RECYCLING TREATED SEWAGE THROUGH CYPRESS
WETLANDS IN FLORIDA
H.I. Odum, K.C. Ewel, W.j, Mirsch, and jW, Ordway
Studies supported by
Rockefeller Foundation and RANN Division of the
National Science Foundation through a joint project;
Cypress Wetlands for W -r, i M.'I..4, niir, Recycling, and
Conservation. Paper presented at "Rockefeller Conlerence
on Waste Recycling on Land" at Bellagio, Italy, July, 1975.
Occasional Publication No,. 1
Center for Wetlands
University of Florida
Gainesvillc 3261 1
RECYCLING TREATED SEWAGE THROUGH CYPRESS WETLANDS IN FLORIDA
H.T. Odum*, K.C. Ewel:, W.J. Mitsch*, and
Center for Wetlands
ULlik-ersi ,' of Florida
*DtepdrtmenC of Ervirunmenial Engineering Sciences
:5~,...1 of Forest Resources and Conservation
I't r..,'. in.' of Clvil Engineering
i t * - Surface -
__ Freshwater Aqulfr
Fig. 1 Generalized profile of a cypress dome ecosystem
showing ....n .jundin. pinelands and wastewater
A natural water management system exists in Florida:
its cypress wetlands catch and hold excessive rains, letting
them percolate slowly to the sr.Pujnd water. The needles
drop in the dry season and the dense trunk biomass shades
the waters, reducing water loss during this critical time. A
test of the feasibility of 'c, cLi',' treated sewage through
cypress wetlands is in its second year in G.hin;i. Ill. Florida
(see Fig. 1}. Wastewater from secondary sewage treatment
plant at a trailer park is being routed into two of the
cypress domes which are found in large numbers in many of
the counties of Florida (40,000 in some counties). A dome
is a r..u4-.hl, circular cypress swamp, 1 25 acres in size,
occupying a saucer-shaped depression which receives water
from the surrounding higher ground. The trees are tallest in
the center and shorter at the sides i. .- an inverted bowl, or
One of the domes receiving waste was severely burned
prior to flow, and the other is typical of the wet ponds that
have only been burned -1 *i.l '. A normal acid-water pond
is being used as a control, while another burned cypress
dome is receiving hard _r ..,iiJ water, such as is now found
in many cypress ponds in disturbed areas in south Florida.
The burned, unburned, soft water, hard water, and
wastewater variations represent major classes of conditions
in Florida. Phases of research are measuring uptake of
nutrients by the components of the i'c.I.,1. Tim, heavy metal
concentrations, changes in ,iiinL matter, microbial
concentrations, tree growth, changes in ecological commu-
nlties, microclimate, mosquito populations, and econorric
potentials in stimulating this type of land use as a
replacement for tertiary treatment (Odum and Ewel, 1974).
Figure 2 is an aerial photograph of the experimental
area six years prior to the study. The dark, isolated areas
arc cypress domes; the light areas are pine plantaLions. This
Fig. 2 Aerial view of cypress domes in pine plantations
before forest fire or sewage disposal.
STALW-.i4 l \
1IIS1LU SE*&AE _ /
TGBcN wTE T I I
TQ'TfA NITRmEtN tsnCETrRAT ION, rrq-N11
TOTAL PHOSPHORUS CONCENTRATIONI, mn9-P/t
Fig. 5 Nutrient concernirations of created s- ,,L , standing
wa:er, and groundwater of cypress dome receiving
sewage, I little rr no surface outflow is assumed
at an application rate of one inch per week,
Groundwater percolaiin values are from a well 3 m
directly below the dome while surface groundwater
values ire from a shallow well 50 m downstream
of Ihe dome,
f;- 6 Coliform budget (total and fecal) for cypress dome
receiving treated sewage (U. Price and J. Fox).
55 xii" I FECAL
1? ' . C ..,, TAi 104 *-W TTAL [
\STAI.W� CR 681 X P07 FEC-L --1
I29 X 0 - TVTAL - V9
5T11 i x id' FECEL
Se X 10" FECAL - 9l% A54K
COLIFORM BuDGET, # COLIFORMS/MONTH
week in rainfall and two inches in runoff, and experimental
domes receive an additional inch per week of wastewater or
groundwater. The dorre has some of the properties of a
giant pressure lilter: one inch per week is estimated to pass
from the pool in the dome to the groundwater.
The diagram (Fig. 5) shows the nutrient concen-
trations in the wastewater fhi..'.g.; into the dome, in the
standing water, and in the groundwaters emerging .in.
under the dome. Pi-..lniin.ii. indications are that most .f
the nutrients are being taken out or the waters, although we
don't know yet whether they are being routed from
sediments to increased tree :.. i',. Sediments below domes
are calcareous and may be pJ'..-.Wir, phosphorus from
rLr,...l i;ng waters by differential precipitation, as shown
by Gilliland (1973) lor Florida ecosystems in general,
Snii uI,, changes in coliform bacteria counts are shown
in I-"'., 6, Some leakage of sewage around wells drilled into
the first dome resulted in detectable bacterial concen-
trations in the groundwater, no wells or other deep
installations were put into the second dome, preventing this
C... .i' ,.]' e studies by &illunc, et al., (1975) showed
leakage of viruses as well into the groundwater. So far, none
have appeared in the groundwater wells surrounding the
unburned dome that has not been perforated.
I he main features of the cypress dome ecosystem are
shown in Fig. 7. The energy circuit symbols used (Odum,
1 ''1) imply *r*.: Ifjc mathematical relationships. The major
features shown include the autotrophic compartments of
cypress, pine, and understory. Nutrients, water, and organic
peat are important components in the sediments.
Interactions with the outside include fire, logging, drainage,
and nutrient loading.
The ecosystems with normal, acid water are fairly
diversified with some aquatic plant growth; productivity in
the pond is particularly high in the early spring before
cypress leaves have reappeared. Dissolved oxygen is about
one-fourth of saturation. After fire, a floating algal cover
and then a duckweed ,.'." i1i appeared in the two burned
domes, presumably stimulated by the released nutrients.
The burned dome receiving treated sewage, however,
retained a solid duckweed cover, causing near tero oxygen
levels in the water. The duckweed in the burned
groundwater dome decreased to scattered patches. Nitrogen
was depleted there also, presumably due to denitrification
by microbes in anaerobic water and uptake by duckweed
Normal cypress domes are characterized by submerged
aquatic *'..' ijin, bladderworE (Utriciuaria sp.) in this
case. Emergent vegetation is common in ., lri.. ,:, areas,
a .=,. ......
degree of interspersion of the two ecosystem types is very
1.pK al of the north Florida landscape. The current layout
of the experimental areas is I.i,.'hL..J in Fig. 3, which shows
the location of the three domes affected by fire. The fourthh
is about ten miles from this site.
Th1 geologic strata are described in a thesis by
Cutright (1974}, and are shown in F';. 4. His studies
showed the movement of superficial groundwater through
the sands beneath the dome. During the winter dry season,
the ponds of the cypress domes are perched several *...
above the groundwater; in the summer rainy season, the
ta.ur.iJd, i conLacts the lower surface of the dome,
Dfrin,a the summer each dome receives about an inch per
Fig. 3 General site plan of cypress domes receiving
treated :. ' and L.i ,.i . 'A. i
� 4 idealized east-west cruss-sec tiin through the
cypress dome receiving sewage, �i. .iri4 .... .1. -i ,1
strata, test wells, and general groundwater flow
WaE INM FT
Fi: 7 Energy flow diairm of major compartments of
a cypress dome ecosystem inn, I.,Jin g in terac lions
of fire, tree !iL*.E. and drainage.
Fi 8 Major organic storage and flows in the cypress
dome receiving sewage, Flowss are in . ...r ! Lni
matter2 -yr while storage are in g-organic
while most of the vegetation is character *rn. :i. clumped
around the bases of the trees and knees. 1 1.- thick cover of
duckweed soon disappeared in the dome receiving
groundwater, but has presisted in the sewage dome and
I'.I..IIc .ILj in the second dome ri.:e.' ,-,4 sewage almost as
soon as flow began. One species of duckweed returned to
the ....un,.I...ir[, dome when a rookery .i immature white
ibis took up temporary residency. The duckweed fronds
contain nitrogen concentrations of 5 to G - much higher
than the oLher species of vegetation.
Baldcypress .ecilin, planted in all the domes grew
more slowly at first in the sewage domes, but all seedlings
, ) - ,,-- - �- --E
showed a high survival rate, and in the second year the
seedlings in the sewage dome are growing faster than those
in the groundwater dome (J.B. MiI'i h,. Cypress seeds
require drawdown to germinate, so natural regeneration
may have stopped in the three experimental areas except
around the edge.-. New tupelo S LLiin,. have been found
growing in the sewage dome.
T he intensity of the li L , increased by considerable
draining in the area during the last few years, was so great
that most of the hardwoods and pines in the two badly
burned domes were - ilk-.1. Over 95% of the cypress trees
survived, however. Foliage put out in the spring t, llI',.i e
the fire was very abnormal, springing adventitiously from
the trunk and hanging close to the trunks, opening up the
canopy considerably. New limbs were red-barked. Ihe
second year's growth has been more normal, however, and
many of the branches show the spread of *i.IJ'.
- .0- 21.2 q C/kg dry blI ,tt /
" I - T , - -
I -1 >- "
/ � ^____
.o 400 ly/day
0 a 2 I6 20
1 ,4 9 Productivity and transpiration of cypress leaves
as measured by CO2 gas metabolism unit. Data
are from October 1973 B.. .:.,
Burns and Burr, 1974}.
. 2 i.
6A4 tkl H2j/* dry leat w./ y
- 4 8 2 6 20 u
Fig, 10 Productivity and transpiration of cypress leaves
as measured by CO2 gas metabolism unit- Data
are from October 1974 (Cowles, 1974).
characteristic .. baldcypress rather than pondcypress.
The organic cycle within the dome is shown in Fig, 8.
Measurements of litterfall and tree metabolism indicate that
net primary pi I.i; 1. is greater than zero and that the
trees are still tr....;ng The high input to the pond from
autochthonous organic materials is due to the proliferation
and deposition of duckweed. ',pie-- has a Aull leaf
biomass per unit area .ii ..-mll the photosynthesis and
transpiration per area of leaf is normal. Insolation and rates
of ph ..i. . -,,ilhsish for two consecutive years are shown in
,i 9 (Bayley e al. 197-4 and Fig. 10 (Cowles, 1974).
Mosquito fish as well as several other species are
commonly found in cypress domes. All the experimental
ponds are being seeded with a combination of common
freshwater species. Self-design properties of the ecosystem
may result in an interface ecosystem with organisms using
all the resources; � ,,iAinI will eliminate the delay caused h;
species having to gain access by natural means,
The duckweed cover is changing the insect population
considerably, at present fr..tCing moth and shore flies
GROSS PRiMAR"f PRODUCTION
"62 q C/"g dry,
Li ------- A ^ -
go with cypress as contrasted with those that generate
marshes, wet prairies and r, rn.,b without cypress F',, 18).
The cypress arc in the sites with the lower position and
longer period of submergence.
At Wildwood, Florida, wastes from the town have
passed into a r .... Jl..,. swamp for 19 years, Recent
analysis ci these by Brown et af., (1974) showed colilorrn
reduction from 1.6 million MPN to 'H'1 million MPN per
Fig, 1 2 Cypress dome model used
I or computer simulation.
which are common in sewage areas. Many migratory
songbirds, potential carriers of viral encephalitis, were
observed to be attracted by the flies in the sewage domes,
The mosquito species with disease potentials are listed in
Surrounding the domes are slash pine seedlings planted
by Owens-Illinois Incorporated Irom whom the land is
leased- One of the beneficial effects of maintaining domes
0 10 20 30 40
DISTANCE FROM DOME, m
Fig. 11 Soil moisture as a Iunction of distance I rum
edge of cypress dome receiving sewage. Curves
are fur 3 depths in the soil (P. Okorie and
wei during the dry L... ri as natural domes tend to be,
may be in l,....Ii-L: ,,r..r....i li,, forests damp, reducing the
damage by fire and increasing growth of pine trees.
Preliminary data cL'.,h i ~ l by P. Okorie on the effects of the
increased water levels on ..1iI moisture in the surrounding
area are shown in Fig. 11.
The possible interplay of sun, nutrients, water, fire,
and harvesting over the long range was studied by W. ,
Mitsch (1 i'''} in computer simulation models which
incorporated much of the information already discussed.
The model simulated is in Fig. 12 and samples of the graphs
are given in Figs. 13-15. Cutting and draining together
increase the effects of fire, thereby reducing productivity,
whereas fire alone or .,.rli.; alone (without driirnn,.- is not
a major effect and may accelerate the dominance of the
Field workers on the project have commented on the
pleasant microclimate within the dome; measurements arc
being made to determine the differences between radiation
balance, temperature, saturation deficit and wind velocity,
in the domes and surrounding pinclands (see Fig. 16.)
Spectral ,1 .liiJ from - II. Ilir photographs of Lee County
(Capehart et alt, 1975) showed cypress as having higher
reflectance in infrared than pine or Mekaileuca
leurodendron. Reflectance was higher in spring than in
other months and higher in dry cypress.
Cypress lands are only half as expensive as uplands in
Florida, and they are aesthetic attractions in many places.
Savings in housing costs may be realized by liIl.,;iij4 in or
near cypress swamps. Architects associated with the project
seized the need for an equipment shed as an opportunity to
test the concept ol building houses within domes lHouses
have been built in Naples, I 1a .I with cypress swamps as
yards. They are elevated as well, Drainage ditches have
lowered the groundwater in the vicinity to some extent.
Another pa t of the project is monitoring cypress
growth and nutrient relations in swamps and strands where
waters flow in from larger drainage areas and out ., iiI
more as in a river floodplain. Here, there are believed to be
more nutrients available per tree, water levels may be more
regular in dry season, and master growth per tree can he
shown (Fig. 17), Trees are of the baldcypress variety.
At the virgin Corkscrew Swamp site, Collier County,
Florida (Duever eft /., 19741), as part ol this project, have
documented the hydruopeiod and groundwater levels that
Mosquitoes Caught in Ramp Traps in Sewage Dome
(Over I . of Catch)
(HDavis, L. Berner and D. Darne)
Culex spp. (other)
Urano Laenia sapphirina
% ft Cac.li DIsease Potenfial
encephalitis in birds
04 WATER DEPTH
20 40 60 10 100
Fig. 13 Simulation results lor model in Fig. 12 tor
1.i nh. nirbed conditions.
100 ml, phosphorus reduction from 7ppm to 0.1 ppmr, and
increased ai .."..h of the trees by a third.
A new energy investment ratio principle :''a>'i,
1975) can also be used to estimate economic worth (see
Fig. 19a). The average purchased energy (in fossil fuCl
equivalents) that an area can attract depends on a free
resident resource that supplements bought energy in
generating a yield. It is the basis for attracting economic
investment in the first place and the resident energy helps
keep prices competitive. An activity that supplements
bought energy with less than 1.0 kcal Fossil fuel equivalent
(2000 kcal solar energy) I.ii every 2.5 kilocalorie bought
(directly or indirectly from 2.5 kcal o[ fossil fuel work) will
not be competitive.
Based on the nation's economy as a whole, 1 kcal
fossil fuel equivalent (I FL) of sunlight on the average has
been .it JL '.', 2,5 kcal FFE of fossil fuels to the resident
.,..r,..,in.'. The ratio will change as the nation's energy
intensiveness changes. Figure 19 shows a general J, iii
to( the disposal of 2.8 anrLlir ... c.iinI per day of nutrient
wastewater. The high energy quality of Lhe nutrients gives
Fig, 14 Simulation results for model in Fig. 12 in which
cypress, harvesting occurs when cypress biomass
reaches 15 kg/mz.
O 4 ' 60 0 100
,' 60- ---DEAD
40- HARVEST HARVEST
20 40 60 90 100
Fig. 15 Simulation results for model in Fig. 12 where
cypress harvesting and 10 year fire occur, the
latter due to low water conditions.
that flow a value of 8.0 X 10Q kcal FFE/year if it is
matched by enough -ni.Li'hIr (intensity x ir..i to get a net
increase in 8.0 X '1. kcal F f L,'yr from the natural system.
This increase in resident natural energy flow can therefore
theoretically attract 20.0 X 108 kcal FFE/year from
external sources. The model shows that the 1Il'I.,..,[t of the
2.8 mgd of secondary sewage into a natural system has the
capability of increasing the money flow in the local
economy by 8"1 (qll,'year or approximately $0.08/1000
gallons of secondary sewage.
Fig, 16 Microclimate differences between interior of an
undisturbed cypress dome and the interior of
the surrounding pinuland in M.,ich 1 - . U K Heimburg).,
In a cypress system disposal scheme (Fig. '",p. a
loading rate of one inch per week allows the water to
percolate slowly through the soil to the aquifer, leaving
nutrients behind. The nutrients increase the metabolism of
the cypress trees over a large area, and the economic flow is
amrrplai.id with the production of high quality cypress
wood. The purchased energies attracted to this system
include the work of harvesting, the work of "ini.hin. the
wood, and p ..-i..' even the work Ii Iuuiiliig a cypress
TREE PENEITS ryPy Gil
7 Relationship between cypress net p -.I.... I%, and
tree density for three dk,1.,c,,, cypress associations
in the Withlaccochee State Forest in west-c central
Florida. Cypress-mixed hardwood associes generally
indicated riverine or strand swamps '-vii It 1975),
04 wATER DEPTH
G6OuC--WAI Uw.VLs - tUAfm
,4 . . !;* 12 t. .1' 4 h ' . .5 , 1 .
AP MAY JSU JUL MJ XPT OCT
Fig, 18 kI. I. *.lir'.,I between seasonal water levels and
ecosystem type found in the Big Cypress Swamp
in Southwest Florida (M, Duever, E. Carlson .and
L. l,.ip.-ilk, 1974).
wood house or some other manufactured goods. Here the
free work of the sun, water, wind, and land have
supplemented the purchased economy, P t',II.i , calcu-
lations show the investment ratio in this case to be 2.1.
The Water Pollution Control Act of 1972 requires that
cities 1I;, .. Qlh.'u I the country should be treating wastewater
with the "best r ' .i il..ihl, treatment" .** ],,1. 1, "'i..:. '.and
to meet this goal the U.S. Environmental Pr.i, .' -."-
Agency estimates that construction costs for advanced
waste treatment plants *,,il be over twenty billion dollars
(Dobrzynski, 1975). The maintenance and operating costs
of these physical-chemical plants (most conventional
prirnr .' and secondary plants use biological processes) will
also hbe high due to the large amounts .I .c..a .ul.[i,- n
chemicals and energy needed for their operation. A cost
analysis for a ten million .ll... I per day plant based on an
"'Engineering News-Record" Construction Cost Index of
2000 (the EN-R Construction Cost Index for October 2,
1975 was 2278.7) found that a JT* -. hI"1 removal and a
92% phosphorus removal would cost approximately $.50
per 1000 gallons iW.Ill., 1974) and a comparison of
nutrient removal processes for domestic waste by Eliassen
and 1.h,..b .. 1.-',, (1969) 1....... that for both nitrogen
and phosphorus removal efficiencies of 80% or more the
costs ranged ii ..n. 0,17 to $1 'I o per 1000 gallons treated.
Table 2 gives the unit cost for cypress wetland disposal at
4.25 per 1000 -z.ilh.ns, but it should be noted that using a
higher loading rate with the existing system would lead to a
much smaller unit cost. A loading rate of two inches per
week for instance, would lead to a cost of approximately
S,13 per 1000 gallons.
Fig. 19 (a) Definition .,f investment ratio: purchased
feedback divided by free natural inflow whire
both are expressed in fossil fuel equivalents.
Ratio for U.S.A. in fossil fuel equivalents is
(b) Iheoretical investment ratio model showing
energy II. ..-. to be expected (in kcal f, .... I
fuel equivalents x 10 ,'yr) for the disposal
of 2.8 million .I' l. 11,, per day treated se'. .,
into a natural system.
FV; 20 Calculated natural and purchased energy flows
applied to the disposal of 2-8 mgd treated sewage
into cypress domes. For every natural calorie of
fossil luel equivalent of increased productivity,
2.1 calories fossil fuel equivalent are purchased
in this case.
Cost Analysis for Cypress Wetlands Disposal 6n 25,000 GPD
Cost Items Cost
Capital and Installation Cost
Wetwell (6 ft. dia., 10 ft. deep) $ 1,
Pumps (2 ces., 1-I/2 hp.) 2,0
6" P.V,C.. Pipe and I mll.ac (1500 ft.) 4,5
4" P.V.C. Pipe and Fitnings (3100 ft.) 7,1
Annual Operating Cost
Power @ 5 cents/kwh
M.in ICeiai Ce
Labor (1 1/2 hrisjweek @ $1.u *?. I
Total Unit Costs
Annual Capital Cost
Wetwell and Pump (20 yrs. @ ': I 2
P,V,C, Pipe (40 yrs. @ 8 I t
Annual Operating Cost 1,2
TOTAL UNIT COSTS
S/1000 gal. $ 0.:
Bayley, S., L. humi and D, Burr. 1974, Community
metabolism of selected species in a cypress dome in
Alachua County. P.;.., 501-513 in Cypr-es wetlands for
water management, rc:.:...i,. and conservation, Annual
report to NSF and Rockefeller Foundation, Center for
Wetlands, University of Florida.
Brown, S., S. Bayley and J. Zoltck. 1974. Preliminary
results of long term effects of sewage effluent on water
tu :. amnd tree growth in swamplands. P..c. 894-907 in
Cypress wetlands for water management, icc.L linI,, and
( .,,...jli.,-n, Annual report to .',t and Rockefeller
Foundation,. Center for Wetlands, University of Florida.
Capchart, B.L., J,. Ewel, Is k. Sedlik, R.L. 1.,,r,
and ,A. Browder. I'-47. Remote S--,-in. Survey of Spread
ol 4aelideuca. F .... report to National Park Service, U.S.
Department olf Interior. Center for Wetlands, University of
Florida. 113 pp,
Cowles, S.W. F' '74, Primary productivity of cypress
trees, Pages 514-533 in Cypress wetlands for waler
management, recycling and conservation, Annual report to
NSF and Rockefeller Foundation. Center for Wetlands,
University ol Florida.
Cutright, B.1, 1974. H.d -...'....* of a cypress
swainp, north-central Alachua County, Florida. M.S. thesis,
University of Florida 83 pp.
Dobrzynski, J. 1975. " technology and Twenty Billion
r.l ,," Advanced Wastewater Treatment Effort": Engi-
neering News Record, Vol. 195, No. 13. pp. 18-31.
McGraw-Hill, New York.
Duever, Mj., J.E. Carlson and L.A. Riopelle. 1974.
Water budgets and comparative study of virgin Corkscrew
Swamp. Pages 595-634 in Cypress wetlands for water
management, recycling and conservation, Annual report to
NSF and R... ki rK Ili r Foundation. Center for Wetlands,
University -.1 Florida,
Eliassen, R. and G. TI hobanoglous. 1969. Removal i.-
nitrogen and phosphorus from wastewater. Environ. S&i.
ilil -.,I., i-M W,. 1973, Man's impact on the phosphorus
cycle in Florida, Ph.D. J.-r;rLji..,.i, University of Florida.
Mitl,, W.J. 1975. Systems analysis of nutrient
disposal in cypress wetlands and lake ecosystems in Florida.
Ph.D. dissertation, University of Florida. 421 pp.
Odum, H.T, and M.T. Brown, eds, 1',.4 Carrying
Capacity for Man and Nature in South Florida. Final report
to National Park Service, U. S. Department of Interior and
State of I lorida, Division of State Planning. Center I .,
Wellands, University of FL.. ,ida. 886 pp.
Cl ir,,, H.T. I n.' I, Environment, Power and Snoiey.,
Wiley Interscience, New York. 331 pp.
Odum, H,T. and K.C. Ewel. 1974. Cypress Wetlands
lor Watie M 11... Ii ri, ELL.cling and Conservation, First
Annual Report to National Science Foundation F -'J'
Division) and Rockefeller Foundation, 947 pp.
Wallis, I.G. 11',4. "Balance b . 11 Waste Treatment
and Waste Discharge in the U.S. 1957-2000"; Water
Pollution Control Federation journal; Vol. 46 No. 3. pp.
iVi.-.in-, f'.M., A.L. Lewis, C.W. Mountain and L.V.
Pierce. 1975. Demonstration of virus in groundwater after
effluent discharge on soil. Appl.. MILi. i.i..I, 29: 751-757.