ArticlePDF Available

Algal turf scrubber (ATS) floways on the Great Wicomico River, Chesapeake Bay: Productivity, algal community structure, substrate and chemistry1


Abstract and Figures

Two Algal Turf Scrubber (ATS) units were deployed on the Great Wicomico River (GWR) for 22months to examine the role of substrate in increasing algal productivity and nutrient removal. The yearly mean productivity of flat ATS screens was 15.4 g center dot m-2 center dot d-1. This was elevated to 39.6g center dot m-2 center dot d-1 with a three-dimensional (3-D) screen, and to 47.7g center dot m-2 center dot d-1 by avoiding high summer harvest temperatures. These methods enhanced nutrient removal (N, P) in algal biomass by 3.5 times. Eighty-six algal taxa (Ochrophyta [diatoms], Chlorophyta [green algae], and Cyan-obacteria [blue-green algae]) self-seeded from the GWR and demonstrated yearly cycling. Silica (SiO2) content of the algal biomass ranged from 30% to 50% of total biomass; phosphorus, nitrogen, and carbon content of the total algal biomass ranged from 0.15% to 0.21%, 2.13% to 2.89%, and 20.0% to 25.7%, respectively. Carbohydrate content (at 10%-25% of AFDM) was dominated by glucose. Lipids (fatty acid methyl ester; FAMEs) ranged widely from 0.5% to 9% AFDM, with Omega-3 fatty acids a consistent component. Mathematical modeling of algal produ-ctivity as a function of temperature, light, and substrate showed a proportionality of 4:3:3, resp-ectively. Under landscape ATS operation, substrate manipulation provides a considerable opportunity to increase ATS productivity, water quality amelioration, and biomass coproduction for fertilizers, fermentation energy, and omega-3 products. Based on the 3-D prod-uctivity and algal chemical composition demonstrated, ATS systems used for nonpoint source water treat-ment can produce ethanol (butanol) at 5.8x per unit area of corn, and biodiesel at 12.0x per unit area of soy beans (agricultural production US).
Content may be subject to copyright.
Walter H. Adey
, H. Dail Laughinghouse IV
Department of Botany, National Museum of Natural History, Smithsonian Institution, Washington, D.C., 20013, USA
John B. Miller
Chemistry Department MS5413, Western Michigan University, Kalamazoo, Michigan 49008, USA
Lee-Ann C. Hayek
Statistics and Mathematics, National Museum of Natural History, Smithsonian Institution, Washington, D.C., 20560, USA
Jesse G. Thompson, Steven Bertman, Kristin Hampel, and Shanmugam Puvanendran
Chemistry Department MS5413, Western Michigan University, Kalamazoo, Michigan 49008, USA
Two Algal Turf Scrubber (ATS) units were deployed
on the Great Wicomico River (GWR) for 22 months to
examine the role of substrate in increasing algal
productivity and nutrient removal. The yearly mean
productivity of flat ATS screens was 15.4 g m
This was elevated to 39.6 g m
with a three-
dimensional (3-D) screen, and to 47.7 g m
avoiding high summer harvest temperatures. These
methods enhanced nutrient removal (N, P) in algal
biomass by 3.5 times. Eighty-six algal taxa (Ochrophyta
[diatoms], Chlorophyta [green algae], and Cyan-
obacteria [bluegreen algae]) self-seeded from the
GWR and demonstrated yearly cycling. Silica (SiO
content of the algal biomass ranged from 30% to 50%
of total biomass; phosphorus, nitrogen, and carbon
content of the total algal biomass ranged from 0.15%
to 0.21%, 2.13% to 2.89%, and 20.0% to 25.7%,
respectively. Carbohydrate content (at 10%25% of
AFDM) was dominated by glucose. Lipids (fatty acid
methyl ester; FAMEs) ranged widely from 0.5% to 9%
AFDM, with Omega-3 fatty acids a consistent
component. Mathematical modeling of algal produ-
ctivity as a function of temperature, light, and
substrate showed a proportionality of 4:3:3, resp-
ectively. Under landscape ATS operation, substrate
manipulation provides a considerable opportunity to
increase ATS productivity, water quality amelioration,
and biomass coproduction for fertilizers, fermentation
energy, and omega-3 products. Based on the 3-D prod-
uctivity and algal chemical composition demonstrated,
ATS systems used for nonpoint source water treat-
ment can produce ethanol (butanol) at 5.83per unit
area of corn, and biodiesel at 12.03per unit area of
soy beans (agricultural production US).
Key index words: Algal Turf Scrubber
; biochemistry;
nutrient removal; productivity enhancement; species
Abbreviations: 2-D, two-dimensional; 3-D, three-
dimensional; ATS
, Algal Turf Scrubber; DHA,
docosahexaenoic acid; EPA, eicosapentaenoic acid;
FAMEs, fatty acid methyl esters; GC, gas chromato-
graph; GWR, Great Wicomico River; HDPE, high
density polyethylene; ICP-OES, inductively coupled
plasma optical emission spectroscopy; MSD, mass
selective detector; PUFA’s, polyunsaturated fatty
The Algal Turf Scrubber (ATS) is an ecologically
engineered, floway system for utilizing algal photo-
synthesis to control a wide variety of water quality
parameters. Developed in the early 1980s at the
Smithsonian Institution as a biomimicry of coral
reef primary productivity, ATS was initially used as a
tool to manage an extensive series of living micro-
cosm and mesocosm models of wild ecosystems
(Adey and Loveland 2007). Applied to closed living
models, ATS units functioned to control nutrients,
oxygen levels, carbonate systems, including calcifica-
tion (through CO
control) and to minimize toxic
compounds from the local human-engineered envi-
ronment. ATS also allowed the development of
planktonic communities and water borne reproduc-
tion in model ecosystems, as it has little effect on
the planktonic component.
Algal Turf Scrubber was successfully scaled-up by
HydroMentia, Inc. for nutrient removal from point-
source and semi point-source open waters during
the 1990s and early 21st century. ATS use ranged
from aquaculture and tertiary treatment of sewage to
agricultural canal amelioration of nutrients (Adey
et al. 2011). By 2007, eight scaled-up ATS systems
Received 28 June 2012. Accepted 13 December 2012.
Author for correspondence: e-mail:
Editorial Responsibility: A. Buschmann (Associate Editor)
J. Phycol. 49, 489–501 (2013)
©2013 Phycological Society of America
DOI: 10.1111/jpy.12056
had been built and operated from coast to coast,
mostly in the southern tier of states, including a
2.83 ha Tilapia operation in Falls City, Texas, that
produced commercial quantities of fish for 9 years,
and a 0.1 ha, ~950,000 L d
tertiary sewage system
in the northern Central Valley of California (Craggs
et al. 1996). The ATS process has been used for
water quality amelioration in a wide variety of envi-
ronments and some of the earlier projects provided
water quality analyses (Craggs et al. 1996, Stewart
2006). However, broad-based analytical studies, have
been lacking, especially those that could allow the
assessment of byproduct utilization of biomass.
Earlier non-point source work with ATS, had con-
centrated on the cleaning of environments rich in
either hard benthic substrates (i.e., rock or
branches) with abundant algal turfs as biofilms, or
aquatic flowering plants that contained periphyton,
epiphytically, on their stems. Those ATS systems
were highly dominated by filamentous chlorophytes,
particularly species of the genera Cladophora, Spiro-
gyra, Microspora, Ulothrix, and Rhizoclonium. Diatoms
and cyanobacteria were usually present, especially as
epiphytes, but rarely provided significant biomass
(Adey et al. 1993, Craggs et al. 1996). Later work
on larger water bodies showed a greater dominance
of diatoms and very high levels of species diversity,
but generally lower productivity, as some filamen-
tous diatoms tended to shear-off the floways before
harvest (Sandefur et al. 2011, Laughinghouse 2012).
Algal Turf Scrubber was developed as a simple con-
tinuous flow low-cost means of utilizing algal photo-
synthetic and productivity potential using attached
algae. Algal harvest and water/algal separation can
be easily accomplished by suspending water flow and
allowing the system to drain by gravity, since the algae
are attached to the substrate. A wide variety of scrap-
ing and vacuuming methods apply. When nutrients
are moderately high and solar energy moderately
abundant, productivities ranging from 25 to
45 g m
were common on ATS systems
(Adey and Loveland 2007, Mulbry et al. 2008, Adey
et al. 2011). The combined use of ATS to reclaim
nutrients from eutrophic waters and to produce algal
biomass coproducts, such as biofuels, soil amend-
ments, or high-value extracts, reduces the costs of
both processes. Equally important, the costs of algal
production using ATS can be considerably lower than
photobioreactor methods (Adey et al. 2011).
As a cohort of living organisms, including lesser
amounts of bacteria, protists, and small inverte-
brates, the algal turf communities of ATS systems
contain varying quantities of compounds essential
to life: carbohydrates, lipids, proteins, genetic mate-
rials, structural components like silica, and a pleth-
ora of other organic and inorganic substances. It is
known that individual species have different carbon
(C), nitrogen (N), phosphorus (P), and silicon
dioxide (SiO
) bioaccumulation rates (Cade-Menun
and Paytan 2010) as well as unique responses to
light and temperature (Hessen et al. 2002, Wang
and Lan 2011). Algal turfs can also contain extracel-
lular materials, including terrigenous sediments. In
this study, we analyzed the as-harvested algal turf
biomass as a whole, rather than an analysis of indi-
vidual species.
A large range of macro- and micro-algal species
have rigid carbohydrate-based cell walls containing
large quantities of simple and complex carbohy-
drates (Noseda et al. 1999). The structural carbohy-
drate composition of the diverse algal species may
vary as the amount of sunlight, nutrients, and har-
vesting periods also change (Imbs et al. 2009,
Skriptsova et al. 2010). Moreover, a large subset of
algal species, particularly the Chlorophyta, uses
carbohydrates as storage compounds (Raven and
Beardall 2003). Regardless of biological function, if
the carbohydrates can be liberated from the algal
biomass the potential exists to produce bio-alcohols
(ethanol and butanol) through fermentation (Potts
et al. 2011, Wang and Lan 2011). The specific advan-
tages of developing algae as a feedstock for bio
-alcohol production include the following: nutrient
absorption from eutrophic waters, oxygen injection
into those waters, negligible competition with terres-
trial agricultural starch or sugar crops for valuable
arable land, a high growth-rate, and the potential for
carbon dioxide absorption (Singh and Olsen 2011).
Algal lipids have been studied extensively as a source
for biodiesel production (Chisti 2007). However,
some algal lipids have greater value in nonfuel prod-
ucts. Polyunsaturated fatty acids (PUFAs) are a com-
ponent of the lipid profile in algae. This compound
class, especially the omega-3 fatty acids, appears
essential for human health. Algae are a potential
alternative source to fish oil for essential fatty acids
anchez et al. 2008, Inhamuns and Franco
2008). This article presents the productivity, commu-
nity structure and chemical analyses of algal biomass
produced from ATS floways on the Great Wicomico
River (GWR) near Reedville, Virginia. In addition,
we investigated the role of algal growth substrate in
controlling ATS production and the relative effects
of temperature and light.
The GWR is a small tributary that lacks significant fresh
water input, located on the west-central shore of the Chesa-
peake Bay, Virginia. It is mesohaline in character, with yearly
salinity ranging from 11 to 18. The experiments were carried
out on a 70 m long dock located on a 1 km wide embayment,
about 6 km from the open Chesapeake Bay (Fig. 1). During
this study, a semidiurnal tidal current (0.1 kn maximum)
flowed perpendicular to the dock’s axis. The ATS Floways
(#1 and #2) were 0.61 m wide fiberglass troughs, with 15.2 m
length (1% slope) and 24.4 m length (2% slope), respec-
The floways had oscillating input trays (Fig. S1, see Sup-
porting Information) that provided a surging motion in the
water column moving down the floway (Carpenter et al.
1991, Adey and Loveland 2007). Two submerged centrifugal
pumps (38 L min
flow) were installed on each floway and
placed on the outer part of the dock, with separate PVC pipe
lines (24.4 m long, #1; 15.2 m long, #2) leading to the oscil-
lating input trays. The pumps were placed on separate electri-
cal circuits and had an automatic backup electrical generator
in the event of failure of the commercial power source. Inter-
ruption of water flow occurred only during short-term non-
drying harvests over the 22 months of floway operation. This
is a critical element of ATS operation, since 46 weeks is typi-
cally required to build a mature, fully productive community
of algae (Fig. S2, see Supporting Information).
Water samples were taken on several occasions in both
autumn and spring for analysis of nitrogen and phosphorus,
and the samples were collected from the river near the pump
intakes and on the inflow and outflow of both floways. Sam-
ples were taken directly in 500 mL acid-washed plastic bottles
and frozen shortly after collection. Analyses were carried out
by the water quality laboratory at the Virginia Institute of
Marine Science. Water temperature and pH were measured
by hand with a Hanna Instruments HI 9024 microcomputer/
pH meter. Readings were taken weekly or bi-weekly, before
harvest at the river input and in the incoming and outflowing
water on the floways, and occasionally every two h from 0600
to 2200.
The two floways were studied from August 2009 to June
2011 (22 months). On Floway #1, tests were carried out on
algal growth and dynamics relative to different substrate
types, with an emphasis on comparing the traditional two-
dimensional (2-D) design to a variety of three-dimensional
(3-D) designs (Fig. S3, see Supporting Information). Growth
experiments on Floway #2 were primarily focused on measur-
ing productivity on a single type of 3-D screen prototype
developed after the initial Floway #1 studies showed a strong
increase in production on some 3-D types. In addition, two
short-term experiments on CO
introduction and lipid trig-
gers were undertaken on Floway #2 and will be described
fully elsewhere. All algal settlement on the floways “self-
seeded” from the GWR water and no filters were employed.
Floway #1. The traditional substrate in the ATS system is a
flat plastic screen. An HDPE screen of 3 95 mm mesh is
typical, although a wide range of mesh sizes and screen mate-
rials have been tested. On the scale of the ATS floway and
the attached algal community, these screens are 2-D struc-
tures. The dominant diatom communities that occurred on
Chesapeake ATS systems quickly attach to these “standard”
screens, but depending on the species, their filaments/chains
can constantly “shear-off” in the moderate energy environ-
ment of an ATS, producing a lower standing crop and ulti-
mately lower water remediation capabilities and biomass
accumulation than communities higher in filamentous green
This experiment was established primarily to determine
how to prevent the loss of fragile diatom filaments, and prin-
cipally involved examining the efficacy of 3-D screens/fabrics
as support structures. A wide variety of off-the-shelf, deep pile
throw rugs, with 12 cm thick loose fibers, were tested along
with special, more open variants produced for this project by
the carpet company InterfaceFLOR (La Grange, GA, USA).
Multiple layers of 2-D screens, and open plastic fabrics used
for soil retention were also examined. Most provided some
improvement in diatom retention over simple 2-D screens,
but unfortunately were readily degraded by solar UV. Several
screens were hand woven, with the specific purpose of struc-
turing a growth environment with the limitations of diatoms
in mind. Braided Dacron fibers (2 cm long) were attached to
a 5 mm mesh basal screen (Fig. S3). The Dacron was
employed because it would provide for minimal degradation
by solar UV. The braided fibers were used to provide maxi-
mum attachment surface for the diatoms. Two types, one with
a coarser braid (labeled #14) and the other with a fine,
“hairy” surface (labeled #17) were used. These were estab-
lished in the central part of the ATS test floway. Standard 2-D
and other 3-D ATS screens were arrayed both above and
below for comparison.
Sampling of test screens, and whole floway harvest were
generally performed every 7 d in the summer and 14 d in
the winter. Depending on the treatment, sample size on Flo-
way #1 varied from 0.13 to 0.14 m
; the standard sample size
on Floway #2 was 0.14 m
. There was a small amount of varia-
tion in harvest period due to weather and during the spring
and fall when switchover occurred from one harvest schedule
to the other. On the day of harvest, water flow on the ATS
floway was stopped and the system was allowed to drain for
one half to 1 h, depending on humidity and temperature. At
no time was the floway allowed to fully dry, as it was essential
not to reduce the viability of the (“seed”) algal community
remaining on the floway after harvest. Harvest of all test
screens, and the intervening 2-D screen that extended down
the whole floway, was achieved by hand with a small wet-dry
shop vacuum. While the vacuum severs and removes the pre-
dominance of the algae on the growing screen surface, deliv-
ering the biomass to the storage chamber of the shop
vacuum, a basal layer of short algal filaments remains
attached to the floway screens to quickly re-initiate growth.
Each test sample was more fully drained on coarse and
fine filters immediately after harvest, and the samples and
their filters dried at 50°C55°C in an Excalibur Food Dehy-
drator (Excalibur, USA) until weight was stable for 6 h (typi-
cally 4872 h total drying). The final dry weight was recorded
at 0.1 g and the samples stored, in plastic containers at
10°C12°C, for future reference. The wet, bulk sample of
whole floway harvest (about one half of the floway surface,
with 2-D screen) was placed in plastic 20 L buckets from
which several separate aliquots, each taken after stirring, were
extracted and placed in one liter glass, food-storage jars and
immediately placed in a freezer.
Starting in the 11th month of Floway #1 operation (July
2010), several aliquots were taken from the bulk sample,
FIG. 1. Floway #1 (left) and Floway # 2 (right) on the Great
Wicomico River off the Central Chesapeake Bay.
before freezing, for algal species analysis, placed in plastic
sample bags and kept refrigerated until analysis. At that time,
poorly performing 3-D screens were also removed, leaving a
substrate of 2-D screen on much of the bed of the floway for
the remainder of the experiment. Since the 3-D screens were
sampled before the whole floway harvest and the bulk sample
taken, 80%90% of Floway #1 biomass was provided from 2-D
Floway #2. The 3-D substrate used on Floway #2 was coar-
ser than the “optimal” hand-woven 3-D substrates on Floway
#1 (#14 and #17, Fig. S3). This was produced by the carpet
company InterfaceFLOR to examine the possibility of large-
scale 3-D screen production. Algal biomass accumulation on
the Floway #2 substrate (both on the smaller samples in Flo-
way #1 and on the whole Floway of #2) was about two-thirds
that of the optimal screens (#17, #14, 3-D) for the same 12-
month period (see Results). Floway #2 was not fully opera-
tional until summer 2010, and the lack of funding required a
shutdown early in the summer of 2011. Thus, a full data set
was not available for the highly productive summer interval.
However, Floway #2 provided the critical analysis of a 3-D
screen growth substrate as applied to a whole floway unit,
rather than the smaller individual sample screens.
Algal species analyses. Random aliquots from the bulk sam-
ple were homogenized and a 5 mL subsample was blended
with water to a 400 mL final concentration. A subsample of
100 mL was separated and standard Lugol’s solution was added
to sediment the sample; it was allowed to stabilize in prepara-
tion for quantitative analyses. Identification to lowest taxo-
nomic level possible and relative abundance counts were
conducted using a Zeiss inverted microscope (Zeiss, Oberko-
chen, Germany) at 4009, following the Uterm
ohl procedure
ohl 1958). Samples are maintained in a liquid herbar-
ium in Rm. E-117 at the National Museum of Natural History,
Smithsonian Institution, Washington, D.C., USA.
Algal chemical analyses.Analytical methods: Sources and
preparation of the materials, reagents, and samples used for
chemical analysis of the ATS biomass are described in Appen-
dix S1, in the Supporting Information.
Ash analysis: The weight percent ash in each algal sample
) was determined according to the method
ASTM E 1755 2007 (ASTM, 2007 #27).
Nutrient analysis: Carbon, hydrogen, and nitrogen content
of algal samples was analyzed using a Leco Corporation True-
CHN analyzer. Lyophilized algae were oven-dried at
105°C for one h and stored in a desiccator prior to analysis.
The instrument was calibrated before each set of analyses
using an EDTA standard and blank checks were run between
every 10 and 15 analyses. Triplicate measurements were made
for all samples.
Total phosphorus levels of algal samples were determined
by a digestion and colorimetric method (Dick and Tabatabai
1977). Lyophilized algae (0.030.08 g) were digested in a solu-
tion of 2 M sodium hydroxide and liquid bromine at 250°C,
heating to dryness. The samples were then reconstituted with
formic acid and 0.5 M H
and analyzed by colorimetry
using a Perkin Elmer Lambda Bio 20 spectrophotometer at
kmax =711 nm. Triplicate measurements were made for all
The total silicon of algal samples was measured by a modi-
fied literature method (Reay and Bennett 1987). Briefly, a
solution of potassium tetraborate and potassium nitrate was
heated at 100°C in clean oxidized nickel crucibles, until com-
pletely evaporated, then cooled to ambient temperature.
A weighed portion (2550 mg) lyophilized algae was added
to the crucibles and heated in air at 300°C for 2 h. Solid
potassium hydroxide (0.4 g) was added to each crucible,
heated until molten (~>410°C) and held at temperature for
30 min. After cooling, the soluble residue was extracted with
20 mL of water and centrifuged. Aliquots of the supernatant
were analyzed by inductively coupled plasma optical emission
spectroscopy (Perkin Elmer Optima 2100DV ICP-OES). Trip-
licate digestions were performed for each biomass sample
and triplicate measurements made for all digestions. Total
was calculated from the total silicon, which presumed
no other significant silicon compounds.
The proportion of biogenic silica was determined using a
time-digestion procedure modified from that of Conley and
Schelske (1993). Briefly, algal biomass (3050 mg) was
weighed into six 50 mL polypropylene centrifuge tubes, to
each of which was added 40 mL of 5 wt.% sodium carbonate
solution. All but one of the tubes was heated to 70°C under
continuous stirring. A 1 mL aliquot was removed from the
unheated tube immediately, and at intervals (1, 2, 3, 4, 5,
and 24 h) a 1 mL aliquot was taken from a previously
un-sampled heated tube. The sodium carbonate in each of
the sampled aliquots was quenched immediately with 5%
nitric acid. The acidified samples were diluted and analyzed
by ICP-OES. The proportion of silica that is biogenic was
determined by performing a linear fit of the silicon in solu-
tion versus digestion time for times longer than 2 h; the bio-
genic silicon is the y-intercept of the fitted line, as described
in the literature method.
Carbohydrate analysis: Preparation of the ATS algal biomass
for carbohydrate analysis was a three-step process: (i) hydroly-
sis with strong acid to convert all carbohydrates to their constit-
uent monosaccharides, (ii) reduction of the monosaccharides
to their corresponding alditols, and (iii) derivatization of the
monosaccharide hydroxyls with acetyl groups. The derivatized
carbohydrates were identified and quantified by Gas Chroma-
tography Mass Spectrometry (GC-MS) analysis.
Lipid analysis: Fatty-acid based lipids from the lyophilized
algae samples were converted to the corresponding fatty acid
methyl esters (FAMEs) and analyzed by gas chromatography
(GC). Reporting fatty-acid based lipids on a FAMEs basis
leads to a systematic underestimate of somewhat less than 1%
in the absolute quantity for triglyceride lipids, and a system-
atic overestimate of 3%4% for any phospholipids. However,
these systematic deviations are generally small compared to
the variability in the overall analytical method.
Modeling of productivity. In this study, we also analyzed pri-
mary productivity as a function of light and temperature.
Light data were derived from NASA monthly predictions for
Annapolis, MD; Washington, D.C.; and Charlotte, NC. The
mean of these three data sets was equally matched against
the Smithsonian Institution SERC Laboratory daily light data
for the Rhode River, MD, just south of Annapolis. Tempera-
ture data were taken on site, as presented above.
Temperature and pH. Out-flowing water tempera-
ture changed little from river temperature on these
floways, although atmospheric temperature extremes
were observed as brief deviations of the discharge
temperature when compared to the river tempera-
ture (Fig. S4, see Supporting Information). River and
floway incoming water had a mean yearly pH of 8.25
(SD 0.15). At night, there was little change in pH,
however, through mid-day (09001500 h), pH rose
uniformly down Floway #2, reaching 8.5 (SD 0.24) by
the middle of the floway and 8.64 (SD 0.15) at the
effluent, as CO
was extracted from the water column
by algal photosynthesis.
Nutrients. The GWR had moderate nutrient levels
during both fall and spring (Table 1), and these
levels were not likely limiting for the production lev-
els seen on the screens of Floway #1. By paired sam-
ple t-test, TN (t
=2.245, P<0.029) and TP (t
=3.322, P<0.002) were both significantly
reduced between river source water and floway efflu-
ent. The significant TP reduction highly was due to
phosphorus precipitation with elevated pH. In the
analysis of the harvested biomass, no effort was
made to distinguish between extracellular and intra-
cellular phosphorus.
Productivity floway #1. The 3-D screens showed a
consistently greater level of algal biomass than the
2-D screen with summer production about five times
that of mid-winter production on both types of
growth substrate (t
=5.959, P<0.001; Fig. 2). Lar-
gely following temperature and light, the 3-D
screens produced at a level of about 2.5 times that
of the 2-D screens; the yearly mean for both 3-D
screens combined was 36.9 g m
, and that
for the combined 2-D screens was
15.4 g m
. Separate analysis of the 3-D
screens showed that production on the two screens
followed each other closely. However, the screen
with the finely braided and somewhat hairy surface
produced biomass 15%20% higher than the screen
with the coarser fibers. The difference was relatively
small, but highly significant (t
P0.0001), even though the two screens were sit-
uated only one meter apart on the floway. Three-
dimensional screen #17 was placed downstream of
3-D screen #14, and 3-D screen #17 appeared to be
at a disadvantage for production due to lesser nutri-
ents and suspended particulates. Biomass productiv-
ity on 2-D screens placed at about the 10% and
80% positions on Floway #1 followed each other
closely over the entire period of study and there was
no significant difference in production between the
two positions (t
Floway #2. The primary difference between the
two floways was the addition of 3-D substrate on the
entire length of Floway #2 (Fig. 1). The hand-woven
3-D substrates described above (Fig. S3) were nearly
three times as productive as the flat screens used in
previous studies. However, the only 3-D substrate
available in large quantity at the time of construc-
tion was one of the InterfaceFLOR types being
tested on Floway #1. We understood the characteris-
tics of this screen and how it related to highly pro-
ductive hand-woven type 3-D substrates (Fig. 2);
however, that information was only available as test
rectangles. This was the first time that 3-D screens
had been employed in an environment rich in dia-
toms and it was essential that we understood how
the 3-D substrate would perform on whole floways.
The manufactured 3-D screen used in Floway #2,
when deployed as a test square in Floway #1, pro-
duced biomass at approximately 67% of the opti-
mum 3-D screens (Fig. 2). This was the mean of the
manufactured types; the optimum manufactured
screen used on Floway #2 performed at a 13%
higher level. Thus, we would have expected the
upper part of Floway #2 to perform at about 80% of
its hand-woven 3-D equivalent. On the upper half of
Floway #2, during August and September, the manu-
factured screen produced at 4050 g m
while on floway #1, the 3-D screens were producing
at 6070 g m
and the 2-D screens at
2030 g m
. Therefore, 3-D screens can
TABLE 1. Mean nutrient (N, P) levels in the Great Wicomico River and in the outflow of Floways #1 & #2 during Fall 2010
and Spring 2011 in mg L
Fall 2010 mg L
River Outflow Floway #1 Outflow Floway #2 Spring 2011 mg L
River Outflow Floway #1 Outflow Floway #2
PN 0.1008 0.0683 0.0447 PN 0.239 0.134 0.079
PP 0.0066 0.0038 0.0023 PP 0.0126 0.006 0.0012
TDN 0.3234 0.2783 0.2442 TDN 0.3536 0.3493 0.3176
TDP 0.0095 0.007 0.0054 TDP 0.015 0.009 0.0059
TN 0.4242 0.3466 0.2889 TN 0.5926 0.4833 0.3966
TP 0.0161 0.0108 0.0077 TP 0.0276 0.015 0.0071
Six sampling periods of two samples each Five sampling periods of two samples each
PN, particulate nitrogen; PP, particulate phosphorus; TDN, total dissolved nitrogen; TDP, total dissolved phosphorus.
Each number represents 12 weekly samples (Fall 2010) and five weekly samples (Spring 2011). See text for statistical analysis.
FIG. 2. Biomass productivity on Floway #1, comparing 2-D
(dashed lines), 3-D (solid lines) and Interface (bold dotted line)
growth substrate (screens). The smooth curves are sine functions
fit to the two data sets. Note the one-year mean production
marked for the different screen types.
significantly increase biomass production on long
floways as well as on small test screens, but to
achieve optimum production the right screen con-
figuration is necessary.
Algal biodiversity.Floway #1: A total of 86 algal taxa
belonging to seven different phyla dominated by
Ochrophyta (54%), Chlorophyta (24%), and Cyano-
bacteria (22%) were found on this system (Rhodo-
phyta and Dinophyta provided less than 1% of the
flora). The most diverse phylum was Ochrophyta
(diatoms) with 78% of the total taxa (67 of 86 taxa)
coming from this group. Ochrophyta was also the
most abundant algal group on the system during
the study period 54% of total relative abundance.
The most abundant taxa were Berkeleya spp. (B. fen-
nica Juhlin-Dannfelt, B. fragilis Greville, and B. ruti-
lans (Trentepohl ex Roth) Grunow; 20%), Gloeothece
sp. (13%), Ulva intestinalis Linnaeus (10%), Ulothrix
sp. (9%), Melosira spp. (M. monoliformis (O. F.
Muller) C. Agardh and M. nummuloides C. Agardh;
9%), Lyngbya salina Gomont (7%), and Achrochaete
sp. (7%). Other taxa occurring on the floway indi-
vidually accounted for less than 2% of the total
abundance. The most frequent taxa (>84%) were
Achnanthes spp. (i.e., Achnanthes brevipes C. Agardh),
Amphora spp., B. rutilans,Mastogloia spp., M. nummu-
loides, Stauronella sp., Navicula spp., and Grammato-
phora spp.
Periphyton growing on these engineered systems
were dynamic (Fig. 3a). Diatoms were always present
and abundant; different species dominanted season-
ally. For example, Tabularia tabulata
(C. Agardh) Snoeijs, Nitzschia sigmoidea (Nitzsch)
W. Smith, and N. sigma (Kutzing) W. Smith were
most abundant from July to September, but were
absent during colder months when other taxa, such
as Nitzschia nana Grunow, Thalassionema nitzschiodes
(Grunow) Mereschkowsky, and Thalassiosira sp.,
appeared. However, as a group, chlorophytes tended
to increase during late spring through summer and
reached 70% of total abundance for a short period
in the spring. Cyanobacteria were abundant in fall
due to the large amount of two taxa: first, in mid-
September, Gloeothece sp. dominated the periphyton;
later in the fall, L. salina dominated.
Floway #2: On this floway, 98 algal taxa, belonging
to seven different phyla dominated by Ochrophyta
(68%), Chlorophyta (7%), and Cyanobacteria
(24%) were found (Rhodophyta and Dinophyta
together contributed 1% of the flora). The most
diverse phylum was Ochrophyta (diatoms) with 74%
of the total taxa (73 of 98 taxa) coming from this
group, while also being the most abundant algal
group on the system during the study period (68%
of total relative abundance). Other phyla accounted
for roughly 1% of total relative abundance. There
were 7 taxa of dinoflagellates compared to only 2
taxa on Floway #l. The most frequent taxa (>85%)
were Achnanthes spp. (i.e., A. brevipes), Amphora spp.,
B. rutilans, Grammatophora spp., Licmophora spp.,
M. nummuloides,Navicula spp., and T. tabulata. The
most abundant taxa were Berkeleya spp. (B. fennica
and B. rutilans; 19%), Lyngbya cf. salina (19%),
T. nitzschiodes (14%), M. nummuloides (7%), and
U. intestinalis (6%). The other taxa occurring on the
floway accounted for 3% or less of total abundance
The periphytic community on Floway #2 was also
seasonally dynamic (Fig. 3b). Diatoms were always
present, though there were changes in community
structure occurring throughout the year. Cyanobac-
teria increased from August to November (up to
60% of total abundance, dominated by L. salina).
Although present several times during the year,
chlorophytes peaked in abundance in late spring
(40% of the total relative abundance); both U. intes-
tinalis and Ulothrix spp. were responsible for this
Seasonality in species biomass and abundance was
marked on both floways. While the beginning of the
seasons were not exactly the same on the two flo-
ways, they were close enough to be regarded as
system characteristics dependent on a mix of light
and temperature. While both floways have the same
sets of species, seasonally and in relative abundance,
a few differences were marked. Acrochaete sp. and
Ulothrix sp. were important summer and spring spe-
cies on Floway #1, yet were unimportant on Floway
#2; the diatom Thalassionema nitzschiodes was the
dominant species in the winter on Floway #2, yet
was virtually absent on Floway #1.
The side-by-side floways had the same water
source, light, and temperature; yet had different
screen types (2-D vs. 3-D) and algal productivities.
The floways also differed in slope (Floway #1: 1%;
Floway #2: 2%); the higher slope was added to
Floway #2 to compensate for the effects of a
rougher screen surface on water flow. With the con-
tinuous 3-D substrate on Floway #2, the diatoms
Berkeleya and Melosira replaced the green alga Ulva,
especially in spring and summer.
FIG. 3. Seasonal relative abundance of algal species, by phyla
on Floways #1 (a Upper) and Floway #2 (b lower; July 2010 to
June 2011).
Algal turf biomass composition.Nutrients (C, N, P, Si):
Algal biomass was harvested from upper and lower
regions of Floway #1 and #2, at various times from
October 2010 through May 2011. There was no clear
seasonal trend in any of the C, N, and P of harvested
biomass, although N was somewhat higher and C
somewhat lower in late winter or early spring than in
early summer (Fig. 4). On the basis of the Redfield
ratio, we expected that the N, P, and C levels should
be correlated; however, there was no correlation
<0.07) between C and either N or P in the algal
biomass. To provide an overall assessment of nutrient
accumulation, each data set in Figure 4 was averaged;
the mean values are collected in Table 2. The C con-
tent of the algal biomass from Floway #2 was some-
what higher than that from Floway #1, with lower
Floway #2 biomass being higher in C than either
Floway #1 or upper Floway #2 (Table 2); P and N con-
tent of lower floway #2 were also somewhat higher.
Pair-wise comparisons between lower Floway #2 and
Floway #1 showed that the differences were generally
significant (Table 2).
The total quantity of SiO
decreased during the
late winter when the water temperature was cooler
and terrestrial runoff events were less frequent. In
any event, there was considerable variability in total
(Table 3). The basic origin of SiO
in the har-
vested material was biogenic and terrigenous. The
amount of biogenic SiO
was measured on two sam-
ple dates (23 February 2011 and 23 May 2011) with
a mild, basic digestion (Table 3).
Algal turf carbohydrates. The carbohydrate content
of the algal turf biomass was measured for a time
series of samples from Floway #1 and upper and
lower Floway #2, and also lacked a clear seasonal
correlation (Fig. 5, a and b). In comparison to the
population dynamics plots (Fig. 3, a and b), higher
carbohydrate levels corresponded to higher propor-
tions of Chlorophyta and Cyanobacteria in the turf
community. It was also clear that the most abundant
and variable monosaccharide component was
Fatty-acid lipids. Algae samples from Floway #1 and
upper and lower Floway #2 were analyzed for overall
fatty-acid based lipid content, especially the nutri-
tionally important x-3 polyunsaturated fatty acids ei-
cosapentaenoic acid (EPA) and docosahexaenoic
acid (DHA). The weight percentage of FAMEs on an
AFDM basis (g
) showed similar tem-
poral variation for all of the floways (Fig. 5c). By
comparison with the relative phyla abundances
(Fig. 3, a and b) higher lipid contents corresponded
to higher proportions of Ochrophyta. EPA and DHA
levels generally were correlated (Fig. 5, d and e),
with EPA levels systematically higher than those of
TABLE 2. Overall mean carbon, nitrogen, and phosphorus
composition of algal turf harvested from the Great
Wicomico River ATS systems.
Mean wt%/g g
Floway 1 0.147 0.051 2.125 0.070 20.88 2.36
Floway 2 (all) 0.177 0.052 2.438 0.590 22.22 0.52
Floway 2
0.162 0.019 2.680 0.301 21.53 2.60
Floway 2
0.213 0.059 2.890 0.463 25.67 2.9
FIG. 4. Temporal variation in levels of carbon (triangles),
nitrogen (crosses), and phosphorus (910; diamonds) measured
in algae from Floway 1 (top) upper Floway 2 (middle) and lower
Floway 2 (bottom). Carbon is plotted on the secondary axis. Error
bars represent the sample standard deviations of triplicate analy-
ses. Lines are linear fits.
DHA. Curiously, PUFA levels did not correlate with
variations in species distribution at the phyla level, as
the overall lipid levels did. Instead, specific genera
were more important contributors to the PUFA con-
tent, perhaps Thalassionema or Berkeleya.
Modeling of productivity. Algal production on the
GWR test floways was a function of both light and
temperature (Fig. 6). Mean production on 3-D
(#17) screen substrate was consistently about
39greater than that on the 2-D screens, as demon-
strated earlier for the yearly mean. Algal production
was related to temperature (Regression: F
= 131.272, P0.0001) and light (Regression: F
= 74.921, P0.0001); the regression line for tem-
perature had a higher slope than for light, and pro-
duction was more variable as a function of light
than temperature. The higher variation likely
resulted from the regional light database as com-
pared to the on-site temperature data. The ratio of
slopes between the two showed that temperature
provides 61.5% of the total rise and light about
Temperature and light were mutually important
on driving productivity (Fig. 7). To describe these
data, we chose a linear model (eq. 1) after compari-
son with multiple polynomials because the linear
equation fit with equal significance and provided
equivalent predictive power. The model described
the algal production on 3-D screen #17 as a com-
bined function of temperature and light (T-L),
where productivity P(T-L) had units of g m
and T-L had units of °C-mol photons d
P(T-L) ¼9:05 þ0:04(T-L) ð1Þ
Equation 1 was used to estimate the “true” values
of the six depressed data points in the plot of Fig-
ure 2 that we ascribed to excessive temperature
excursions during harvest. The time-dependent pro-
ductivity of the 3-D test screen #17 was plotted in Fig-
ure 8, with the six data points replaced by the
estimated values. An empirical time-dependent pro-
ductivity function P(t) derived from equation 1 and
the Temperature-Light function was described by
equation 2, where P(t), P
and P
had units of
;tand t
had units of days; and xhad
units of days radian
. The P
term was the average
annual productivity, and has a value of
35 g m
, while P
was the difference between
the annual minimum (winter) and maximum (sum-
mer) productivity and had a value of 32 g m
. Together, eqs 1 and 2 empirically described
the performance of the 3-D screen #17 growing
P(t) ¼P0Ptsinð2pðt-t0Þ=xÞð2Þ
Adey et al. (2011) reviewed the status of algal pro-
ductivity as used in ecologically engineered systems
to produce algal biomass and remove nutrients from
waste waters, and they compared ATS systems to that
of typical suspended microalgal production in biore-
actors. The results of this study showed that provid-
ing 3-D substrate structure, extending above the
traditional flat 2-D screen, significantly enhanced dia-
tom retention in ATS systems and therefore biomass
productivity. Enhanced productivity likely also
resulted from the increase in attachment surface and
the resulting higher density of chl and other photo-
synthetically active pigments. This was seen in the
significantly improved productivity of the finer
braided 3-D screen as compared to the coarser
variety. However, some of the enhancement also lies
in the physical disruption of laminar flow that diatom
filament strength cannot tolerate. The surface char-
acter of the 3-D fibers needs to be further investi-
gated, as major improvements in production seem
likely with this parameter.
Biomass production rates during the summer of
2010 showed a series of dips that were not charac-
teristic of ATS floways. In late August, the dips were
observed and correlated to high harvest tempera-
tures on the floways; the harvest mode was sub-
sequently shifted from late-afternoon to early-morn-
ing and biomass production dips ceased. At the first
dip, the incoming river water temperature was about
27°C. When the dips were finally prevented, the
temperature was about 29°C, falling from a peak of
nearly 34°C. Approximately, 1.5 h was required to
harvest this floway; as described earlier, test screen
sampling was carried out first, followed by a general
floway harvest. A large part of the water on the flo-
way was allowed to drain in this process to facilitate
separation of the algal biomass from the entrained
water. On very hot, sunny afternoons, the algae and
remaining water likely exceed 50°C and these tem-
peratures would be fatal to the remaining algal seed
community. The high-temperature excursions dur-
ing harvesting were likely responsible for the strong
dips in biomass production during the summer of
During the late spring and early summer of 2011,
all harvests took place in the early morning to
minimize the influence of the sun warming a drying
floway. Biomass harvests in early June 2011 on these
3-D screens had exceeded 90 g m
(Fig. 2);
TABLE 3. Season variations in total and biogenic silica in
algal turf harvested from the Great Wicomico River ATS
Harvest dates
Mean wt%/g g
Total SiO
Biogenic SiO
NovemberDecember 41.4(6.2) a
FebruaryMarch 27.2(4.0) 2.2(2.0)
MayJuly 52.1(4.5) 27.6(2.3)
FIG. 5. (a and b) Temporal variation of total carbohydrates from the algae harvested from Floway #1 (F1-diamonds), upper Floway #2
(F2U-X), and lower Floway #2 (F2L-triangles). Error bars represent the sample standard deviations of replicate analyses. (c) Temporal vari-
ation of fatty acid lipids (as FAMEs) from the algae harvested from Floway #1 (F1-diamonds), upper Floway #2 (F2U-X), and lower Floway
#2 (F2L-triangles). Error bars represent the sample standard deviations of replicate derivatizations. (d and e) Temporal variation of (top)
EPA and (bottom) DHA (as the respective FAMEs) from the algae harvested from Floway #1 (F1-diamonds), upper Floway #2 (F2U-X),
and lower Floway #2 (F2L-triangles). Error bars represent the sample standard deviations of replicate derivatizations.
the river temperature at that time was 27°C. By mid-
June, with no apparent reduction in production,
and with water temperatures approaching 30°C, sig-
nificant dips had not occurred. Mean yearly produc-
tivities on these floways were likely higher than
those we presented earlier. Based on the hypothesis
that occasional very high-temperature excursions on
the ATS floways during the summer harvests in June
and July 2010 caused artificially low measured bio-
mass production, the annual mean and seasonal
increase P(t) may be even higher than the data
(Fig. 2). For example, three harvests during the
early summer of 2011 exceeded 90 g m
Thus, the round-topped maroon curve shown in
Figure 9, and derived from equation 2 with a higher
, provided the more likely representation of the
production curve. In this case, the yearly mean pro-
duction on 3-D substrates on ATS systems in the
mid-Chesapeake Bay should be between 45 and
50 g m
and the summer peak should be
near 80 g m
. This would provide a signifi-
cant increase in the efficiency of both nutrient
removal and algal biomass production in ATS
systems employed on non-point-source waters.
3-D screens were not only more productive than
2-D screens, they also provide an algal biomass with
a higher organic and nutrient content (Table 2).
While the difference in nutrient content was not
large (Δ%P20%; Δ%N 15%), with the greater
productivity, the net nutrient removal was 39that
of the 2-D screens. With the temperature-compen-
sated productivity demonstrated, the net nutrient
removal was almost 49larger.
The use of appropriately configured 3-D sub-
strates allowed retention of diatom cells and fila-
ments in the ATS environment, with its key element
of moderated turbulence and mixing. Three-dimen-
sional substrates also allowed greater packing of
algal cells and therefore greater density of chl a
and accessory pigments. These features were the
key elements in maximizing algal photosynthesis
and productivity in photobioreactors. In planktonic
algal culture, the floating algal cells present an
inherent impediment to cell packing, produce light-
shading, and made constant exchange with the
water around those cells difficult (Adey et al. 2011).
In contrast, for cultures of benthic and other
periphytic algal culture, ATS 3-D substrates pro-
vided numerous opportunities to increase produc-
tivity, with future engineering optimizing the key
parameters that currently limit photosynthesis and
FIG. 6. Productivity of 3-D (#17) screen (bold) and combined
2-D screens (regular) as a function of temperature and light. The
surfaces were fit with Kriging correlation using Origin 8.6 graphing.
FIG. 7. Primary algal production on 3-D (#17) screen plotted
against light. The line is a linear least-squares fit.
FIG. 8. Productivity of 3-D (#17) test screen as a function of
time (estimated data have been substituted for six periods of har-
vest losses as described in the text). The line represents the
empirical productivity function given by equation 2.
When diatoms were a significant component of
biomass production on non-point-source ATS flo-
ways, production on 3-D substrates can be consider-
ably enhanced. However, if nutrients were moderate
to low in the incoming water, nutrient limitation
(including CO
limitation) was possible down the
floways (Table 1). If nitrogen removal was the target,
shortening the floway, while maintaining total floway
area, may lead to more economical removal. If phos-
phorus is the primary target, then longer floways with
nitrogen injection may allow very high pH levels, with
consequent phosphorus precipitation, providing the
most economical operation (Craggs et al. 1996).
Research has shown that the biochemically fixed
Redfield ratio can range among taxa or can change
depending on what type of macromolecules the alga
is storing, or whether the nutrient is surface
adsorbed (Geider and La Roche 2002, Sanudo-
Wilhelmy et al. 2004). As demonstrated by the signifi-
cantly higher P content in the downstream portion
of Floway #2, phosphorus precipitation was likely
occurring in the generally higher pH environment of
the low part of the floways, especially the longer
Floway #2 (Craggs et al. 1996). In any case, the lack
of correlation between C, N, P, as noted earlier, sug-
gests that the community growth was not N or P-lim-
ited, independent of carbon limitation.
Silicon dioxide is an essential nutrient for the
metabolism and growth of diatoms, whose cell walls
are composed of hydrated amorphous silica. Thus,
diatom cell growth was highly coupled to ambient
levels, as shown in several studies (Martin-
equel et al. 2000, Claquin et al. 2002). Carbon
and nitrogen levels in algae are specifically linked
with photosynthesis. Since silicate is not directly
coupled to photosynthesis, environmental factors
such as incident radiation should not directly affect
biomass SiO
levels. Although we hypothesize that
the variation in composition could be due to algal
species composition, terrigenous materials, making
up roughly half of the silica, may be more impor-
tant in the variability of silica content (Table 3).
Thus, even though the total Si content was lower in
winter when diatom relative abundance was maxi-
mum, the winter sample (23 February 2011) showed
a higher proportion of biogenic silica than the sum-
mer sample did (23 May 2011) when Chlorophyta
abundance was higher.
Being able to control the composition of the ATS
algal communities will also be important for product
application using the biomass as a feedstock. For
example, oil-based biofuels (e.g., biodiesel) required
high-lipid biomass to be economical and encouraging
diatoms would be beneficial for biodiesel application.
Specialized lipid products, such as PUFAs for nutri-
tional supplements, may require species-specific
accommodations. In contrast, for carbohydrate-based
fuels like ethanol or butanol, communities domi-
nated by Chlorophyta would likely be more useful.
Regardless of the fate of the carbon-based compo-
nent of the biomass, the nitrogen and phosphorus
that are accumulated by the algal turf from the
source water can be converted into appropriate fertil-
izer products. As demonstrated by Mulbry et al.
(2008), the organic component also delays wash-out
of nutrients, further enhancing the value of the fertil-
izer product.
One of the goals of this study was to assess the bio-
fuel potential of ATS-grown algal biomass and com-
pare it to other biofuel feedstocks. The USDA reports
recent corn harvest averages of about 973 g m
(155 bu acre
, USDA 2012), and an
ethanol fermentation yield of 8.48910
gal bu
) for corn (USDA 2010). This translates to
corn ethanol yields of 0.40 L m
. Corn is
66% starch (USDA 2010), which leads to a yeast fer-
mentation ethanol yield of 1.28 910
starch. Given the measured ~24% carbohydrate con-
tent of the biomass harvested from the Great Wicomi-
co floways, and presuming equivalent ethanol yield
from fermenting algal carbohydrates as from corn,
we predict that an ATS installation with 3-D screens
could produce about 2.24 L m
ethanol, or
about 5.6 times greater than for corn feedstock. Moreover,
even the relatively low (~4%) lipid content of the algal turf
biomass could produce 0.65Lm
of biodie-
sel, as compared to 0.055L m
from soy-
beans, assuming recent U.S. harvests of
282 g m
(42 bu acre
2012), soy is 20.6 mass% oil, and an oil-to-biodiesel
conversion efciency of 82.4 mass% (USDA 2009).
Algal turfs are thus superior to current terrestrial crops
as feedstock for biofuel production.
Algal Turf Scrubber is the only significant, non-
point-source technology for water quality amelioration
in large watersheds. However, the area requirements
of ATS are often cited as the single significant detrac-
FIG. 9. Production curves of Figure 6, modified to fit equa-
tion 2 for summer production, assuming morning harvest to
avoid lethal heating of residual floway algae during harvest. In
this mode of operation, it is estimated that yearly mean biomass
production at this location will be 47.7 g m
tion. The high biofuel production capability created
by use of 3-D substrates is critically important to ATS
selection, since in many large watersheds vast acreage
of corn and soy is planted for biofuels. Fertilization of
these two crops is a major source of nutrients and
hypoxia of waterways. Production of biofuels with
ATS, coupled with nutrient removal, is a logical
process for controlling nutrification and hypoxia.
Floway construction, operation, data collection, and algal
analyses in this study were made possible by funding from
HydroMentia, Ecological Systems Technology, The Lewis
Foundation, Smithsonian Institution, and Statoil through the
Virginia Institute of Marine Science. Chemical analyses were
supported, in part, by funding from the U.S. Department of
Energy (DE-FG3608GO88049), the Smithsonian Institution
(T08CC10068), and the Virginia Institute of Marine Science
(77263B/12683). We are grateful for discussions with Patrick
Kangas of the University of Maryland, and Emmett Duffy and
Elizabeth Canuel of the Virginia Institute of Marine Science.
We thank the staff of InterfaceFLOR for their considerable
efforts to simulate a manufactured version of the 3-D screen.
E. Adey built the two floways studied, carried out many of the
harvests, and participated in data analysis. J. Barker worked
to provide excellent manuscript production.
Adey, W., Kangas, P. & Mulbry, W. 2011. Algal Turf Scrubbing:
cleaning surface waters with solar energy while producing a
biofuel. Bioscience 61:43441.
Adey, W. & Loveland, K. 2007. Dynamic Aquaria: Building and
Restoring Living Ecosystems, 3rd ed. Academic Press, Elsevier,
508 pp.
Adey, W., Luckett, C. & Jenson, K. 1993. Phosphorus removal
from natural waters using controlled algal production. Restor.
Ecol. 1:2939.
Cade-Menun, B. & Paytan, A. 2010. Nutrient temperature and
light stress alter phosphorus and carbon forms in culture-
grown algae. Mar. Chem. 121:2736.
Carpenter, R., Hackney, J. & Adey, W. 1991. Measurements of pri-
mary productivity and nitrogenase activity of coral reef algae
in a chamber incorporating oscillatory flow. Limnol. Oceanogr.
Cequier-Sanchez, E., Rodrıguez, C., Ravelo, A. & Zarate, R. 2008.
Dichloromethane as a solvent for lipid extraction and assess-
ment of lipid classes and fatty acids from samples of differ-
ent natures. J. Agric. Food Chem. 56:4297303.
Chisti, Y. 2007. Biodiesel from microalgae. Biotechnol. Adv. 25:294
Claquin, P., Martin-Jezequel, V., Kromkamp, J., Veldhuis, M. &
Kraay, G. 2002. Uncoupling of silicon compared with carbon
and nitrogen metabolism and the role of the cell cycle in
continuous cultures of Thalassioria pseudonana (Bacillariophy-
ceae) under light, nitrogen and phosphorus control.
J. Phycol. 38:92230.
Conley, D. & Schelske, C. 1993. Potential role of spongespicules
in influencing the silicon biogeochemistry of Florida lakes.
Can. J. Fish. Aquat. Sci. 50:296302.
Craggs, R., Adey, W., Jessup, B. & Oswald, W. 1996. A controlled
stream mesocosm for tertiary treatment of sewage. Ecol. Eng.
Dick, W. & Tabatabai, M. 1977. Determination of orthophosphate
in aqueous solutions containing labile organic and inorganic
phosphorus compounds. J. Environ. Qual. 6:825.
Geider, R. & La Roche, J. 2002. Redfield revisited: variability of C:
N:P in marine microalgae and its biochemical basis. Eur. J.
Phycol. 37:117.
Hessen, D., Færøvig, P. & Andersen, T. 2002. Light, nutrients,
and P:C ratios in algae: grazer performance related to food
quality and quantity. Ecology 83:188698.
Imbs, T. I., Shevchenko, N. M., Sukhoverkhov, S. V., Semenova,
T. L., Skriptsova, A. V. & Zvyagintseva, T. N. 2009. Seasonal
variations of the composition and structural characteristics of
polysaccharides from the brown alga Costaria costata.Chem.
Nat. Compd. 45:78691.
Inhamuns, A. & Franco, M. 2008. EPA and DHA quantification in
two species of freshwater fish from Central Amazonia. Food
Chem. 107:58791.
Laughinghouse, H. D. 2012. Studies of Periphytic Algal Turf Scrubbers
along the Chesapeake Bay: Community Structure, Systematic and
Influencing Factors. PhD dissertationUniversity of Maryland,
College Park, 274 pp.
equel, V., Hildebrand, M. & Brzezinski, M. A. 2000. Sili-
con metabolism in diatoms: implications for growth. J. Phycol.
Mulbry, W., Kondrad, S. & Pizarro, C. 2008. Treatment of dairy
manure effluent using freshwater algae: algal productivity
and recovery of manure nutrients using pilot-scale algal turf
scrubbers. Bioresour. Technol. 99:813742.
Noseda, M. D., Siumara Tulio, S. & Duarte, M. E. R. 1999. Poly-
saccharides from the red seaweed Bostrychia montagnei: chemi-
cal characterization. J. Appl. Phycol. 11:3540.
Potts, T., Du, J., Paul, M., May, P., Beitle, R. & Hestekin, J. 2011.
The production of butanol from Jamaica Bay macroalgae.
Env. Progr. Sustain. Energy 31:2936.
Raven, D. & Beardall, J. 2003. Carbohydrate respiration and
metabolism in algae. In Larkum, A. W. D., Douglas, S. E. &
Raven, J. A. [Eds.] Photosynthesis in Algae. Academic Publish-
ers, Dordrecht, Kluwer, pp. 2058.
Reay, P. & Bennett, W. 1987. Determination of amorphous silica
and total silica in plant materials. Anal. Chim. Acta 198:14552.
Sandefur, H., Matlock, M., Costello, T., Adey, W. & Laughing-
house, H. D. IV 2011. Seasonal productivity of a periphytic
algal community for biofuel feedstock generation and nutri-
ent treatment. Ecol. Eng. 37:147680.
Sanudo-Wilhelmy, S., Tovar-Sanchez, A., Fu, F., Capone, D.,
Carpenter, E. & Hutchins, D. 2004. The impact of surface-
adsorbed phosphorus on phytoplankton Redfield stoichiome-
try. Nature 432:897901.
Singh, A. & Olsen, S. I. 2011. A critical review of biochemical
conversion, sustainability and life cycle assessment of algal
biofuels. Appl. Energy 88:354855.
Skriptsova, A. V., Shevchenko, N. M., Zvyagintseva, T. N. & Imbs,
T. I. 2010. Monthly changes in the content and monosaccha-
ride composition of fucoidan from Undaria pinnatifida (Lami-
nariales, Phaeophyta). J. Appl. Phycol. 22:7986.
Stewart, A. 2006. Preliminary Engineering Assessment for a Comprehen-
sive Algal Turf Scrubber Based Nutrient Control Program for the Su-
wannee River in Florida. Suwannee River Water Management
District, Live Oak, FL, 18 pp. 18 Sheets.
USDA. 2009. Energy life-cycle assessment of soybean biodiesel.
(Agricultural Economic Report Number 845). Available at:
Biodiesel91409.pdf (last accessed 05 December 2012).
USDA. 2010. 2008 Energy balance for the corn-ethanol industry.
(Agricultural Economic Report Number 846). Available at:
final.pdf (last accessed 05 December 2012).
USDA. 2012. National Agricultural Statistics Service: Charts and
Maps Field Crops. Available at:
Charts_and_Maps/Field_Crops/index.asp (last accessed 07
November 2012).
Utermohl, H. 1958. Zur Vervollkommnung der quantitativen Phy-
toplankton-Methodik. Mitt. Int. Ver. Limnol. 9:138.
Wang, B. & Lan, C. 2011. Biomass production and nitrogen and
phosphorus removal by green alga Neochloris oleobundans in
simulated wastewater and secondary municipal wastewater
effluents. Bioresour. Technol. 102:563944.
Supporting Information
Additional Supporting Information may be
found in the online version of this article at the
publisher’s web site:
Appendix S1. Sources and preparation of the
materials, reagents and samples used for chemical
analysis of the ATS biomass.
Figure S1. Surging device at the input of Flo-
way #1. This dump-trough, mounted on a single
nylatron rod axis, fills with water and dumps
approximately every 30 s, providing an oscillatory
flow in the ATS floway. Note the dense buildup
of diatom biomass (primarily Berkeleya) on the flo-
Figure S2. Floway #1 during the spring of 2011.
At this time a mix of Ulva intestinalis,Berkeleya ruti-
lans, and Ulothrix sp. dominated the floway with
dense, stringy masses.
Figure S3.Hand-woven 3-D growth substrate
(vertical, braided fibers attached at the nodes of a
5 mm mesh, spiral-wound, basal screen) overlying
the “standard” 2-D, 2 95 mm mesh screen.
Figure S4. Temperature of the Great Wicomico
River at the ATS water intakes, and the discharge
temperature of Floway #1.
... g DW m − 2 d − 1 ) achieved in this study at HWV of 0.04-0.6 m s − 1 over the three-month NZ summer period are comparable to those (from 5 to 25 g DW m − 2 d − 1 ) previously reported for Algal Turf Scrubbers treating surface waters such as polluted stream, rivers and agricultural drainage (Sandefur et al., 2011;Adey et al., 2013;Kangas and Mulbry, 2014;Kangas et al., 2017;Yan et al., 2018). In particular, the biomass productivity of the FANS operated with ACT <2.5 min (~11.7 g DW m − 2 d − 1 , Table 2) in this study is comparable to that (14.3 ± 2.0 g DW m − 2 d − 1 in July-Aug 2008 and 10.0 ± 1.0 g DW m − 2 d − 1 in July-Aug 2009) found for a pilot-scale ATS (91 m long × 0.3 m width; HWV: 0.26 m s − 1 ; ACT: 5.8 min calculated based on an estimated the same 7 mm water depth of this study) treating water from the Susquehanna River, Pennsylvania, USA (Kangas et al. (2017). ...
... The relatively low organic matter (or high ash) content in the biomass is likely due to the dominance of diatoms and the resulting high silica (SiO 2 ) content of the algal biomass, as has been shown in previous ATS studies. Adey et al. (2013) reported that the silica content in the algal turf (harvested from ATS operated near Great Wicomico river near Chesapeake bay, USA) ranged from 30% to 50% of total biomass. Another cause of the high ash content of the harvested biomass could be due to filtration of fine colloidal inorganic particles contained in the river water (e.g., inert sediment particles such as fine sand, silt and clay particles). ...
... The total nitrogen content of the dried FANS harvested algal biomass was unaffected by the ACT and was typically ~2%, which is comparable to literature values (1.5-3.0%) for ATS biomass (Adey et al., 2013;Kangas and Mulbry, 2014;Kangas et al., 2017;Yan et al., 2018). This indicates that up to 0.24 g N m − 2 d − 1 of the river water nitrogen was removed by algal assimilation followed by algal harvest ( Table 2). ...
We investigated the effect of algal contact time (ACT) and horizontal water velocity (HWV) on the performance of pilot-scale Filamentous Algae Nutrient Scrubbers (FANS) treating river water during the NZ summer. The FANS floways were seeded with a mixture of four New Zealand native filamentous algal species (Oedogonium sp., Cladophora sp., Rhizoclonium sp., and Spirogyra sp.) and allowed to establish over one month. River water was pumped onto the top of each FANS at different flow rates (2, 4 or 8 L min⁻¹) to give ACTs from 0.6 to 10.1 min depending on FANS length (6–24 m) and HWV from 0.04 to 0.16 m s⁻¹. FANS inflow and final outflows were monitored three times a week for nitrate and DRP concentrations and FANS algal biomass was harvested weekly. Average biomass productivity was significantly higher on the FANS with shorter ACT. For example, biomass productivity of the 24 m length FANS with 2.5 min ACT were 67% higher (11.2 g DW m⁻² d⁻¹) than that with four times the ACT (10.1 min). Irrespective of the HWV the biomass productivity declined down the length of the floways (with longer ACT) and the decline was greater at lower HWV. The decreased biomass productivity at lower HWV (and/or higher ACT) was likely attributable to the daytime carbon limitation of photosynthesis (at pH > 9.5) and heat stress with elevated daytime water temperature (at >30 °C). Despite the short ACT (<10.1 min) the single pass pilot-scale FANS effectively removed both nitrate-N and DRP from the river water, with >35% removal of both NO3–N (from 0.49 to <0.32 mg N L⁻¹) and DRP (from 0.14 to <0.09 mg P L⁻¹). Both the nitrogen and phosphorus content of the harvested algal biomass were unaffected by both HWV and ACT and typical (N: ∼2.0%; P: 0.2–0.3%) of the literature values (N: 1.5–3.0%; P: 0.15–0.32%). Compared with constructed wetland nutrient removal (0.1 g N m⁻² d⁻¹; 0.08 g P m⁻² d⁻¹), the FANS achieved up to 2.5-fold higher nitrogen removal (0.24 N m⁻² d⁻¹) through algal nitrogen assimilation followed by subsequent algal harvest and up to 4-fold higher phosphorus removal (0.34 g P m⁻² d⁻¹) through a combination of algal phosphorus assimilation and some P-precipitation under photosynthesis-mediated elevated daytime pH levels (pH > 9.0). This research indicates that FANS have the potential to require less than half the land area of constructed wetlands for the same level of nitrogen removal and that they require only a few weeks to establish to achieve full performance. Moreover, FANS have the further benefit of resource recovery for beneficial re-use of harvested algal biomass for animal feed, fertiliser, or biofuel.
... 115 Typically, attachedalgae systems are applied to water treatment operations and focused on autotrophic and/or mixotrophic growth conditions. [115][116][117] However, because of the inherent mixed-culture algae community dynamics, 117 there are challenges associated with controlling both the types of algae that appear, the stability of the community and the resulting harvested biomass quality. In most instances, the biomass will be high ash (>50% of the harvested material), with varying proportions of biogenic and abiogenic ash. ...
... 115 Typically, attachedalgae systems are applied to water treatment operations and focused on autotrophic and/or mixotrophic growth conditions. [115][116][117] However, because of the inherent mixed-culture algae community dynamics, 117 there are challenges associated with controlling both the types of algae that appear, the stability of the community and the resulting harvested biomass quality. In most instances, the biomass will be high ash (>50% of the harvested material), with varying proportions of biogenic and abiogenic ash. ...
... The nitrogen and thus protein composition will vary, though as long as nutrients are not limiting, protein will be the dominant fraction of the organic content. 117 ...
High-protein algal biomass is an important bio-commodity that has the potential to provide a new source of sustainable protein products. Herein is a critical review that identifies 1) the most relevant sustainability findings related to the processing of proteinaceous algal biomass to higher value protein products and 2) the potential pathways to improve life cycle assessment (LCA) and techno-economic analysis (TEA) metrics, including life-cycle carbon dioxide equivalent (CO 2 eq), life cycle energy, and minimum selling price (MSP) of these products. The critical review of the literature revealed a large variation in model input parameters relating to these metrics. Therefore, a Monte Carlo analysis was conducted to assess the risk associated with these input variations. To understand the uncertainties that propagate into high-protein algae to products' systems, we reviewed more than 20 state-of-the-art unit operations for algal biomass processing., including cell disruption, protein solubilization, protein precipitation and purification, and protein concentration. We evaluated displacement of proteinaceous products by algal-bioproducts, including ruminant feed, aquaculture feed, protein tablets, and biopolymers and biopolyesters, with prices in the market ranging from 1.9 to 120 $ kg-1 protein. This review realized that the MSP of ruminant and non-ruminant feed ranges from 0.65±0.56 to 2.9±1.1 $ kg-1 protein, and bioplastics' MSP ranges from 0.97 to 7.0 $ kg-1 protein. Regarding LCA metrics, there is limited research on life cycle energy in proteinaceous biomass concentration and bioproduct systems, reported at 32.7 MJ kg protein-1, for animal feed displacement. Animal feed emissions in the literature report negative fluxes, representing environmental benefits, as low as-3.7 kgCO 2 eq kg-1 protein and positive fluxes, i.e., global warming potential, as high as 12.8 kgCO 2 eq kg-1 protein. There is limited research on bioplastics life cycle emissions reported at 0.6 kgCO 2 eq kg-1 protein. In general, the studies to date of algae-derived protein bioproducts showed similar life cycle emissions to soybean meals, nylon, polymers, and polystyrenes. Our risk analysis realized that more than 50% of scenarios can result in negative-net life cycle CO 2 eq emissions. This review and risk analysis assess and demonstrate the scenarios that improve economic and environmental sustainability metrics in high-protein algal bioproducts systems.
... Filamentous Algae Nutrient Scrubbers (FANS) and similar algae turf scrubbers (ATS) are ecologically engineered, artificial streams that grow attached filamentous algae and associated periphyton to treat polluted water (Craggs 2001;Adey et al. 2011Adey et al. , 2013Sutherland and Craggs 2017). FANS consist of a shallow, gently sloped floway with a liner and sometimes an overlying screen to which the algal "turf" attaches. ...
... FANS typically grow a mixed-species assemblage of filamentous algae (Mulbry et al. 2008a;Adey et al. 2013;D'Aiuto et al. 2015;Sutherland et al. 2020). The algae are established by either letting them naturally colonise the floway or seeding the floway surface with mixed algal species collected from nearby water bodies. ...
Full-text available
Filamentous algae nutrient scrubbers (FANS) have demonstrated potential for cost-effective and sustainable nutrient bioremediation of a wide range of wastewaters. Typically, FANS are seeded with a mixed assemblage of algae species, however, growing a monoculture of one species on FANS could facilitate biomass use by providing a more consistent and high-quality substrate for end-product applications. To date, a standardised bioassay to assess the productivity and nutrient removal of filamentous algae attached to a bottom substrate (that could help identify promising species for FANS monoculture) has not been developed. Therefore, we developed a microscale filamentous algae nutrient scrubber (µFANS) and a protocol to establish monocultures of freshwater filamentous algae to compare performance in terms of attachment capability, nutrient removal and biomass production. Four common filamentous algae species (Cladophora sp., Oedogonium sp., Rhizoclonium sp. and Spirogyra sp.) were seeded by evenly distributing and rubbing the biomass onto µFANS textured liner to “hook” algal filaments, providing initial physical attachment. Within 14 days, a “lawn” of the seeded algae had established and the “hooked” biomass had attached biologically. Depending on species, biological attachment resulted from either holdfast development from filaments that grew from settled zoospores, growth of rhizoids or adhesion of filament fragments to mucilage. Biomass productivity of each species ranged from 2.2 to 5.3 g DW m⁻² day⁻¹ while nutrient removal rates ranged from 8.8 to 28.4 mg NO3 g⁻¹ DW day⁻¹ and 2.2 to 8.1 mg PO4 g⁻¹ DW day⁻¹. Oedogonium sp. was the best performing species overall, with the strongest holdfast attachment, high biomass productivity (mean 4.2 g DW m⁻² day⁻¹) and high nutrient removal rates (mean 21.8 mg NO3 g⁻¹ DW day⁻¹; 5.6 mg PO4 g⁻¹ DW day⁻¹).
... Some systems are presented in Fig. 1-11. Algal footprint or surface productivities in those systems under different conditions are also displayed in Table 1 Microfluidic flow cell system (Nielsen et al., 2011), c) Flat plate parallel horizontal photobioreactor system (Schnurr et al., 2013), d) Algal turf scrubber system (Adey et al., 2013), e) Revolving algal biofilm system (Gross and Wen, 2014), f) and g) Rotating algal biofilm reactor (Bernstein et al., 2014), h) Algadisk lab scale reactor (Blanken et al., 2014), i) and j) Twin-layer system (Schultze et al., 2015). ...
... show an excellent performance in terms of biomass productivity (Blanken et al., 2017; 28 B -Microalgae cultivation systems Huang et al., 2016;Li et al., 2017). For instance, an ATS system was yearly employed by Adey et al. (2013) to examine the effect of support material on algal productivity. Results of this work reported productivity as high as 60 -70 g m -2 d -1 in August and September, and a yearly average productivity of 36.9 g m -2 d -1 . ...
Microalgae biofilm-based technology has been pointed out as a promising alternative to suspended cells cultivation. However, a better understanding of the impact of process operational factors on biofilm development and productivity is actually needed to fully confirm the potential of biofilm-based processes for biomass/compounds production at large-scale. This will help identifying stressful factors and optimizing this kind of processes, and it will also provide new knowledge on immobilized cell behavior, which remains poorly understood.The aim of this thesis is to assess the effect of operational factors like light, support and inoculum on growth, productivity, and physiological properties of sessile cells during biofilm development. To do so, C. vulgaris was immobilized on porous materials including fabric supports and membranes. Immobilized cell behavior (growth, composition, photosynthetic activity) was investigated under different conditions of light intensity, supports and inoculum density /physiology. In a continuously submerged reactor, cells acclimated within only 3 days to the new biofilm lifestyle. Low productivities were though obtained, which may be related to high detachment or support properties. Consequently, cell retention capability on five fabrics was tested afterwards in order to find promising supports. Accordingly, Terrazzo with the smallest pore size and mesh density exhibited the highest cell retention capability, representing thus a potential support for large-scale biofilm cultivation. Immobilized cells distribution on fabrics, biofilm development and activity were found to be strongly affected by the support material. Another biofilm-based system (perfused system) where cells are immobilized on a membrane was then used to study the combined effect of light intensity and inoculum density on biofilm developpement under more controlled conditions. Results showed that high cell density affects negatively the productivity on fabrics and membranes which may be linked to limitation in light and nutrients. Combined conditions of light and cell density which promote either biomass or lipids production were also identified. Inoculum acclimation to 350 and 500 µmol m-2 s-1 benefited biomass production and avoided photo-inhibition, respectively, suggesting the importance of considering photo-acclimation on biofilm-based systems set-up. For the first time, the acclimation process of cells switching from planktonic to the immobilized state was lastly studied. Unexpectedly, within the first hours, cells increased their size, accumulating carbohydrates, and decreasing the Chl a content. It is likely that changes in environmental conditions trigger this behavior, allowing the cells to acclimate to the new lifestyle.
... Developing new bioproducts from microalgae is a suitable alternative for this use. In addition, microalgae biomass production in an ATS system is potentially economically viable [10]. It is a simple system, such as the 1 m ramp pilot ATS system installed by Martini [11] in a lake known to have eutrophicated waters. ...
Full-text available
The application of algal turf scrubber (ATS) systems for simultaneous bioremediation and biomass valorization can promote the mitigation of nitrogen, phosphorus, and organic contaminants in water bodies, helping reduce both the environmental impacts and the costs related to bioactive compound production. The present study performed a life cycle assessment (LCA) based on the production of pigments from the periphytic biomass at a pilot-scale ATS system (SISGEN register A256CC3). For the LCA, the functional unit was defined as 10 kg of pigments, and 10 years was considered the reference flow. The Ecoinvent 3 database and IMPACT 2002 + (V2.10) were used in SimaPro software (version 8.04) with data from the ATS system cultivation (construction and operation), drying, and pigment extraction. As a result, the damages and impacts of most categories were mainly related to the drying step, followed by extraction. However, even when considering a permanent demand for electricity (for the hydraulic pump), the operation of the ATS system presented fewer impacts in its construction and operation than traditional cultivation systems. Additionally, applying an ATS system to produce biomass for pigment extraction had low impacts on most categories, mainly due to water use from a catchment lake. New scenarios are proposed to minimize impacts, indicating that the ATS system with solar drying can be an interesting option to produce pigments from periphytic biomass simultaneously and promote polluted lake bioremediation. Graphical Abstract
... The biofilm is a mesocosm of benthic bacteria, proand eukaryotic microalgae, and fungi embedded in their extracellular matrix (Fig. 1). Through complex mutual interactions and synergistic effects among the different species, the algal-bacterial biofilm can sequester a wide range of nutrients, for example, N, P, and K [14]. Thus, ATS systems are studied as alternative biological wastewater treatment technologies [15,16], but so far, they have not been systematically evaluated as substrates for biogas production by anaerobic digestion. ...
Full-text available
This study investigated the anaerobic digestion of an algal–bacterial biofilm grown in artificial wastewater in an Algal Turf Scrubber (ATS). The ATS system was located in a greenhouse (50°54′19ʺN, 6°24′55ʺE, Germany) and was exposed to seasonal conditions during the experiment period. The methane (CH 4 ) potential of untreated algal–bacterial biofilm (UAB) and thermally pretreated biofilm (PAB) using different microbial inocula was determined by anaerobic batch fermentation. Methane productivity of UAB differed significantly between microbial inocula of digested wastepaper, a mixture of manure and maize silage, anaerobic sewage sludge, and percolated green waste. UAB using sewage sludge as inoculum showed the highest methane productivity. The share of methane in biogas was dependent on inoculum. Using PAB, a strong positive impact on methane productivity was identified for the digested wastepaper (116.4%) and a mixture of manure and maize silage (107.4%) inocula. By contrast, the methane yield was significantly reduced for the digested anaerobic sewage sludge (50.6%) and percolated green waste (43.5%) inocula. To further evaluate the potential of algal–bacterial biofilm for biogas production in wastewater treatment and biogas plants in a circular bioeconomy, scale-up calculations were conducted. It was found that a 0.116 km ² ATS would be required in an average municipal wastewater treatment plant which can be viewed as problematic in terms of space consumption. However, a substantial amount of energy surplus (4.7–12.5 MWh a ⁻¹ ) can be gained through the addition of algal–bacterial biomass to the anaerobic digester of a municipal wastewater treatment plant. Wastewater treatment and subsequent energy production through algae show dominancy over conventional technologies. Graphical abstract
... Optimum cultivation temperature highly depends on the algae species present (algal community composition) [36][37][38]. For outdoor periphytonbased systems, community composition and productivity changes are often observed because of the seasonal light and temperature variations [39]. Light intensity and temperature are usually a set of correlated parameters, whether in outdoor or indoor cultivation systems, with higher light levels being associated with increased temperatures. ...
Attached algal cultivation systems come with the promise of improving yield, efficiency, and consistency of algae-based/algal biomass products and processes. Immobilized biomass is inherently more concentrated than the biomass obtained from planktonic cultivation. This can make harvesting and downstream processing more feasible. The wastewater treatment capabilities of the attached systems are well established. However, the efficiency and applications of the attached systems could be enhanced by developing the fundamental understanding of algae-substrate interactions needed for the rational design of cultivation systems that maximize pollutant removal, biomass product production, or both. Yet, algae-substrate interactions have been rarely studied in periphyton-based attached systems. There is a need for the development of reactors and methods that enable the investigation of substrate effects at an intermediate scale that bridges fundamental periphyton studies (in natural systems) and large-scale application-focused studies on systems like algal turf scrubbers (ATS). Indoor, intermediate-scale flow ways can provide a balance between these two extremes. Still, close attention needs to be paid to the influence of design and growth factors/parameters such as inoculation method, light exposure, flow conditions, pH, and nutrient types and concentrations. In this work, we present a design for a multi-channel indoor flow way photobioreactor alongside an assessment of how key design and growth parameters affect the attached cultivation of the periphytic filamentous green algae Stigeoclonium tenue on polymeric substrates. The results from this research highlight that carefully designed indoor flow way reactors can overcome the limitations of lab-scale photobioreactors in studying algae-substrate interactions in terms of repetition, flow uniformity, and cost-effectiveness.
... The diatom strain Haslea ostrearia, grown in immobilized cell photobioreactor using agar gel layers, yielded a twofold higher cell growth [32]. Earlier studies on algae production in the biofilm cultivation system showed dry biomass productivity as high as [33][34][35]. The productivity reported in such an ecologically engineered system is five times more than that achieved in conventional microalgal cultivation (open pond/raceway pond) systems. ...
The rapidly depleting fossil fuel reserves with rising greenhouse gas levels (GHGs) in the atmosphere necessitate exploring alternate sustainable energy options. Biofuels from microalgae are emerging as a viable renewable energy resource owing to their inherent characteristics of higher biomass and lipid yield per hectare compared to other terrestrial bioenergy feedstocks. In this context, the present communication highlights the prospects of microalgal biofuel and other value-added products produced in a decentralized microalgal biorefinery in the flood plains (gazani lands) of the west coast of India. The spatial extent of potential sites for diatom cultivation estimated in three districts along the Indian west coast was 1940 ha. The opportunities for establishing biorefineries using diatoms as renewable bioenergy feedstocks were investigated through species prioritization, seasonal availability, tolerance, and biochemical composition analyses. Nitzschia and Amphora sp. were prioritized for lab-scale productivity studies based on their tolerance and macromolecular composition. When cultivated in a prototype biofilms-based bioreactor designed using gravel stones as substrates, Amphora sp. Yielded 16 times more productivity (0.56 g L⁻¹) than conventional shake flask cultures. Design of a diatom biorefinery and its mass budgeting considering 100 kg dry biomass yielded ∼15–24 kg of biodiesel. Techno-economic assessment of biodiesel with value-added products of glycerol, biogas, and biofertilizer demonstrated a biodiesel production cost of 30.08–59.52 INR/kg of biodiesel. Harvesting cost in a hybrid mode using mechanized scrubbers and manual labour was estimated as 20 INR/kg of biomass.
Diminishing fossil fuel reserves, escalating oil prices, higher greenhouse gas footprint, and the adverse effects of climate change have propelled increased research focus during the twenty-first century on sustainable renewable energy transitions. Algal biofuel is gaining global interest due to its potential to convert biomass into a range of bioenergies and other value-added products. Life Cycle Assessment (LCA) aids in quantifying the environmental benefits of algal biodiesel over plant-oil-based biodiesel. The present chapter presents the environmental footprint of biodiesel from microalgae. It is compared with the terrestrial oil yielding feedstocks of first- (palm) and second-generation (Jatropha)-derived biodiesel by considering the mass and energy of production processes starting from feedstock generation to biodiesel production (cradle to gate analysis). The lifecycle impact of different generations of the biodiesel was assessed using OpenLCA software to understand the potential health and environmental implications (GHG, etc.)/soundness. Process-wise energy expenditure estimation shows a 68% and 45% reduction in energy expenditure and GHG emissions in algal biodiesel compared to first- (palm) and second-generation (jatropha) biodiesel, respectively. Results also reveasled a GHG mitigation potential in terms of direct GHG emission savings of 84, 90, and 95% for palm, Jatropha, and microalgal biodiesel compared to conventional fossil diesel.
Full-text available
The Salton Sea in the Imperial Valley of California is a threatened ecosystem. To address the challenges related to trace element accumulation and eutrophication, a 270 m long attached periphytic algae flow-way was deployed at the Alamo River Wetlands, a tributary of the Salton Sea. To assess opportunities for efficient generation of biomass from reclamation of run-off-derived nutrients, the quantities of available and biosorpbed nutrients, organics, and trace elements were monitored. Over the course of 2 years, persistent algal biomass production was achieved at an average ash-free biomass productivity of 5.8 ± 2.7 g/m²/day for the full flow-way length. Overall, harvested biomass consisted of high ash (76.3 ± 5.4%) and low lipid (1.1 ± 1.0%) content, which are typical of periphytic algal biomass. Nitrogen removal rates of 530 ± 190 mg N/m²/day, phosphorous removal rates of 14 ± 6 mg P/m²/day, and 73 ± 25% of BOD (Biological Oxygen Demand) removal rates were achieved. Furthermore, spatial variations of the biomass productivity along with nutrient removal rates were observed to have a decreasing trend over the length of the flow-way, while N and P contents of the biomass showed increasing trend, indicating variable nutrient utilization efficiency as an important factor for system scaling. Temperature and solar irradiation were found to be key environmental factors for biomass productivity and nutrient removal rates. However, stable uptake of nutrients, organics, and metals in the biomass, despite intermittent variation of the analyte concentrations in the source water, indicate the resilience of attached periphytic algae biomass production at dilute nutrient concentration regimes. Trace metal analysis of the water from the surrounding area revealed levels exceeding federal toxicity guidelines for Selenium and Copper. Significant bioaccumulation of these and other metals contaminants were also identified in the harvested biomass, including Nickel, Chromium, Cadmium and Lead. Together these findings demonstrate several potential value propositions for attached algae cultivation from agricultural runoff-impacted surface waters, including remediation of N/P nutrients, organics, and common toxic metals, concomitant with biomass production.
Full-text available
As human populations have expanded, Earth's atmosphere and natural waters have become dumps for agricultural and industrial wastes. Remediation methods of the last half century have been largely unsuccessful. In many US watersheds, surface waters are eutrophic, and coastal water bodies, such as the Chesapeake Bay and the Gulf of Mexico, have become increasingly hypoxic. The algal turf scrubber (ATS) is an engineered system for flowing pulsed wastewaters over sloping surfaces with attached, naturally seeded filamentous algae. This treatment has been demonstrated for tertiary sewage, farm wastes, streams, and large aquaculture systems; rates as large as 40 million to 80 million liters per day (lpd) are routine. Whole-river-cleaning systems of 12 billion lpd are in development. The algal biomass, produced at rates 5 to 10 times those of other types of land-based agriculture, can be fermented, and significant research and development efforts to produce ethanol, butanol, and methane are under way. Unlike with algal photobioreactor systems, the cost of producing biofuels from the cleaning of wastewaters by ATS can be quite low.
Full-text available
A simple and precise colorimetric method of determining orthophosphate in aqueous solutions containing labile organic and inorganic P compounds is described. It involves a rapid formation of molybdenum blue color by the reaction of orthophosphate with molybdate ions in the presence of ascorbic acid trichloroacetic acid and citrate arsenite reagents and complexation of the excess molybdate ions to prevent further formation of blue color from the phosphate derived from hydrolysis of the acid labile P compounds. The color is stable up to 24 hr. The method is sensitive and accurate, and it permits determination of microgram quantities of orthophosphate in samples containing large amounts of acid labile P compounds. Tests with a wide range of condensed phosphate and organic phosphate compounds showed that none of the P compounds studied interfered with this method. Results by this method are compared with those obtained by the method of Murphy and Riley.
Bostrychia montagnei was submitted to aqueous extraction at 25 and 85 ^°C. The purified polysaccharide extracts represent ∼ 17% of the dried alga. Galactose is the principal monosaccharide component of these extracts (60.8–70.4 mol%). 3,6-Anhydrogalactose and its 2- O-methyl derivative are also present in smaller amounts (16.2–22.0 mol%), as well as other methylated sugars, such as 6- O- (6.5–7.8 mol%) and 2-O-methylgalactose (0.2–2.1 mol%). Xylose (4.1–8.1 mol%) and glucose (0.7–2.6 mol%) were also detected. The aqueous extracted polysaccharides (25 ^°C) were separated by anion-exchange chromatography into six sulphated galactan fractions with negative specific rotations and another two with high xylose contents and positive specific rotations. The sulphated galactans all have an agar type backbone modified by partial O-methyl substitution on O-6 or O-2 of the galactosyl units. The latter substitution is also present in varying degrees of 3,6-anhydrogalactose.
This study ascertained the technical potential of producing biofuel from a naturally occurring macroalgae. The algae examined grow in Jamaica Bay, New York City, on water containing nitrates, phosphates, and carbon dioxide that comes from the atmosphere. The process consisted of manual and mechanical harvesting, drying, grinding, and subjecting the algal matter to acid hydrolysis to extract carbohydrates to form an algal sugar solution. Fermentation of that solution to butanol was performed with butanol ultimately removed by distillation. An average of 15.2 g/L of reducing sugars was extracted in the hydrolysate showing that macroalgae (Ulva lactuca) have significant usable carbohydrates after hydrolysis. It was found necessary to remove the excess solids from the hydrolysate prior to fermentation, as the productivity fell by 75% if this was not done. With the bacterial strains (Clostridium beijerinckii and C. saccharoperbutylacetonicum) and the algal sugar solutions used, an acetone butanol ethanol (ABE) fermentation was used to make butanol. The butanol concentration in the fermentation broth reached about 4 g/L, which is close to the theoretical value for the sugar concentration obtained, and compares well (when adjusted for sugar concentration in the media) with values reported in the literature for other systems. The recovery of reducing sugars in the media during the pilot study was 0.29 g butanol/g sugar. © 2011 American Institute of Chemical Engineers Environ Prog, 2012
Amorphous silica in plant material was dissolved with potassium carbonate solution, after nitric acid/hydrogen peroxide digestion. Total silica in plant material and some minerals was obtained after ashing and fusing in nickel crucibles with potassium hydroxide containing potassium tetraborate/nitrate. The relative standard deviation for determinations of silicon by these methods in plant material was <2% for plant material containing 2–2100 μmol Si g−1 (dry weight).
Algal biomass is a promising feedstock for biofuel production. With a high lipid content and high rate of production, algae can produce more oil on less land than traditional bioenergy crops. Algal communities can also be used to remove nutrients from impacted waters. The purpose of this study was to demonstrate the ability of an algal turf scrubber (ATS)™ to facilitate the growth of periphytic algal communities for the production of biomass feedstock and the removal of nutrients from a local stream. A pilot-scale ATS was implemented in Springdale, AR, and operated over the course of a nine-month sampling period. System productivity over the nine-month operating time averaged 26gm−2d−1. Total phosphorus and total nitrogen removal averaged 48% and 13%, respectively. The system showed potential for biomass generation and nutrient removal across three seasons.