.9, specific conductance printed ‘161 S/cm2’ (almost certainly 161 uS/cm2, likely dropped mu-symbol in PDF extraction; UNIT CONVERSION ONLY: 0.161 dS/m used in EC column), background nitrate 0.28 mg/L, background phosphate 0.15 mg/L, no detectable TAN. Raceway effluent (facility-wide operating range, not this trial’s measured value) elevates TAN to 0.08-0.49 mg/L with nitrate/phosphate increases <=0.05 mg/L over spring baseline (range only, no trial mean given for this facility-wide figure). Greenhouse: unheated 7.9 x 14.6 m, double polyethylene film + polycarbonate end walls, roll-up sides (lower 1.2 m vented), 32x32 mesh screening. Channel: plywood/EPDM(45 mil)-lined trough, 38 cm W x 2.44 m L, 15 cm water depth, 2.5 cm inlet ball valve, 5 cm outlet drain standpipe. Fish facility: trout (Oncorhynchus mykiss) reared in 8-section linear raceways (each 7.3x0.9 m rearing zone + 1.8x0.9 m quiescent zone), facility capacity 3629 kg maintained via periodic harvests - fish were the effluent source only; fish were not co-located with, nor measured alongside, the plant trial (channel inflow from raceway effluent was temporarily stopped during the 4-h removal measurement itself).
buzbyScalingAquaponicSystems2014-T2
Fish
| Field | Value |
|---|---|
| Fish | Rainbow trout (Oncorhynchus mykiss) |
Water
| Field | Value |
|---|---|
| Water type | Aquaculture effluent (spring-fed, flow-through trout raceway effluent) |
| Water classification | Flow-through |
| Aq pH | 6.9 |
| EC | UNIT CONVERSION ONLY: 0.161 dS/m (161 uS/cm2 as printed, p.40) |
| Water temperature | 12.5 |
| TAN / NH4-N | range only: initial NR (89% reduction stated, Fig.1) - final 0.06 mg/L |
| NO2-N | NR (never analyzed; only TAN, NO3, PO4 measured per Methods 2.3) |
| NO3-N | range only: 0.30 (initial) - 0.11 (final) mg/L, stated in Results 3.1 |
Plant
| Field | Value |
|---|---|
| Plant | Nasturtium (Tropaeolum majus ‘Whirlibird Mix’) |
| Details | Sown directly into 32-cell Styrofoam trays w/ vermiculite; 3 trays placed in channel immediately after sowing; nutrient removal measured in Experiment 1 (crop comparison), single 4-h batch trial, hourly sampling after TAN spike |
| Plant Category | Edible flowering ornamental |
| Days Plant after transplant | NR (Experiment 1 gives no plant-age figure; plants only ‘visually monitored to confirm growing well’ before measurement) |
System & Setup
| Field | Value |
|---|---|
| System type | Flow-through plywood/EPDM-lined trough channel (38 cm W x 2.44 m L x 15 cm water depth); plants in floated Styrofoam trays; inflow temporarily stopped and submersible pump used for mixing during the batch removal trial |
| Media Details | Vermiculite (Therm-O-Rock East Inc., Grade 3A) in 32-cell Styrofoam seed-starter trays, 3 trays/channel |
| Climate control | Y (Thermostatically-driven louvered shutter (east wall) + exhaust fan (west wall); roll-up sides venting lower 1.2 m; unheated double-poly/polycarbonate greenhouse, no supplemental heating) |
| Nutrient supplemented | Y (Channels spiked with TAN (ammonia) ~30 min before each removal trial ‘to maximize the range of conditions the channels might operate under’ (Methods 2.2) - an experimental TAN spike for the removal-rate assay, not a routine fertigation/nutrient-dosing regime) |
| Equipment | Submersible magnetic-drive statuary pump (Model SP 330, Tetra) for in-channel mixing; 0.5 hp submersible pump (raceway tail box to distribution manifold); 2.5 cm ball valve (channel inlet); 5 cm drain standpipe (channel outlet, maintains 15 cm depth) |
| Control Parameters | Channel inflow temporarily stopped for the duration of each removal trial (batch/static-water mode); TAN spike applied ~30 min pre-trial; water sampled at t=0 then hourly to 4 h (plus t=0.5 h when biomass was substantial) |
| Combination | Trout raceway effluent (flow-through, spring-fed) directed to a temporarily isolated plant channel for a batch nutrient-removal assay; fish and plant components physically separate during measurement |
Site
| Field | Value |
|---|---|
| Region | North America |
| Country | United States |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | mg/L (TAN, NO3-N, PO4 removal); g fresh/dry weight (destructive biomass sampling, oven-dried 102 degC) |
| Statistic Details | Linear and exponential regression of nutrient concentration vs. time (r2 reported, Tables 1-2); no significance test (ANOVA/t-test) comparing crops or plant ages is reported |
| Statistically analysed | Y |
| Replicates (n) | 1 |
Experimental Remarks: TRIAL DEFINITION: T2 = Experiment 1, nasturtium channel, single 4-h batch nutrient-removal trial (TAN-spiked, inflow stopped, hourly sampling). Paired comparison crop = lettuce (T1), same experiment/conditions; no hydroponic control in this paper (AP vs HP left empty; FUE/WUE = NA). Replication = 1 channel (unreplicated); n=3 plants destructively sampled from this channel for biomass (sub-sample, not an independent replicate). | WARN-CHECK Phosphate unit basis: Methods 2.3 states phosphate analyzed as ‘phosphate-P (4500-P E. Ascorbic acid method)’ (implies mg P/L), but Results text and Fig. 1/2 y-axes label the same values ‘PO4’ (e.g. ‘initial concentration was 0.14 mg/L’, axis ‘mg/L PO4’). Paper never states which basis the numbers use (~3.07x apart: PO4 vs PO4-P). No dedicated phosphate column exists in this schema, so this affects only the NO COLUMN PO4 figures below, recorded exactly as printed by the paper (unconverted). | NO COLUMN: PO4 (as printed, see unit caveat above) initial 0.14 mg/L, final 0.05 mg/L (63% removal); exponential removal rate 0.26 mg/L/h, normalized 0.042 mg/L/kg FW/h, retention time to 50% removal 2.7 h, r2=0.99 (Table 1); linear removal rate 0.02 mg/L/h, normalized 0.004 mg/L/kg FW/h, retention 2.8 h, r2=0.97. DIN (TAN+NO3) exponential removal rate 0.43 mg/L/h, normalized 0.070 mg/L/kg FW/h, retention 1.6 h, r2=0.98; linear removal rate 0.19 mg/L/h, normalized 0.030 mg/L/kg FW/h, retention 2.4 h, r2=0.98; overall %DIN removed over 4 h = 80%. Source/system water N:P ratio = 6.6 (cited via Koerselman & Meuleman 1996 as indicative of N-limitation). Facility-wide (not trial-specific) source spring: flow ~1325 L/min, pH 6.9, specific conductance printed ‘161 S/cm2’ (almost certainly 161 uS/cm2, likely dropped mu-symbol in PDF extraction; UNIT CONVERSION ONLY: 0.161 dS/m used in EC column), background nitrate 0.28 mg/L, background phosphate 0.15 mg/L, no detectable TAN. Raceway effluent (facility-wide operating range, not this trial’s measured value) elevates TAN to 0.08-0.49 mg/L with nitrate/phosphate increases <=0.05 mg/L over spring baseline (range only, no trial mean given for this facility-wide figure). Greenhouse: unheated 7.9 x 14.6 m, double polyethylene film + polycarbonate end walls, roll-up sides (lower 1.2 m vented), 32x32 mesh screening. Channel: plywood/EPDM(45 mil)-lined trough, 38 cm W x 2.44 m L, 15 cm water depth, 2.5 cm inlet ball valve, 5 cm outlet drain standpipe. Fish facility: trout (Oncorhynchus mykiss) reared in 8-section linear raceways (each 7.3x0.9 m rearing zone + 1.8x0.9 m quiescent zone), facility capacity 3629 kg maintained via periodic harvests - fish were the effluent source only; fish were not co-located with, nor measured alongside, the plant trial (channel inflow from raceway effluent was temporarily stopped during the 4-h removal measurement itself).
buzbyScalingAquaponicSystems2014-T3
Fish
| Field | Value |
|---|---|
| Fish | Rainbow trout (Oncorhynchus mykiss) |
Water
| Field | Value |
|---|---|
| Water type | Aquaculture effluent (spring-fed, flow-through trout raceway effluent) |
| Water classification | Flow-through |
| Aq pH | 6.9 |
| EC | UNIT CONVERSION ONLY: 0.161 dS/m (161 uS/cm2 as printed, p.40) |
| Water temperature | 12.5 |
| TAN / NH4-N | NR (Experiment 2 reports removal RATES only, Table 2; concentrations shown only graphically in Fig.2, not read off figure per schema rule) |
| NO2-N | NR (never analyzed; only TAN, NO3, PO4 measured per Methods 2.3) |
| NO3-N | NR (see TAN / NH4-N note; Fig.2 shows nitrate ‘little affected or increased slightly’, no numeric value in text) |
Plant
| Field | Value |
|---|---|
| Plant | Lettuce (Lactuca sativa ‘Red Sails’) |
| Details | Same cultivar/system as Experiment 1 (128-cell Styrofoam trays w/ vermiculite); nutrient removal measured longitudinally in Experiment 2 (plant-age comparison) at 6 weeks from sowing; single 4-h batch trial |
| Plant Category | Leafy vegetable |
| Days Plant after transplant | 42 (UNIT CONVERSION ONLY: 6 weeks from sowing x 7). WARN-CHECK plant-age basis: paper states age as ‘weeks from sowing’ (Table 2 header ‘Time since sowing (weeks)’), not explicitly ‘after transplant’; Methods 2.2 places trays in the channel ‘immediately after sowing’ with no separate transplant event described, so sowing and channel-placement are effectively simultaneous. Recorded under the clearest available basis (sowing-to-measurement interval); UNIT CONVERSION ONLY: weeks x 7 = days. |
System & Setup
| Field | Value |
|---|---|
| System type | Flow-through plywood/EPDM-lined trough channel (38 cm W x 2.44 m L x 15 cm water depth); plants in floated Styrofoam trays; inflow temporarily stopped and submersible pump used for mixing during the batch removal trial |
| Media Details | Vermiculite (Therm-O-Rock East Inc., Grade 3A) in 128-cell Styrofoam seed-starter trays, 3 trays/channel |
| Climate control | Y (Thermostatically-driven louvered shutter (east wall) + exhaust fan (west wall); roll-up sides venting lower 1.2 m; unheated double-poly/polycarbonate greenhouse, no supplemental heating) |
| Nutrient supplemented | Y (Channels spiked with TAN (ammonia) ~30 min before each removal trial ‘to maximize the range of conditions the channels might operate under’ (Methods 2.2) - an experimental TAN spike for the removal-rate assay, not a routine fertigation/nutrient-dosing regime) |
| Equipment | Submersible magnetic-drive statuary pump (Model SP 330, Tetra) for in-channel mixing; 0.5 hp submersible pump (raceway tail box to distribution manifold); 2.5 cm ball valve (channel inlet); 5 cm drain standpipe (channel outlet, maintains 15 cm depth) |
| Control Parameters | Channel inflow temporarily stopped for the duration of each removal trial (batch/static-water mode); TAN spike applied ~30 min pre-trial; water sampled at t=0 then hourly to 4 h (plus t=0.5 h when biomass was substantial) |
| Combination | Trout raceway effluent (flow-through, spring-fed) directed to a temporarily isolated plant channel for a batch nutrient-removal assay; fish and plant components physically separate during measurement |
Site
| Field | Value |
|---|---|
| Region | North America |
| Country | United States |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | mg/L (TAN, NO3-N, PO4 removal); g fresh/dry weight (destructive biomass sampling, oven-dried 102 degC) |
| Statistic Details | Linear and exponential regression of nutrient concentration vs. time (r2 reported, Tables 1-2); no significance test (ANOVA/t-test) comparing crops or plant ages is reported |
| Statistically analysed | Y |
| Replicates (n) | 1 |
Experimental Remarks: TRIAL DEFINITION: T3 = Experiment 2, lettuce channel at 6 weeks from sowing, single 4-h batch nutrient-removal trial (TAN-spiked, inflow stopped, hourly sampling), part of a 3-point longitudinal age series (6/10/13 weeks) on the same cultivar/system as Experiment 1. No hydroponic control in this paper (AP vs HP left empty; FUE/WUE = NA). Replication = 1 channel per age (unreplicated); n=3 plants destructively sampled per age for biomass (sub-sample, not an independent replicate). | WARN-CHECK Phosphate unit basis: Methods 2.3 states phosphate analyzed as ‘phosphate-P (4500-P E. Ascorbic acid method)’ (implies mg P/L), but Results text and Fig. 1/2 y-axes label the same values ‘PO4’ (e.g. ‘initial concentration was 0.14 mg/L’, axis ‘mg/L PO4’). Paper never states which basis the numbers use (~3.07x apart: PO4 vs PO4-P). No dedicated phosphate column exists in this schema, so this affects only the NO COLUMN PO4 figures below, recorded exactly as printed by the paper (unconverted). | NO COLUMN (Table 2, exponential model): channel biomass 9.97 kg fresh weight; DIN removal rate 0.215 mg/L/h, normalized 0.0216 mg/L/kg FW/h; PO4 (as printed, see unit caveat above) removal rate 0.241 mg/L/h, normalized 0.0242 mg/L/kg FW/h. Relative growth rate (units not stated by paper): fastest early in experiment at 0.10, declined to 0.01 by 10 weeks, unchanged to end of experiment (Results 3.2) - offered by authors as explanation for high early biomass-normalized PO4 removal (root development / possible luxury P uptake in young plants). Facility-wide (not trial-specific) source spring: flow ~1325 L/min, pH 6.9, specific conductance printed ‘161 S/cm2’ (almost certainly 161 uS/cm2, likely dropped mu-symbol in PDF extraction; UNIT CONVERSION ONLY: 0.161 dS/m used in EC column), background nitrate 0.28 mg/L, background phosphate 0.15 mg/L, no detectable TAN. Raceway effluent (facility-wide operating range, not this trial’s measured value) elevates TAN to 0.08-0.49 mg/L with nitrate/phosphate increases <=0.05 mg/L over spring baseline (range only, no trial mean given for this facility-wide figure). Greenhouse: unheated 7.9 x 14.6 m, double polyethylene film + polycarbonate end walls, roll-up sides (lower 1.2 m vented), 32x32 mesh screening. Channel: plywood/EPDM(45 mil)-lined trough, 38 cm W x 2.44 m L, 15 cm water depth, 2.5 cm inlet ball valve, 5 cm outlet drain standpipe. Fish facility: trout (Oncorhynchus mykiss) reared in 8-section linear raceways (each 7.3x0.9 m rearing zone + 1.8x0.9 m quiescent zone), facility capacity 3629 kg maintained via periodic harvests - fish were the effluent source only; fish were not co-located with, nor measured alongside, the plant trial (channel inflow from raceway effluent was temporarily stopped during the 4-h removal measurement itself).
buzbyScalingAquaponicSystems2014-T4
Fish
| Field | Value |
|---|---|
| Fish | Rainbow trout (Oncorhynchus mykiss) |
Water
| Field | Value |
|---|---|
| Water type | Aquaculture effluent (spring-fed, flow-through trout raceway effluent) |
| Water classification | Flow-through |
| Aq pH | 6.9 |
| EC | UNIT CONVERSION ONLY: 0.161 dS/m (161 uS/cm2 as printed, p.40) |
| Water temperature | 12.5 |
| TAN / NH4-N | NR (Experiment 2 reports removal RATES only, Table 2; concentrations shown only graphically in Fig.2, not read off figure per schema rule) |
| NO2-N | NR (never analyzed; only TAN, NO3, PO4 measured per Methods 2.3) |
| NO3-N | NR (see TAN / NH4-N note; Fig.2 shows nitrate ‘little affected or increased slightly’, no numeric value in text) |
Plant
| Field | Value |
|---|---|
| Plant | Lettuce (Lactuca sativa ‘Red Sails’) |
| Details | Same cultivar/system as Experiment 1 (128-cell Styrofoam trays w/ vermiculite); nutrient removal measured longitudinally in Experiment 2 (plant-age comparison) at 10 weeks from sowing; single 4-h batch trial |
| Plant Category | Leafy vegetable |
| Days Plant after transplant | 70 (UNIT CONVERSION ONLY: 10 weeks from sowing x 7). WARN-CHECK plant-age basis: paper states age as ‘weeks from sowing’ (Table 2 header ‘Time since sowing (weeks)’), not explicitly ‘after transplant’; Methods 2.2 places trays in the channel ‘immediately after sowing’ with no separate transplant event described, so sowing and channel-placement are effectively simultaneous. Recorded under the clearest available basis (sowing-to-measurement interval); UNIT CONVERSION ONLY: weeks x 7 = days. |
System & Setup
| Field | Value |
|---|---|
| System type | Flow-through plywood/EPDM-lined trough channel (38 cm W x 2.44 m L x 15 cm water depth); plants in floated Styrofoam trays; inflow temporarily stopped and submersible pump used for mixing during the batch removal trial |
| Media Details | Vermiculite (Therm-O-Rock East Inc., Grade 3A) in 128-cell Styrofoam seed-starter trays, 3 trays/channel |
| Climate control | Y (Thermostatically-driven louvered shutter (east wall) + exhaust fan (west wall); roll-up sides venting lower 1.2 m; unheated double-poly/polycarbonate greenhouse, no supplemental heating) |
| Nutrient supplemented | Y (Channels spiked with TAN (ammonia) ~30 min before each removal trial ‘to maximize the range of conditions the channels might operate under’ (Methods 2.2) - an experimental TAN spike for the removal-rate assay, not a routine fertigation/nutrient-dosing regime) |
| Equipment | Submersible magnetic-drive statuary pump (Model SP 330, Tetra) for in-channel mixing; 0.5 hp submersible pump (raceway tail box to distribution manifold); 2.5 cm ball valve (channel inlet); 5 cm drain standpipe (channel outlet, maintains 15 cm depth) |
| Control Parameters | Channel inflow temporarily stopped for the duration of each removal trial (batch/static-water mode); TAN spike applied ~30 min pre-trial; water sampled at t=0 then hourly to 4 h (plus t=0.5 h when biomass was substantial) |
| Combination | Trout raceway effluent (flow-through, spring-fed) directed to a temporarily isolated plant channel for a batch nutrient-removal assay; fish and plant components physically separate during measurement |
Site
| Field | Value |
|---|---|
| Region | North America |
| Country | United States |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | mg/L (TAN, NO3-N, PO4 removal); g fresh/dry weight (destructive biomass sampling, oven-dried 102 degC) |
| Statistic Details | Linear and exponential regression of nutrient concentration vs. time (r2 reported, Tables 1-2); no significance test (ANOVA/t-test) comparing crops or plant ages is reported |
| Statistically analysed | Y |
| Replicates (n) | 1 |
Experimental Remarks: TRIAL DEFINITION: T4 = Experiment 2, lettuce channel at 10 weeks from sowing, single 4-h batch nutrient-removal trial (TAN-spiked, inflow stopped, hourly sampling), part of a 3-point longitudinal age series (6/10/13 weeks) on the same cultivar/system as Experiment 1. No hydroponic control in this paper (AP vs HP left empty; FUE/WUE = NA). Replication = 1 channel per age (unreplicated); n=3 plants destructively sampled per age for biomass (sub-sample, not an independent replicate). | WARN-CHECK Phosphate unit basis: Methods 2.3 states phosphate analyzed as ‘phosphate-P (4500-P E. Ascorbic acid method)’ (implies mg P/L), but Results text and Fig. 1/2 y-axes label the same values ‘PO4’ (e.g. ‘initial concentration was 0.14 mg/L’, axis ‘mg/L PO4’). Paper never states which basis the numbers use (~3.07x apart: PO4 vs PO4-P). No dedicated phosphate column exists in this schema, so this affects only the NO COLUMN PO4 figures below, recorded exactly as printed by the paper (unconverted). | NO COLUMN (Table 2, exponential model): channel biomass 20.70 kg fresh weight; DIN removal rate 0.143 mg/L/h, normalized 0.0069 mg/L/kg FW/h; PO4 (as printed, see unit caveat above) removal rate 0.120 mg/L/h, normalized 0.0058 mg/L/kg FW/h. Relative growth rate (units not stated by paper): fastest early in experiment at 0.10, declined to 0.01 by 10 weeks, unchanged to end of experiment (Results 3.2) - offered by authors as explanation for high early biomass-normalized PO4 removal (root development / possible luxury P uptake in young plants). Facility-wide (not trial-specific) source spring: flow ~1325 L/min, pH 6.9, specific conductance printed ‘161 S/cm2’ (almost certainly 161 uS/cm2, likely dropped mu-symbol in PDF extraction; UNIT CONVERSION ONLY: 0.161 dS/m used in EC column), background nitrate 0.28 mg/L, background phosphate 0.15 mg/L, no detectable TAN. Raceway effluent (facility-wide operating range, not this trial’s measured value) elevates TAN to 0.08-0.49 mg/L with nitrate/phosphate increases <=0.05 mg/L over spring baseline (range only, no trial mean given for this facility-wide figure). Greenhouse: unheated 7.9 x 14.6 m, double polyethylene film + polycarbonate end walls, roll-up sides (lower 1.2 m vented), 32x32 mesh screening. Channel: plywood/EPDM(45 mil)-lined trough, 38 cm W x 2.44 m L, 15 cm water depth, 2.5 cm inlet ball valve, 5 cm outlet drain standpipe. Fish facility: trout (Oncorhynchus mykiss) reared in 8-section linear raceways (each 7.3x0.9 m rearing zone + 1.8x0.9 m quiescent zone), facility capacity 3629 kg maintained via periodic harvests - fish were the effluent source only; fish were not co-located with, nor measured alongside, the plant trial (channel inflow from raceway effluent was temporarily stopped during the 4-h removal measurement itself).
buzbyScalingAquaponicSystems2014-T5
Fish
| Field | Value |
|---|---|
| Fish | Rainbow trout (Oncorhynchus mykiss) |
Water
| Field | Value |
|---|---|
| Water type | Aquaculture effluent (spring-fed, flow-through trout raceway effluent) |
| Water classification | Flow-through |
| Aq pH | 6.9 |
| EC | UNIT CONVERSION ONLY: 0.161 dS/m (161 uS/cm2 as printed, p.40) |
| Water temperature | 12.5 |
| TAN / NH4-N | NR (Experiment 2 reports removal RATES only, Table 2; concentrations shown only graphically in Fig.2, not read off figure per schema rule) |
| NO2-N | NR (never analyzed; only TAN, NO3, PO4 measured per Methods 2.3) |
| NO3-N | NR (see TAN / NH4-N note; Fig.2 shows nitrate ‘little affected or increased slightly’, no numeric value in text) |
Plant
| Field | Value |
|---|---|
| Plant | Lettuce (Lactuca sativa ‘Red Sails’) |
| Details | Same cultivar/system as Experiment 1 (128-cell Styrofoam trays w/ vermiculite); nutrient removal measured longitudinally in Experiment 2 (plant-age comparison) at 13 weeks from sowing; single 4-h batch trial |
| Plant Category | Leafy vegetable |
| Days Plant after transplant | 91 (UNIT CONVERSION ONLY: 13 weeks from sowing x 7). WARN-CHECK plant-age basis: paper states age as ‘weeks from sowing’ (Table 2 header ‘Time since sowing (weeks)’), not explicitly ‘after transplant’; Methods 2.2 places trays in the channel ‘immediately after sowing’ with no separate transplant event described, so sowing and channel-placement are effectively simultaneous. Recorded under the clearest available basis (sowing-to-measurement interval); UNIT CONVERSION ONLY: weeks x 7 = days. |
System & Setup
| Field | Value |
|---|---|
| System type | Flow-through plywood/EPDM-lined trough channel (38 cm W x 2.44 m L x 15 cm water depth); plants in floated Styrofoam trays; inflow temporarily stopped and submersible pump used for mixing during the batch removal trial |
| Media Details | Vermiculite (Therm-O-Rock East Inc., Grade 3A) in 128-cell Styrofoam seed-starter trays, 3 trays/channel |
| Climate control | Y (Thermostatically-driven louvered shutter (east wall) + exhaust fan (west wall); roll-up sides venting lower 1.2 m; unheated double-poly/polycarbonate greenhouse, no supplemental heating) |
| Nutrient supplemented | Y (Channels spiked with TAN (ammonia) ~30 min before each removal trial ‘to maximize the range of conditions the channels might operate under’ (Methods 2.2) - an experimental TAN spike for the removal-rate assay, not a routine fertigation/nutrient-dosing regime) |
| Equipment | Submersible magnetic-drive statuary pump (Model SP 330, Tetra) for in-channel mixing; 0.5 hp submersible pump (raceway tail box to distribution manifold); 2.5 cm ball valve (channel inlet); 5 cm drain standpipe (channel outlet, maintains 15 cm depth) |
| Control Parameters | Channel inflow temporarily stopped for the duration of each removal trial (batch/static-water mode); TAN spike applied ~30 min pre-trial; water sampled at t=0 then hourly to 4 h (plus t=0.5 h when biomass was substantial) |
| Combination | Trout raceway effluent (flow-through, spring-fed) directed to a temporarily isolated plant channel for a batch nutrient-removal assay; fish and plant components physically separate during measurement |
Site
| Field | Value |
|---|---|
| Region | North America |
| Country | United States |
Results & Statistics
| Field | Value |
|---|---|
| Measured Unit | mg/L (TAN, NO3-N, PO4 removal); g fresh/dry weight (destructive biomass sampling, oven-dried 102 degC) |
| Statistic Details | Linear and exponential regression of nutrient concentration vs. time (r2 reported, Tables 1-2); no significance test (ANOVA/t-test) comparing crops or plant ages is reported |
| Statistically analysed | Y |
| Replicates (n) | 1 |
Experimental Remarks: TRIAL DEFINITION: T5 = Experiment 2, lettuce channel at 13 weeks from sowing, single 4-h batch nutrient-removal trial (TAN-spiked, inflow stopped, hourly sampling), part of a 3-point longitudinal age series (6/10/13 weeks) on the same cultivar/system as Experiment 1. No hydroponic control in this paper (AP vs HP left empty; FUE/WUE = NA). Replication = 1 channel per age (unreplicated); n=3 plants destructively sampled per age for biomass (sub-sample, not an independent replicate). | WARN-CHECK Phosphate unit basis: Methods 2.3 states phosphate analyzed as ‘phosphate-P (4500-P E. Ascorbic acid method)’ (implies mg P/L), but Results text and Fig. 1/2 y-axes label the same values ‘PO4’ (e.g. ‘initial concentration was 0.14 mg/L’, axis ‘mg/L PO4’). Paper never states which basis the numbers use (~3.07x apart: PO4 vs PO4-P). No dedicated phosphate column exists in this schema, so this affects only the NO COLUMN PO4 figures below, recorded exactly as printed by the paper (unconverted). | NO COLUMN (Table 2, exponential model): channel biomass 56.65 kg fresh weight; DIN removal rate 0.248 mg/L/h, normalized 0.0044 mg/L/kg FW/h; PO4 (as printed, see unit caveat above) removal rate 0.116 mg/L/h, normalized 0.0020 mg/L/kg FW/h. Relative growth rate (units not stated by paper): fastest early in experiment at 0.10, declined to 0.01 by 10 weeks, unchanged to end of experiment (Results 3.2) - offered by authors as explanation for high early biomass-normalized PO4 removal (root development / possible luxury P uptake in young plants). Facility-wide (not trial-specific) source spring: flow ~1325 L/min, pH 6.9, specific conductance printed ‘161 S/cm2’ (almost certainly 161 uS/cm2, likely dropped mu-symbol in PDF extraction; UNIT CONVERSION ONLY: 0.161 dS/m used in EC column), background nitrate 0.28 mg/L, background phosphate 0.15 mg/L, no detectable TAN. Raceway effluent (facility-wide operating range, not this trial’s measured value) elevates TAN to 0.08-0.49 mg/L with nitrate/phosphate increases <=0.05 mg/L over spring baseline (range only, no trial mean given for this facility-wide figure). Greenhouse: unheated 7.9 x 14.6 m, double polyethylene film + polycarbonate end walls, roll-up sides (lower 1.2 m vented), 32x32 mesh screening. Channel: plywood/EPDM(45 mil)-lined trough, 38 cm W x 2.44 m L, 15 cm water depth, 2.5 cm inlet ball valve, 5 cm outlet drain standpipe. Fish facility: trout (Oncorhynchus mykiss) reared in 8-section linear raceways (each 7.3x0.9 m rearing zone + 1.8x0.9 m quiescent zone), facility capacity 3629 kg maintained via periodic harvests - fish were the effluent source only; fish were not co-located with, nor measured alongside, the plant trial (channel inflow from raceway effluent was temporarily stopped during the 4-h removal measurement itself).