Pine tree evapotranspiration

MISSING IMAGE

Material Information

Title:
Pine tree evapotranspiration
Series Title:
Florida Water Resources Research Center Publication Number 62
Physical Description:
Book
Creator:
Riekerk, H.
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Notes

Abstract:
Evapotranspiration data of a young slash pine tree (Pinus elliottii) is presented. The information was obtained with a weighing lysimeter placed in the poorly-drained soil of a flatwoods site. The installation had a sensitivity of about 0.5 mm water. Average seasonal evapotranspiration was 2.4 mm/day for the autumn months, 1.2 mm/day for the winter months, and 5.7 mm/day for the spring months. Equipment failures due to high humidity and lightning damage prevented reliable measurement of evapotranspiration for the summer. Potential evaporation was calculated with the Penman equation using data from a nearby weather station. Total potential evaporation was 1440 mm for the year of measurement. The seasonal ratios of measured evapotranspiration to calculated potential evaporation were 0.92 for the autumn months, 0.44 for the winter months, and 0.89 for the spring months. Measurements will be continued for several years until root restriction begins to have an effect.

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution and holding location.
System ID:
AA00001539:00001


This item is only available as the following downloads:


Full Text







PINE TREE EVAPOTRANSPIRATION


By



H. Riekerk
(Principal Investigator)



PUBLICATION NO. 62



FLORIDA WATER RESOURCES RESEARCH CENTER



RESEARCH PROJECT TECHNICAL COMPLETION REPORT


OWRT Project Number A-039-FLA

Annual Allotment Agreement Numbers

14-34-0001-9010
14-34-0001-0110
14-34-0001-1110


Report Submitted:


March 24, 1982


The work upon which this report is based was supported in part
by funds provided by the United States Department of the
Interior, Office of Water Research and Technology as
authorized under the Water Resources Research
Act of 1964 as amended.















ACKNOWLEDGEMENTS


This study was supported by a three-year grant from the Florida
Water Resources Research Center of the University of Florida, Gainesville,
Florida. Funding was matched by the School of Forest Resources and
Conservation throughout the study period. I very much appreciate
the patience of Dr. James Heaney for my struggles with the field
installations, and the help from Ms. Mary V. Robinson with the
administrative and accounting procedures. Special thanks go to
Mr. Gary Ashby and Mr. Larry Korhnak for their persistent technical
assistance with the project.















ABSTRACT


Evapotranspiration data of a young slash pine tree (Pinus elliottii)
is presented. The information was obtained with a weighing lysimeter
placed in the poorly-drained soil of a flatwoods site. The installation
had a sensitivity of about 0.5 mm water.

Average seasonal evapotranspiration was 2.4 mm/day for the autumn
months, 1.2 mm/day for the winter months, and 5.7 mm/day for the spring
months. Equipment failures due to high humidity and lightning damage
prevented reliable measurement of evapotranspiration for the summer.

Potential evaporation was calculated with the Penman equation using
data from a nearby weather station. Total potential evaporation was
1440 mm for the year of measurement.

The seasonal ratios of measured evapotranspiration to calculated
potential evaporation were 0.92 for the autumn months, 0.44 for the winter
months, and 0.89 for the spring months.

Measurements will be continued for several years until root restriction
begins to have an effect.
















TABLE OF CONTENTS


ACKNOWLEDGEMENTS. . .

ABSTRACT .




I. INTRODUCTION

II. LITERATURE REVIEW

Forest Water Management .

Forest Water Use Information

Methods of Water Use Management

III. PROCEDURES

Site Description . .

Lysimeter Installation .

IV. RESULTS

Installation of Lysimeter

Evapotranspiration Data .

V. DISCUSSION

Lysimeter Performance. .

Pine Tree Evapotranspiration

VI. CONCLUSIONS .. .

LITERATURE CITED. .

APPENDIX .


Page



. ii













5



. 10

. 12




. 14

. 21




. 26

. 26


. 29

. 30

. 37








2





CHAPTER I
INTRODUCTION


Forests are the dominant vegetation type in Florida covering about
15.7 million acres (6.4 million ha) (45% of the total land area) of which
about 175,000 acres (70,000 ha) are replanted annually (Bechtold and
Sheffield, 1980; U.S.D.A., 1977). Annual income during 1979 was estimated
at 132.5 million dollars including sales from farm forests (Greene et al. 1980).
Vegetation cover and its stage of development are major determinants of
water fluxes in the hydrologic cycle. Forest regeneration starts on
nearly bare soil, where evaporation dominates water loss, and progresses
through herbaceous development to full site occupation by the forest trees
which expand water loss through intensive evapotranspiration.

Water yields from forest lands under a given precipitation regime
vary with the stage of forest development (Hibbert 1967; Anderson et
al., 1976). Early successional stages of uplands in North Carolina
may initially yield up to 40% more runoff than the mature stage (Swank
and Helvey, 1970). Similar studies in central Florida flatwoods suggested
100 to 250% increases in runoff after forest vegetation removal (Swindel
et al., 1982). Such water yield changes are controlled mainly by rates
of evapotranspiration as modified by changes in vegetation cover (leaf
area index) and interception, rooting depth, surface roughness, annual
insolation and albedo (Douglass and Swank, 1975; Robins, 1965).

Annual insolation varies little over north-central Florida because
aspect influences are insignificant. Therefore, manipulation of the
distribution and extent of each forest successional stage over time and
space represents an important tool for controlling soil moisture storage,
aquifer recharge, and runoff from most of Florida's landscape. Water
supply and water quality are rapidly becoming crucial issues for water
policies in Florida (Lynne and Kiker, 1976; Maloney 1980; Hutchinson,
1981).

The objective of this study was to evaluate the relationship of the
Penman equation of potential evaporation to actual evapotranspiration
from a developing flatwoods pine plantation. Daily estimates of
atmospheric demand from weather station information were compared with
daily evapotranspiration data from a nearby weighing lysimeter installed
in a young pine plantation. This report covers the first three years of
construction, calibration and evapotranspiration data of the installation.
Monitoring will be continued for another decade until root restriction
and crown closure are expected to significantly affect the lysimeter tree.















CHAPTER II
LITERATURE REVIEW

Forest Water Management


Forest water management in Florida traditionally has been focused
on drainage of excess water from wet areas for better equipment access
and tree growth (Young and Brendemuehl, 1973; Klawitter and Young, 1965).
Water management control by shallow ditches (0.6 m deep) did not affect
water table levels, but 1.5-m-deep ditches improved drainage to the
point of drought conditions for slightly higher elevations in the
flatwoods landscape (Kaufman et al. 1977). A seasonally fluctuating
water table was less beneficial for tree growth than a constant-level
water table (White and Pritchett, 1970).

Little attention has been given to non-structural water management
control by manipulation of evapotranspiration. Forest evapotranspira-
tion represents the major pathway (60-90%) of precipitation disposal in
Florida. Manipulation of evapotranspiration can be achieved by
selecting appropriate tree species with respect to crown structure and
vigor of growth (Krygier 1971; Verry, 1976; Van Lill et al., 1980).
Vegetation with deeper and more extensive root and canopy systems tends
to use more water (Pritchett, 1980; Duncan and Terry, 1982). Also,
it has been shown that coniferous tree canopies increase water use
mainly due to more rain interception and evaporation as compared to
deciduous broadleaf trees (Swank and Douglass, 1974; Maxwell, 1976).
The dense needle-like foliage of Australian pine (Casuarina glauca)
of south Florida may be expected to significantly increase rain
interception.

Major management control of areal evapotranspiration, however, is
by judicious harvest patterning (Douglass, 1965). Vegetation removal
reduces evapotranspiration in some proportion of the amount of leaf area
removed. Consequently soil moisture increases often to the point of
saturation bringing the water table to the surface and generating runoff
(Williams and Libscomb, 1981). This process begins in the wetlands
adjacent to streams and expands the runoff source area uphill with
continual rainfall inputs (Hewlett and Hibbert, 1967). Douglass and
Swank (1975) estimated for Appalachian highlands the annual water yield
increase (AQ inch/yr) from percent of basal area cut (xi) and solar
radiation (x2 microlangleys) by: AQ = 0.0024 (xl/x2)1-446. For north-
central Florida the calculation amounts to a 30 cm yield increase after












clearcutting. Clearcut harvesting followed by windrowing for site
preparation of a pine flatwoods site in north-central Florida increased
water yield by 12 cm (Riekerk et al., 1980). Minimum site preparation by
drum chopping quickly regenerated considerable weed and sprout vegetation
resulting in an increase of only 5 cm runoff/year. Hibbert (1967) in a
worldwide literature review reported average increases of 25 cm (range
3 to 45 cm) water yield from clearcut upland forests. Experimental
manipulation of evapotranspiration by species conversion and vegetation
removal generated a range in runoff changes from minus 20 cm to plus 40
cm, respectively (Stone et al. 1978).

As an example, areal evapotranspiration and water yield from a
large pine flatwoods watershed may be regulated to maintain a sustained
increase of water yields assuming a 25-year harvesting rotation and a
100% increase of runoff due to commonly used clearcut harvesting and
regeneration practices. A 25-year rotation schedule causes 4 percent of
such a large watershed to be harvested each year. Doubling the yield
from this clearcut area results in an annual yield increase of 4
percent from the total watershed. Triple the runoff from the clearcut
area would result in an 8 percent annual yield increase from the total
watershed (Riekerk, 1982).

Recent interest in tree farming for fuelwood with fast-growing
exotic species may drastically alter these water yield fluctuations.
Rapidly growing dense stands of Melaleuca quinquenervia in south Florida
are thought to transpire substantially more water than the more open native
vegetation (Woodall, 1980). Similarly, high-density fuelwood plantations
could considerably increase transpiration as well as rain interception,
reducing watershed yields significantly below those of native forest
or wet savanna vegetation (Hibbert, 1969; Ursic, 1974). Forest growth at
a high planting density soon occupies the total growing site, necessitating
a much shorter harvesting rotation (Wells and Crutchfield, 1974). Therefore,
doubling of runoff from 20% of the area each year by a 5-year harvesting
rotation would result in a 20% increase of annual water yield from the
total fuelwood-farm watershed. Rapid coppice sprouting however would
reduce this effect.

Forest Water Use Information

Some crude evapotranspiration information may be derived from
annual precipitation and runoff data, neglecting differences in storage
between years, and assuming groundwater loss to be only a few percent
of the precipitation input (Lee, 1970). Using these assumptions Klein
et al. (1975) estimated a 1220 mm/yr evapotranspiration rate from the
Big Cypress Swamp in south Florida. This rate represented 90% of the
precipitation input. Speir et al. (1969) reported undisturbed runoff







5




from a forest and range watershed of Taylor Creek in South Florida as
27% of the 129 cm/yr precipitation input. Subsequently, more detailed
accounting for aquifer recharge and changes in storage yielded an estimate
of 86 cm evapotranspiration per year for this area (Knisel et al., 1982).
Faulkner (1975) calculated evapotranspiration as 66% of 135 cm/yr rainfall
using flow-net analysis of geohydrologic data from uplands in north-central
Florida. Riekerk et al. (1978) reported 39% runoff from a poorly-drained
flatwoods forest in north-central Florida after 153 cm precipitation in
a wet year. Similar data for drier years (126 cm/yr precipitation) yielded
about 10% runoff (Riekerk, 1981). Campbell (1979) reported runoff from
a flatwoods forest in north central Florida as 5 and 10 percent with 105
and 88 cm/yr precipitation during dry years. Turner et al. (1977) from a
forested-agricultural watershed in north Florida measured an average of
12.3% runoff with an average of 127 cm/yr precipitation. Such data suggest
that a significant portion of incoming precipitation is disposed of by
evapotranspiration as compared to runoff and deep seepage.

These calculations are based on measurements from experiments on
research watersheds. A wider application requires a state-wide data
base for representative physiographic provinces and forest vegetation
types. Direct measurements of actual evapotranspiration from forest
vegetation at different stages of successional development could
considerably refine forest water management guidelines.

Dohrenwend (1977) calculated with the Holdridge method (Holdridge,
1967) annual potential and "actual" evaporation rates for Florida from
biotemperature data (Figure 1). The calculated value for Gainesville in
north-central Florida underestimated within 6 percent the corresponding
value measured by Bartholic and Buchanan (1976). Agreement of calculated
potential evaporation values with lake evaporation data was good for
southern Florida but underestimated measurements for north Florida.
Burns (1978) measured annual evapotranspiration from Fakahatchee Swamp
at 1024 mm which approximates the 1000 mm value calculated by Dohrenwend
for the area. Parker et al. (1955) measured a range of 890 to 1525 mm
annual evapotranspiration for a variety of vegetation covers near West
Palm Beach. Potential evaporation for that area is about 1250 mm/year
and evapotranspiration as calculated by Dohrenwend is about 1000 mm/yr.
Preliminary information from tension lysimeters in north Florida's sand
hills suggested a water use rate of more than 90% of 120 cm precipitation
during a relatively dry year at the initial stages of pine forest
succession (Riekerk et al. 1981).

Methods of Water Use Measurement

Measurements of forest evapotranspiration as reported above contain
large errors and are poorly replicated in time and space. Predictions










6















TSoo
^ ^ ^ ^ .. ,^ / \





LEGEND
SI Clay Hilla
P= Upl.iand
[ Lowlands I00
-- Ev,.prtsirat on
---Wate Surplus/r \/ \ ^ S











lotow
noo
\ \ |oo





9 mm
















Figure 1. Physiographic provinces and annual evapotranspiration (mm)
regimes (Dohrenwend, 1977).












of evapotranspiration at any location in Florida for a given successional
stage of a forest type are difficult to make. Utilization of climatological
data, or more recently aircraft imagery (Jackson et al., 1981), in empirical
or theoretical equations Thornthwaite, 1948; Penman, 1948; Monteith, 1965)
may provide areal estimates of evapotranspiration such as published by
Dohrenwend (1977). Checking these calculated values against actual field
measurements for the duration of a forest rotation is still required.

Field estimates of evapotranspiration have been made from water
balances of experimental watersheds over the entire development of a
forest (Swank and Helvey, 1970; Rodda, 1976). These units were also large
enough to calibrate satellite imagery (IFAS, 1980). Time resolution of
such water balances is in the order of days. Calibration for watershed
leakage may take a decade. Measurement errors of less than 15% are
considered fair (Lee, 1978).

Measurements of instantaneous vapor fluxes at a measuring point
can be made for short time series with an anemometer and sensitive
temperature, humidity and net radiation sensors (Stanhill, 1969; Hicks
et al., 1975), but rapid degradation of complex electronic instruments
and sensors is yet a serious drawback.

Transpiration of individual branches or even small trees enclosed
in transparent chambers may be measured from vapor analyses of input-
output fluxes of the ventilation air stream (Lee, 1978; Kaufman, 1981).
Temperature control to approximate exterior field conditions is the
largest error component of this method. By definition the method
excludes evaporation of rain normally intercepted by the foliage.

Water use also may be estimated from frequent soil moisture
measurements to assess changes in water content. This technique works
well in deep soils of dry climates where drainage is minimal, or for a
short time interval by excluding rainfall with a soil cover (Metz and
Douglass, 1959).

Daily changes in watertable levels may be used to estimate
evapotranspiration (White, 1932). Assuming early-morning rises to
represent only recharge, this term can be subtracted from the daytime
drawdown to estimate the water use component. Woodall (1980) used this
method for a Melaleuca stand in south Florida and found a good correlation
with potential evaporation and daily insolation. This simple method is
limited to shallow watertables in or near the rooting zone such as occur
in Florida's flatwoods. Daily barometric pressure changes may confound
the observations and need to be accounted for (Turk, 1975). Increased
drainage during low atmospheric pressure was also observed from a
saturated subsoil of a San Dimas lysimeter under pine (Patric, 1974),
and from a sloping forest soil in North Carolina (Hewlett and Hibbert,
1963).











Accurate daily water balance measurements can be made from smaller
forested areas with sealed boundaries such as lysimeters (Van Bavel,
1961, and lysimetric plots (Law, 1957). Boundary restrictions and the
usually disturbed soil profile are the major drawbacks of these methods.
However, lysimeter studies of water balances appear to be a practical
compromise between long-term accuracy, time resolution and cost to
evaluate evapotranspiration during stand development. Walled lysimetric
plots require an impermeable substrate (aquiclude) such as compacted
basal till of glaciated soils (Law, 1957) or a dense clay layer under the
rooting zone (Riekerk et al. 1981). Such plots have undisturbed soils and
can be large enough to alleviate root restriction, but require calibration
or assumptions regarding leakage.

Monolith lysimeters are large containers flush with the surface and
backfilled with soil in an approximation of the original horizons. Drainage
of earlier units was only by gravity, causing a saturated subsoil, a
condition that was absent in the surrounding area (Patric, 1974). Later
units had forced drainage through ceramic tubes to simulate subsoil suction.
The problem of root restriction for large trees remained unresolved except
for the very large lysimeters built and planted with pines near Castricum,
Holland (Van Wyk, 1967).

The weighing lysimeter is a large container buried in the soil but
placed on a weighing mechanism to follow changes in water content continually
(Fritschen et al., 1977). For practical purposes a daily resolution of
0.1 mm sensitivity is sufficient for most studies of water relations in
trees. Fritschen et al. (1973) built a steel weighing lysimeter around the
root ball (3.7 mm diam x 1.2 m depth) of a 28 m Douglas-fir tree. Later
lysimeters were filled with the rootballs of small trees transplanted with
a large crane (Schiess, 1977). Sensitivity of these units was about 0.1
mm water.

Lysimeter water balances may not be very representative for forest
stand evapotranspiration because of the above noted limitations, but
can be used successfully to evaluate plant water relations and to
calibrate other more extensive (meteorological) methods (Lee, 1978; Van
Bavel, 1961). For example, Mustonen and McGuinness(1967) used data from
the grass-covered Coshocton, Ohio, lysimeters to establish correlations
with data from watersheds under different levels of forest management.

Numerous empirical and theoretical equations have been proposed to
describe evapotranspiration using weather data (Gray, 1973). Several
of these estimate "potential evapotranspiration" for a given set of
conditions as a reference. Seasonal crop correction factors derived from
comparisons with actual evapotranspiration measurements have been published
(Van Bavel, 1966). The term "potential evapotranspiration" is misleading
because the transpiration component is not only determined by atmospheric












conditions (Lee, 1980). Some investigators incorporated soil moisture
and plant stomata variables to achieve a closer correlation (Monteith,
1965; Rutter, 1967). A better descriptive term for the predictions from
pure meteorological equations is "potential evaporation."

The Penman equation (Penman, 1948) has been used in this report
because of the sound theoretical foundation, the daily resolution, and the
availability of standard weather data.

The original Penman equation is as follows:

E H E mm/day
A + &


where E
A
&


potential evaporation or atmospheric demand
slope saturated vapor pressure (mm Hg/F)
psychrometric constant = 0.27 mm Hg/ F.


and


H = Rs- (1 r) Rb


where


H = heat budget (mm/day)
Rs = incoming shortwave radiation
Rb = outgoing longwave radiation
r = albedo (percent)


Ea = 0.35 (es ea) (1 + 0.24V)


(cal/cm2/day)
(cal/cm2/day)


where


evaporation at vapor deficit es ea (mm/day)
saturated vapor pressure at Ta (mm Hg)
actual vapor pressure at Ta (mm Hg)
air temperature (F)
wind speed (mph)


and















CHAPTER III
PROCEDURES

Site Description

The study site is located within the research area of the
Cooperative Research in Forest Fertilization (CRIFF) program at the
Austin Cary Forest about 20 km northeast of the University of Florida
(Figure 2). The general area is representative of the extensive flat-
woods forest type in the Gulf-Atlantic lower coastal plain and Florida.
The sandy soils are poorly drained and developed under slash pine
(Pinus elliottii, Engelm.) and longleaf pine (Pinus palustris, Mill.)
forest. The climate is characterized by about 1450 mm/yr rainfall
(Dohrenwend, 1978). Winter storms are associated with passing cold
fronts while summer rains derive primarily from convective storms.
Average maximum air temperatures during January and July are 20.50 C
and 32.70 C, respectively. Average minimum air temperatures are 7.0 C
and 21.60 C, respectively. Evapotranspiratfon is about 900 mm/yr
(Bartholic and Buchanan, 1976).

The research area was cleared and planted to 1000 trees/ha of slash
pine during 1977 (Burger, 1979). A central weather station was established
that records air temperature and humidity, rainfall, water table level,
wind speed and direction, and total and net solar radiation. These
variables include the parameters necessary for calculation of daily
potential evaporation according to the Penman method (see Appendix).

The soil surrounding the weighing lysimeter is a siliceous sandy
hyperthermic ultic haplohumod (Electra series) typical of the poorly
drained flatwoods in north-central Florida. The ash-colored surface
soil (pH = 4.8) is about 53 cm deep with a bulk density of 1.7 g/cm3,
0.7% organic matter, 0.7 meq/100 g CEC, 143 ppm total nitrogen, 20 ppm
total phosphorus, and 60 ppm extractable calcium (CRIFF, 1978). The
underlying spodic horizon is darkly colored by humic-ferric precipitates
and is about 36 cm thick with a bulk density of 1.9 g/cm3. Chemical
properties of this zone include 1.4% organic matter, 2.0 meq/100 g CEC,
219 ppm total nitrogen, 37 ppm total phosphorus, and 25 ppm extractable
calcium. The light colored sandy parent material below grades into clay
at 140 cm depth.












YhI7











- -~ -, -


K $~K' uWji:t
a-A, K
*~ t~itI4v#J*
t 4- 5
-A,


*. *: *. *;

* -A*^


.." ,r ., I .-" : "- "

C).. : ...
WEIGHING -
-.'LYSIMETER


.., ... WEATKER."
.. STATION ; ,

e, *^ :^-1,^ ^, '*-- ,


is


Figure 2. Location of the weighing lysimeter and adjacent weather
station in the pine plantation.


k
-a- -












Lysimeter Installation

A nested double-tank lysimeter was anchored flush with the soil
surface (Figure 3) about 60 m east of the existing weather station.
Each fiberglass tank had two 5-cm-thick reinforcing rings in the 0.6
cm thick wall to prevent circle deformation. The bottom was 0.5 cm
thick fiberglass with reinforcements as explained under RESULTS. The
outside tank (3.2 m diam x 1.4 m deep) provided a level and rigid
base and a dry operating space in the poorly drained soil. Forced
drainage from the inside tank (3.0 m x 1.2 m deep) was from twenty
ceramic filter candles (25 cm x 5 cm diam) buried at the bottom.
Pumping started when the interior water level exceeded the soil surface.

After lysimeter installation, soil settlement and system testing,
pine tree seedlings were planted in and around the lysimeter to
homogenize the plantation area at large. In this fashion the
experiment became nested in a typical landscape and subject to the
meteorological conditions of a developing pine plantation. Weight
changes of the inside tank (resolution about 0.5 mm water) were
monitored by a sensitive differential pressure tranducer and recorded
both on a millivolt chart recorder and an electronic datalogger.
























FORCI
DRAIN


SAND


CLAY /


13








ET






S..o WEIGHT







PPTWATER BALANCE: PPT* RO ET STOE
ED ,RECORDS


















WATER BALANCE: PPTu RO + ET A STO *E


Figure 3. Diagram of hydraulically weighing lysimeter.














CHAPTER IV'
RESULTS

Installation of Lysimeter


The installation of the weighing lysimeter in the poorly-drained
soil of the area caused several problems. The external tank was buried
and anchored through fiberglass extensions at the base during relatively
dry conditions in the spring of 1979. Flotation pressure from the
rising watertable broke the anchorage after the first rains. A second
try during the summer also failed as funding limitations prevented the
use of multiple-well pumping to keep the working area dry. During the
dry fall season the tank was successfully anchored by the rim to 3 m
deep x 10 cm diameter in situ grouted pilings.

Ten lapstrake butyl-rubber hydraulic tubes (15 m long x 5 cm diam)
were filled with de-aired water and tested for pinhole leaks and then
coiled on the bottom (Figure 4). Tire valves vulcanized to one end of
each tube were connected by thickwall plastic tubing to the adjacent
manometer standpipe. Multiple hydraulic tubes are easier to manage,
and if one fails the remaining tubes still remain functional. Also
manometer sensitivity can be changed by disconnecting individual tubes.
Subsequently, the inside tank (213 kg) was centered on the hydraulic
tubes and filled with water for preliminary testing of the system
(Figure 5). The addition of 7400 liters of water brought the manometer
from 96 cm to 297 cm above the bottom of the outside tank. Sensitivity
was 0.4 mm water as read to the nearest mm on the manometer (coefficient
= 2.2). After this test the water was pumped out of the inside tank
and the bottom covered with 5 cm of coarse silica filter sand containing
the 20 ceramic drainage tubes (Figure 6). Soil was backfilled to
approximate the original horizonation and allowed to settle for some
time in saturated condition. Fixed weight additions for calibration
showed sensitivity to be 0.5 mm water. Increased hydraulic pressure
by the heavier soil was brought down to the 3.0 m manometer level
by bleeding excess water out of the hydraulic system, causing the
rubber tubes to become more compressed. The reduction in sensitivity
was probably due to the greater weight of soil and the resulting
larger contact area between the rubber tubes and the bottom of the
inner tank. A slash pine tree seedling was planted in the center of
the lysimeter as part of the surrounding plantation during the winter
of 1980 (Figure 7).









































Figure 4. Hydraulic pressure tubes for weighing lysimeter.





















- -

~: ~
7'


)

a...


U,


It -p




Rip
A '~.-~


Preliminary testing of water-filled weighing lysimeter.


'. **t"\J,: .


Figure 5.









































Ceramic drainage tubes at the bottom of the inside tank.


Figure 6.









































Figure 7. Weighing lysimeter after installation in 1980.












A differential pressure transducer (SETRA* Model 228, 24 VDC) was
delivered and installed during the spring. The manometer line was
connected to one port and a static reference line to the opposite port.
The reference line was to compensate for pressure changes due to
barometric pressure, and due to changes in water density at different
air temperatures. The transducer initially had a linear sensitivity of
7 mV/mm water for a range of + 50 cm differential water pressure. The
electrical output was recorded first only on an Esterline 10 mV/mm
chart recorder advancing at 2 cm/hr. Response time to fixed weight
additions was immediate but stabilization required 2-4 hours. These
early chart recordings were very difficult to interpret because of
erratic and contradictory patterns. Figure 8 summarizes all manometer
recordings uncorrected for re-calibration and manometer coefficients.

An electronic data logger (Campbell Scientific* Model CR-21) with
1 mV sensitivity also was delivered, tested, and installed in the
weather station adjacent to the lysimeter during the fall of 1980.
Datalogger memory storage was about 650 data points, but has been
expanded to 10,000 data points with an on-line cassette recorder.
Weather station sensors and the lysimeter transducer were connected
for recordings of hourly summaries. Cassette tapes were read into
the main frame computer by an RS 232 interface (Campbell Scientific*
Model A 235).

During the wet summer months considerable confusion showed up in
the chart recordings. The recordings also decreased in sensitivity
due to a gradual unbalancing of the inner tank until it finally rested
against the outside tank. A lightning storm severely damaged the
pressure transducer and datalogger.

Analysis of the earlier records revealed a strong influence of out-
side watertable changes on the manometer readings. Only when the water-
table dropped below the lysimeter installation did the recordings show
some regularity. It was surmised that the bottom of the outside tank
was flexing in response to forces from the watertable as well as from
the hydraulic rubber tubes.

During the following dry autumn the lysimeter was dissassembled
and a 10 cm thick slab of reinforced concrete was constructed in the
bottom of the outside tank. The bottom of the inside tank was
reinforced with fiberglassed wood braces (Figure 6). Hydraulic rubber
tubes were replaced and the inside tank was centered again and backfilled
with soil which now was more mixed than before. Sixteen steel skate
wheels bearing against steel strips in the annular space were used to
maintain balance.


*Use of brand names does not imply endorsement.















I SEP I OCT I NOV I DEC I JAN I FEB I MAR I APR I MAY I JUN I JUL I AUG I SEP


'I r I i


I1 1 1 'P Ii


f


0


i 10 20 10 20 i 10 20 10 20 10 20 I 0 20
JUN JUL AUG SEP OCT NOV
1980


S10 20 10 20
DEC I JAN


i 10 20 I 10 20 I 10 20 I 10 20 I 10 20 I 10 20 1
FEB MAR APR MAY JUN JUL
1981


10 20 I 10 20 i 10 20 1 10 20 I 10 20 10 zo
AUG SEP OCT NOV DEC JAN


Figure 8. Record of manometer recordings from the weighing lysimeter.


*Denotes runoff drainage.


400-


380-
360

360'




UJ
340-



300

4 280-
n-
0.




20-

24-0
240-


220


=0 'l i i .I '- -- I II |III 1


l Yl m i -- .v i-- .... .... d ..... a]- i aa u m I i i i I I = =i


I m Illl


n


I OCT I NOV I DEC I JAN I


1I' p


rj


-r"y /











The surrounding young pine plantation had been damaged by the
heavy backhoe used for lysimeter repair and by soil storage piles.
Replacement trees from the 1977 plantation adjacent to the weather
station were transplanted during the spring, including a 2 m tree
centered in the lysimeter (Figure 9). Leaf area of this tree was
estimated during November 1981 at 5.8 m2 resulting in a pine leaf area
index (LAI) of 0.85 m2/m2 for the lysimeter area.

Hydraulic pressure in the manometer rose to 4.0 m by rainfall
during the fall settlement period (Figure 8). Calibration showed
that the sensitivity remained at 0.5 mm. A severe cold spell during
the winter froze the manometer line and the resulting over pressure
damaged the transducer. A break in the hydraulic line dropped the
pressure, but winter rains brought the level back up. A calibrated
lowering of the manometer level by 1.25 m on June 26 reduced the
sensitivity to 0.7 mm (coefficient = 1.4).

As a result of the bottom rigidity and free-standing balance of the
lysimeter general trends associated with periods of drought and rainfall
became apparent in the chart recordings. However, contradictory daily
fluctuations were still present and were found to be correlated with
temperature changes. Such temperature effects have been reported
by others for hydraulic weighing lysimeters (Dylla and Cox, 1973;
Schiess, 1977). Apparently the reference line was not representative
enough for the temperature conditions of the manometer system. The
reference was connected to a larger waterfilled plastic tube located
at the bottom of the annular space of the lysimeter. Temperature
changes at the 1.3 m soil depth are insignificant and temperature
essentially remains constant at 25 C in this soil (Bastos and Smith,
1979). The larger volume of more temperature-stable water improved
the thermal stability somewhat.

High humidity and a lightning storm early in the summer of 1981
again damaged the pressure transducer and datalogger. Periodic record-
ings of the manometer and watertable level presented general impressions
of the changes in water content while instruments were being repaired.
Datalogger sensitivity was diminished to 4 mV/mm water after repair.
During December 1981 the reference was removed altogether and the
manometer plus connecting hydraulic lines protected by a temperature
controlled insulated chamber. The differential pressure transducer was
moved to within 50 cm of the manometer level with one port open to the
atmosphere. This arrangement stabilized the recordings and retained the
linear sensitivity range of the pressure transducer.

Evapotranspiration Data

A data set of daily observations has been summarized in Figure 10.
This information includes air temperature (TEMP), precipitation (PPT),









































Figure 9. Weighing lysimeter installation during the second growing
season with the standard weather station in the background.



































































































In 0 in 0 in
(B fa to In


Figure 10.


Weighing lysimeter and associated climatic data.
temperature, PPT = precipitation, PE = potential

STO = storage, and WT = water table.


Temp = air
evaporation,


.0
N c
0
.0 -,


g0

.0


o
0



0








0



N 0.





o



o


0
0
CN




0


- 0


0
N








0 a








.9 -


0
"N U











calculated potential evaporation (PE), relative water content (STO),
and interior watertable level (WT). The lysimeter water content
data (STO) has been corrected for calibration shifts and associated
changes in the manometer coefficient. Water content values estimated
from rainfall inputs (dashed lines) are supported by the occasional
observations as documented in Figure 8. The tentative nature of these
estimated patterns unfortunately precludes reliable conclusions for
the summer period.

It should be noted that the interior watertable level reached the
tank bottom at the 1.02 m depth. The estimated values in Figure 10 were
reconstructed from rainfall inputs and a specific yield value of 10%.
The storm event of November 10 has been deduced from the watertable record.

The information of figure 10 has been obtained during an exceptionally
dry year. Total precipitation for 1981 was only 88 cm which is 37%
below normal. As a consequence no forced-drainage "runoff" was generated
from the lysimeter tank. This simplified the water balance calculations
considerably. Daily rates of evapotranspiration have been calculated
from the slope of changes in water content during dry periods (Table 1).

Seasonal averages were weighted by the lengths of periods and came
to 2.4 mm/day for the autumn months (October, November, December),
1.2 mm/day for the winter months (January, February, March), and
5.7 mm/day for the spring months (April, May, June). The seasonal
potential evaporation was calculated by the Penman method and came
to 2.7, 5.8, 4.9, and 2.6 mm/day for the winter, spring, summer, and
autumn seasons of 1981, respectively. Total annual potential
evaporation was 1440 mm which was considerably more than the longterm
average of 1100 mm/yr as calculated by Dohrenwend (1978) for the area.

The ratio of evapotranspiration to potential evaporation represents
a "crop factor". Penman (1963) published agricultural crop factors of
0.8 for May-August, 0.6 for November-February, and 0.7 for the transition
months of March-April and September-October. The average seasonal
crop factors obtained by this study were 0.92 for autumn, 0.44 for
winter, and 0.89 for spring.













Table 1. Comparisons of evapotranspiration (ET) and potential evaporation
(PE) for selected periods from December 17, 1980 to January 31, 1982.


ET PE ET/PE
Period mm/day mm/day ratio


Dec 17 Dec 23 1.0 1.2 0.83
Dec 25 Jan 5 2.0 1.3 1.54
Jan 7 Jan 27 1.2 1.7 0.71
Feb 27 Mar 4 1.3 3.9 0.33
Mar 5 Mar 12 1.7 3.7 0.46
Mar 24 Mar 28 1.5 3.9 0.38
Apr 1 May 1 1.9 5.1 0.37
May 2 May 26 2.6 5.8 0.45
May 27 Jun 9 5.7 7.0 0.81
Jun 12 Jun 18 2.7 6.1 0.44


Sep 18 Oct 20 2.2 3.7 0.60
Oct 28 Nov 4 1.3 2.3 0.58
Dec 3 Dec 11 2.0 2.0 1.00
Dec 15 Dec 22 1.9 1.8 1.00
Jan 18 Jan 31 1.3 1.8 0.73














CHAPTER V
DISCUSSION

Lysimeter Performance

The above saga of trials and tribulations represents considerable
frustration often to the point of project dismissal. However, the most
serious problems were overcome, resulting in some useful information.
The summertime failures especially made short thrift of the expectations
at the start of the project. Part of this was due to the application of
existing technology to untried (poorly-drained) soil conditions, and
part was due to the time delays associated with limited resources.
Such growing pains are commonly experienced with these installations
(Fritschen, pers. comm. 1972).

Notwithstanding these problems this large weighing lysimeter
is yet the only successful unit of high sensitivity in north-central
Florida. Furthermore, the lysimeter is unique world-wide in that it
measures water-use by a tree in a poorly-drained soil. All other
lysimeters contain well-drained soils and associated plants.

The information derived from this installation was used to establish
seasonal "crop factors" of water-use by a developing flatwoods pine
plantation. These factors are being verified with data from ongoing
watershed studies of similar plantations in the area. Also, the
controlling processes of water relations are being elucidated in
conjunction with different measurement methods of evaporation,
transpiration, stomatal resistance, and soil moisture.

An alternative simpler method for evapotranspiration measurements
is to build a deep single-tank lysimeter for monitoring a confined
water table to assess daily changes in water content. Unsaturated
soil conditions during the brief dry season could be assessed
periodically with a neutron probe. The tank would prevent deep
seepage and would simplify the daily water balance to precipitation,
runoff and evapotranspiration. As an example, large excavations from
surface mining operations could be lined with heavy-gauge plastic
before refilling with spoil and reclamation with forest vegetation.

Pine Tree Evapotranspiration

The collected data provided for some information on evapotranspiration
by the pine tree planted in the weighing lysimeter. As noted earlier
any extrapolation to the surrounding large-scale plantation has to be
viewed with caution.












The total potential evaporation of 1440 mm/yr was about 30% more
than the longterm average as reported by Dohrenwend (1978). This over
estimate may in part have been generated by the exceptionally high
solar radiation input because the dry year of study had a reduced
cloud cover according to data from a nearby airport.

The ET/PE ratios of 0.89 and 0.92 for the spring and autumn,
respectively, of 1981, suggest a good prediction of evapotranspiration
for these two seasons. However, the low ratio of 0.44 for the winter
season reflects a significant over estimate by the Penman equation.
This metereological method relies primarily on the solar radiation
budget. The relatively high radiation input of the winter period in
effect did not seem to significantly increase the vapor pressure deficit
of frontal air masses flowing into north-central Florida.

The data of Figure 10 and Table 1 show that early winter time
evapotranspiration of 1.2 mm/day slowly increased to nearly 2 mm/day
with increasing air temperatures during March and April. Accelerated
evapotranspiration rates of about 2.6 mm/day during May reflected the
fully activated processes of tree physiology.

During the same period the rate of water table subsidence decreased
slightly after passing the 80 cm depth mark. The football of the transplanted
tree was about 60 cm deep. The capillary fringe above the water table
in these sandy soils is about 20 cm high. These observations combined
suggested that after the water table subsided below the reach of the
majority of tree roots more of the transpired water was derived from
the unsaturated rooting zone.

Evapotranspiration rates during the following period of May 27 to
June 9 averaged 5.7 mm/day which was close to that of potential
evaporation. Air temperatures had reached and exceeded 250 C and a
drought breaking rainstorm recharged the moisture supply in the rooting
zone (Figure 10). A similar rate of 5.5 mm/day was also obtained from
preliminary measurements during the spring of 1980 (Figure 8).

A significant reduction in the rate of evapotranspiration down to 2.7
mm/day was measured during the middle of June. A similar reduction down
to 2.8 mm/day was obtained for the first two weeks of July, 1980 (Figure
8). This reduction occurred after the above noted decrease of moisture
supply from the receding water table. Apparently moisture became
drastically limited when the water table in the lysimeter disappeared
altogether. The interference of the lysimeter bottom with further soil
water table interactions may have depressed the measured evapotranspiration
more so than that of the surrounding plantation.












Nevertheless a real reduction in evapotranspiration occurred during
this dry period which could significantly affect other tree physiological
processes. Rosenzweig (1968) reported a good correlation between primary
productivity at the rate of 2 t/ha for each 10 cm of additional
evapotranspiration. Other observations of this nature for slash pine
in poorly-drained flatwoods (White and Pritchett, 1970) also brought into
question the common assumption that moisture supply is ample throughout
the year.

At this time it is not known whether or not the following summertime
rains kept the surface soil moist enough to provide for a nonlimiting
supply of water to the roots. The water table level did not reappear
until a large storm recharged the lysimeter during late August (Figure 10).

Perusal of the lysimeter observations for the dry period in the
fall of 1981 (Figure 10) showed a rate of 2.2 mm/day. This was equal to
the average evapotranspiration rate of the early spring months under
similar temperature conditions. Some data from early fall of 1980
(Figure 8) approximated a rate of 3.0 mm/day and for early winter about
1.5 mm/day in comparison.

Watertable reduction during the dry spring season amounts to 17
mm/day for the month of April, and 8 mm/day for May when the general
level had substantially subsided. The averaged rate for April-May was
about 13 mm/day. The average watertable reduction over a similar range of
depths during September-October was 19 mm/day. Comparison of these data
with the corresponding evapotranspiration withdrawals gives ratios of
0.11, 0.31, and 0.12 mm evaporated water per mm watertable drop for
April, May, and September-October, respectively. The general rates of
lysimeter weight increase and watertable rise during October-November
1981 show a ratio of about 0.1. The above ratios suggests that the
specific yield of the soil in the lysimeter is about 10%. The higher
value during May probably reflects a relative increase in uptake of
water from the unsaturated rooting zone.














CHAPTER VI
CONCLUSIONS


Installation and maintenance of a large weighing lysimeter and
its instrumentation proved to be troublesome in the poorly-drained
flatwoods soil, especially during the lightning-prone humid summer
months. Recordings by sensitive automated equipment have to be
supplemented with periodic manual manometer readings. Similarly,
state-of-the-art electronic weather sensors and datalogger have
to be backed up by less elegant but more reliable hygrothermographs
and by periodic wind and cumulative precipitation measurements.

The available data show that cool season evapotranspiration
was about 2 mm/day while that for late spring was about 6 mm/day
when soil moisture became replenished by drought breaking rains.
Prediction of the above seasonal rates of evapotranspiration by the
Penman method was good for the autumn and spring seasons.

Water use during the spring drought lowered the watertable below
the rooting zone. Soil water drawn by roots from the unsaturated zone
rapidly became limiting for slash pine transpiration until summer
rains raised the water table again.















LITERATURE CITED


Anderson, H. W., M. D. Hoover, and K.G. Reinhart. 1976. Forests and
Water: Effects of Forest Management on Floods, Sedimentation, and
Water Supply. USDA Forest Service, Pac. SW Forest and Range Exp.
Sta., Rep. PSW-18. 115 p.

Bartholic, J. F., and D. W. Buchanan. 1976. Disposition of Water from
Fruit Crops and Approaches to Increase Water Use Efficiency.
University of Florida Water Resourc. Res. Center Public No. 33.

Bastos, T. X., and W. H. Smith. 1979. Influence of Pine Forest
Removal on Flatwoods Soil Temperature and Moisture Conditions.
IMPAC Report Vol. 4, No. 3. 28 p.

Bechtold, W. A., and R. M. Sheffield. 1980. Forest Statistics for
Florida, 1980. USDA Forest Service, SE For. Expt. Sta. Resource
Bull. SE-58.

Burger, J. A. 1979. The Effects of Harvest and Site Preparation on
the Nutrient Budget of an Intensively Managed Southern Pine
Forest Ecosystem. Ph.D. Dissertation, Univ. of Florida,
Gainesville, FL. 185 p.

Burns, L. A. 1978. Productivity, Biomass, and Water Relations in a
Florida Cypress Forest. Ph.D. Dissertation, Univ. of North
Carolina, Chapel Hill. 214 p.

Campbell, K. L. 1979. Nutrient Loads in Streamflow from Sandy Soils
in Florida. Transactions ASAE 22(5):1115-1120.

CRIFF. 1977. Cooperative Research in Forest Fertilization: Progress
Report 1976-1977. University of Florida, Gainesville, FL,
Institute of Food and Agricultural Sciences, School of Forest
Resources and Conservation, CRIFF Progress Report 1977:131-140.

Dohrenwend, R. E. 1977. Evapotranspiration Patterns in Florida.
Florida Scientist 40(2):184-192.

Dohrenwend, R. E. 1978. The Climate of Alachua County, Florida.
Institute of Food and Agricultural Sciences, Univ. of Florida,
Gainesville, FL. 25 p.












Douglass, J. E. 1967. Effects of Species and Arrangement of Forests
on Evapotranspiration. In "Forest Hydrology" eds. W. E. Sopper
and H. W. Lull, Pergamon Press, New York:1451-1461.

Douglass, J. E., and W. T. Swank. 1975. Effects of Management
Practices on Water Quality and Quantity: Coweeta Hydrologic
Laboratory, North Carolina. In "Municipal Watershed Management",
USDA Forest Service, Gen. Tech. Rep. NE-13:1-13.

Duncan, D. V., and T. A. Terry. 1982. Water Management. In "The
Managed Slash Pine Ecosystem", Ed. R. C. Biesterfeldt, USFS
Southeastern Exp. Sta., Ashville, N.C. (In press.)

Dylla, A. S., and L. M. Cox. 1973. An Economical Hydraulic Weighing
Evapotranspiration Tank. Transact. ASAE 1973:294-295.

Faulkner, G. L. 1975. Flow Analysis of Karst Systems with Well-Developed
Underground Circulation. In "Karst Hydrology and Water Resources",
Proc. U.S.-Yugoslavian Symposium, Dubrovnik; 6.1-6.28.

Fritschen, L. J., L. Cox, and R. Kinerson. 1973. A 28-meter Douglas-
fir in a weighing lysimeter. For. Sci. 19:256-261.

Fritschen, L. J., J. Hsia, and R. Doraiswamy. 1977. Evapotranspir-
ation of a Douglas-fir Determined with a Weighing Lysimter.
Water Resour. Res. 13(l):145-149.

Gray, D. M. 1973. Handbook on the Principles of Hydrology. Water
Information Center, Inc., Huntington, New York. 670 p.

Greene, R. E. L., K. Mathis, L. Polopolus, and J. Holt. 1980.
Economic Data for Florida Agriculture; 1975-1980. Univ. Florida,
Institute of Food and Agricultural Sciences, Gainesville, FL.
152 p.

Hewlett, J. D., and A. R. Hibbert. 1963. Moisture and Energy
Conditions Within a Sloping Soil Mass During Drainage. J. Geophys.
Res. 68(4):1081-1087.

Hewlett, J. D., and A. R. Hibbert. 1967. Factors Affecting the
Response of Small Watersheds to Precipitation in Humid Areas.
In "Forest Hydrology", Eds. W. E. Sopper and H. W. Lull, Pergamon
Press, New York.

Hibbert, A. R. 1967. Forest Treatment Effects on Water Yield. In
"Forest Hydrology", Eds. W. E. Sopper and H. W. Lull, Pergamon
Press, New York. 527-543.











Hibbert, A. R. 1969. Water Yield Changes after Converting a Forested
Catchment to Grass. Water Resour. Res. 5:634-640.

Hicks, B. B., P. Hyson, and C. J. Moore. 1975. A Study of Eddy
Fluxes over a Forest. J. Applied Meterology 14:58-66.

Holdridge, L. R. 1967. Life Zone Ecology. Tropical Science Center,
San Jose, Costa Rica. 206 p.

Hutchinson, B. 1981. Running on Empty. New Florida Special Report,
Oct. 1981, Tequesta, FL. 8 p.

IFAS. 1980. Florida Water Resources NAS 10-9348: Final Report.
Institute of Food and Agricultural Sciences, Univ. Florida. 311 p.

Jackson, T. J., T. J. Schmugge, L. H. Allen, and others. 1981.
Aircraft Remote Sensing of Soil Moisture and Hydrologic Parameters,
Taylor Creek, FL., and Little River, CA; 1979 Data Report. USDA
Agric. Research Service, Agric. Res. Results 13. 36 p.

Kaufman, C. M., W. L. Pritchett, and R. E. Choate. 1977. Growth of
Slash Pine on Drained Flatwoods. Univ. Florida, Florida Agric.
Exp. Sta. Bulletin 792. 30 p.

Kaufman, M. R. 1981. Automatic Determination of Conductance,
Transpiration, and Environmental Conditions in Forest Trees.
Forest Science 27(4):817-827.

Klawitter, R. A., and C. E. Young, Jr. 1965. Forest Drainage
Research in the Coastal Plain. J. Irr. and Drainage. Div. ASCE,
Vol. 91, No. IR 3, Proc. Paper 4456. 1-7.

Klein, H. J., T. Armbuster, B. F. McPherson, and H. J. Freiberger.
1975. Water and the South Florida Environment. U.S. Geol.
Survey Water Resour. Invest. 24-75. 165 p.

Knisel, W. G., Jr., P. Yates, J. M. Sheridan, T. K. Woody, III, L. H.
Allen, Jr., and L. E. Asmussen. 1982. Hydrology and Hydrogeology
of Taylor Creek Watershed, Florida: Data and Analysis. S.E.
Watershed Research Program, USDA Agricultural Research Service
Technical Bulletin. (In press.)

Krygier, J. T. 1971. Comparative Water Loss of Douglas-fir and
Oregon White Oak, Oregon State Univ., Water Resour. Res. Instit,
and School of For. Resour. Res. A-0001-OREG.

Law, F. 1957. Measurement of Rainfall, Interception and Evaporation
Losses in a Plantation of Sitka Spruce Trees. I.A.S.H., Gen.
Ass., Toronto Proc., Vol. 2. 377-411.












Lee, R. 1970. Theoretical Estimates versus Forest Yield. Water
Resour. Res. 6:1327-1334.

Lee, R. 1978. Forest Microclimatology. Columbia University Press,
New York. 276 p.

Lee, R. 1980. Forest Hydrology. Columbia University Press, New
York. 349 p.

Lynne, G. D., and C. F. Kiker. 1976. Water Use in South West Florida
An Economic Perspective. Univ. Florida, Gainesville, Economics
Report No. 82. 70 p.

Maloney, F. E. 1980. Legal Principles of Florida Water Management.
Business and Economic Dimensions 16(l):16-26.

Maxwell, J. 1976. The Effects of Alder Conversion on Streamflow.
Siuslaw National Forest, NW Forest Sciences Lab, Corvallis, OR.
4 p.

Mustonen, S. E., and J. L. McGuinness. 1967. Lysimeter and Watershed
Evapotranspiration. Water Resour. Res. 3(4):989-996.

Metz, L. J., and J. E. Douglass. 1959. Soil Moisture Depletion Under
Several Piedmont Cover Types. USDA, Tech. Bull. 1207. 23 p.

Monteith, J. L. 1965. Evaporation and Environment. In "The State
and Movement of Water in Living Organisms." Ed G. E.Fogg.
Academic Press, New York. 205-234.

Parker, G. G., G. E. Ferguson, S. K. Love, and others. 1955. Water
Resources of Southeastern Florida. U.S. Geol. Survey, Water
Supply Paper 1255. 965 p.

Patric, J. H. 1974. Water Relations of Some Lysimeter-Grown Wildland
Plants in Southern California. USDA Forest Serv, N.E. Forest
Exp. Sta., Upper Darby, PA. 134 p.

Penman, H. L. 1948. Natural Evaporation from Open Water, Bare Soil,
and Grass. Proc. Royal Soc (London) Vol. 193:120-145.

Penman H. L. 1963. Vegetation and Hydrology. Commonwealth Agric.
Bur., Farnham Royal, Bucks, England, Commonwealth Bureau of Soils,
Tech. Commun. 53.












Pritchett, W. L. 1980. Properties and Management of Forest Soils.
John Wiley and Sons, New York. 500 p.

Riekerk, H. 1981. Impacts of Silviculture on Flatwoods Runoff, Water
Quality, and Nutrient Budgets. In "Progress in Wetlands Utili-
zation and Management" Ed. P. M. McCaffrey. Coord. Council
Kissimmee River Valley and Taylor Creek-Nubbins Slough,
Tallahassee, FL. (In press.)

Riekerk, H., S. A. Jones, L. A. Morris, and D. A. Pratt. 1978.
Hydrology and Water Quality of Three Small Lower Coastal Plain
Forested Watersheds. Soil and Crop Sci. Soc. Florida 38:105-112.

Riekerk, H., B. F. Swindel, and J. A. Replogle. 1980. Effect of
Forestry Practices in Florida Watersheds. In "Watershed Manage-
ment '80" Am. Soc. Civil Engin., New York. 706-721.

Riekerk, H., L. F. Conde, J. C. Hendrickson, and W. S. Gain. 1981.
Research on Environmental and Site Effects of Forest Management
Practices in the Lower Coastal Plain. In "First Biennial Southern
Silvicultural Research Conference" Ed. J. P. Barnett. USDA Forest
Service, S. For. Exp. Sta., Gen. Tech. Rep. SO-34:331-339.

Riekerk, H., and D. K. Winter. 1982. Forest Water Management
Principles in Florida. In "Environmentally Sound Water and
Soil Management", Eds. E. G. Kruse and Y. A. Yousef. ASCE,
Irrig. and Drain. Div. 255-267.

Robins, J. S. 1965. Evapotranspiration. In "Methods of Soil Analysis
Part I", Ed. C. A. Black, Agronomy Series No. 9, Am. Soc. Agronomy
Inc., Madison, WI. 286-298.

Rodda, J. C. 1976. Facets of Hydrology. John Wiley and Sons, New
York. 368 p.

Rosenzweig, M. L. 1968. Net Primary Productivity of Terrestrial
Communities Prediction from Climatological Data. Amer. Nat.
102:67-74.

Rutter, A. J. 1967. Analysis of Evaporation from a Stand of Scots
Pine. In "Forest Hydrology" Eds. W. E. Hopper and H. W. Lull.
Pergamon Press, New York.

Schiess, P. 1977. Wastewater Utilization in Forest Soil. In
"Municipal Wastewater and Sludge Recycling on Forest Land and
Disturbed Land." Eds. W. E. Sopper and S. N. Kerr, Pennsylvania
State Univ. Press.












Speir, W. H., W. C. Mills, and J. C. Stephens. 1969. Hydrology of
Three Experimental Watersheds in Southern Florida. USDA, Agric.
Research Service, ARS-41-152. 50 p.

Stanhill, G. 1969. A Simple Instrument for the Field Measurement of
Turbulent Diffusion Flux. J. Applied Meteorology 8(4):509-513.

Stone, E. L., W. T. Swank, and J. W. Hornbeck. 1978. Impacts of
Timber Harvest and Regeneration Systems on Streamflow and Soils
in the Eastern Deciduous Region. In "Forest Soils and Land Use"
Ed. C. T. Youngberg, Dept. Forest and Wood Sciences, Colorado
State Univ. Ft. Collins, CO. 516-536.

Swank, W. T., and J. E. Douglass. 1974. Streamflow Greatly Reduced
by Converting Deciduous Hardwood Stands to Pine. Science 185:
857-859.

Swank, W. T., and J. D. Helvey. 1970. Reduction of Streamflow
Increases Following Regrowth of Clearcut Hardwood Forests.
Intern. Assoc. Sci. Hydrol., Publ. 96:346-360.

Swindel, B. F., C. J. Lassiter, and H. Riekerk. 1982. Effects of
Clearcutting and Site Preparation on Water Yields from Slash Pine
Forests. For. Ecol. and Management. 101-114.

Thorntwaite, C. W. 1948. An Approach Toward a Rational Classification
of Climate. Am. Geographical Review Vol. 38:55-95.

Turk, L. J. 1975. Diurnal Fluctuations of Water Tables Induced by
Atmospheric Pressure Changes. Journal Hydrology 26:1-16.

Turner, R. R., T. M. Burton, and R. C. Harriss. 1977. Descriptive
Hydrology of Three North Florida Watersheds in Contrasting Land
Use. In "Watershed Research in Eastern North America" Ed. D. L.
Correll, Chesapeake Bay Center for Environmental Studies,
Smithsonian Institution, Edgewater, Maryland. 211-227.

Ursic, S. J. 1974. Pine Management Influences the Southern Water
Resource. In "Symposium on Management of Young Pines" Eds. H. L.
Williston and W. E. Balmer. USDA Forest Service, SE For. Exp.
Sta. Athens, GA. 42-49.

Van Bavel, C. H. M. 1961. Lysimetric Measurements of Evapotranspi-
ration Rates in the Eastern United States. Soil Sci. Proc.
25(2):138-142.











Van Bavel, C. H. M. 1966. Potential Evaporation: The Combination
Concept and Its Experimental Verification. Water Resour. Res.
2:455-467.

Van Lill, W. S., F. J. Kruger, and D. B. Van Wyk. 1980. The Effect of
A forestation with Eucalyptus grandis and Pinus patula on Stream-
flow from Experimental Catchments at Mokobulaan, Transvaal. J.
Hydrology 48:107-118.

Van Wyk, W. R. 1967. Forest Hydrology Research in the Netherlands.
In "Forest Hydrology" Eds. W. E. Sopper and H. W. Lull, Pergamon
Press. 65-67.

Verry, E. S. 1976. Estimating Water Yield Differences Between Hard-
wood and Pine Forests: An Application of Net Precipitation Data.
USDA Forest Service, N.C. For. Exp. Sta., St. Paul, Minn., Res.
Paper NC-128. 12 p.

Wells, C. G., and D. M. Crutchfield. 1974. Intensive Culture of Young
Loblolly Pine. In "Symposium on Management of Young Pines" Eds.
H. L. Williston and W. E. Balmer. USDA Forest Service, S&SE For.
Exp. Sta., Athens, GA. 212-229.

White, E. H., and W. L. Pritchett. 1970. Watertable Control and
Fertilization for Pine Production in Flatwoods. Univ. of Florida,
Florida Agric. Exp. Sta. Tech. Bull. 743. 41 p.

White, W. N. 1932. A Method of Estimating Ground Water Supplies
Based on Discharge by Plants and Evaporation from Soil. U.S.
Geol. Survey Water Supply Paper 659. 105 p.

Williams, T. M., and D. J. Libscomb. 1981. Watertable Rise After
Cutting on Coastal Plain Soils. S. Journ. Applied For. 5(l):46-
49.

Woodall, S. L. 1980. Evapotranspiration and Melaleuca. In "Melaleuca
Symposium" Ed. R. K. Geiger. Florida Division of Forestry, Collins
Bldg. Tallahassee, FL: 117-125.

Young, C. E., and R. H. Brendemuehl. 1973. Response of Slash Pine
to Drainage and Rainfall. USDA Forest Service SE Forest Exp.
Sta., Athens, GA, Res. Note SE-186.













SYSTEM MODELING PROGRAM
VERSION 1.3


Title-Calculation of pet and potential rainfall deficit

Initial
/ Dimension RSI (365), TEMP (365), DEWP (365), WIND
/ Dimension DAY (365)
FIXED I
INCON Inc = 0.0
CONST PYSCH = 0.27
CONST ALB = 0.88
CONST BOLTZ = 1.9978E-9
FUNCTION VAPCUR = (-20.0, 0.776), (-15.0, 1.436), (-10


(-5.0, 3.163),
(20.0, 17.535)
(35.0, 42.175)
FUNCTION SI = (
(120.0, 925.0)
(240.0, 885.0)
NOSORT
READ (5, 100
WIND (I), RA
1000 FORMAT (F3.
DYNAMIC
NOSORT


(365), RAIN (365)


.0, 2.149), ...


(0.0, 4.579), (5.0, 6.543), (10.0, 9.209), (15.0, 12.788), ...
, (25.0, 23.756), (30.0, 31.824), ...
, (40.0, 55.324)
0.0, 510.0), (30.0, 580.0), (60.0, 710.0), (90.0, 840.), ...
, (150.0, 990.0), (180.0, 1005.0), (210.0, 950.0), ...
, (270.0, 760.0), (300.0, 630.0), (330.0, 520), (365.0, 500.0)


0)
IN
0,


(DAY (I), RSI (I), TEMP
(I), I = (1,365)
F6.0, F6.1, F6.1, F5.1,


(I), DEWP (I), ...

F6.2)


I = TIME + 1.0
DATE = DAY (I)
DEWPC = (DEWP (I) -32.))*0.5555
TEMPD = (TEMP (I) -32.0)*0.5555
TEMPA = TEMPD + 0.2
TEMPB = TEMPD 0.2
ABST = TEMPD + 273.0
ETA = NLFGEN (VAPCUR, TEMPD)
EST = NLFGEN (VAPCUR, DEWPC)
SLPVAP = (NLFGEN (VAPCUR, TEMPA) NLFGEN (VAPCUR, TEMPB)) /0.72
NAPDEF = 0.35*(ETA EST)*(I.0 + 0.24*WIND(I))
NN = ((RSI (I)?NLFGEN(SI,DATE)) -0.18)/0.55
SHRTWV = RSI(I) ALB /58.0
LONGWV = BOLTZ*ABST**4.0*(0.56 0.092*EST**0.5)*(0.l + 0.9*NN)
H = SHRTWV LONGWV
PET = ((SLPVAP/PYSCH)*H + VAPDEF)/((SLPVAP/PYSCH) + 1.0)
TERMINAL
METHOD RECT
PRTPLT PET (DATE, VAPDEF, H)
PRTPLT TEMPD (DATE)
TIMER DELT = 1.0, FINTIM = 214.0, OUTDEL = 1.0
END


APPENDIX


CONTINUOUS




Full Text

PAGE 1

PINE TREE EVAPOTRANSPIRATION By H. Riekerk (Principal Investigat6r) PUBLICATION NO. 62 FLORI DA WATER RESOURCES RESEARCH CENTER RESEARCH PROJECT TECHNICAL COMPLETION REPORT OWRT Project Number A-039-FLA Annual Allotment Agreement Numbers 14-34 ... 0001-901 0 14-34-0001-0110 14-34-0001-1110 Report Submitted: March 24, 1982 The work upon which this report is based was supported in part by funds provided by the United States Department of the Interi or, Offi ce of Water Research and Technology as authori zed under the Water Resources Resea rch Act of 1964 as amended.

PAGE 2

i ACKNOWLEDGEMENTS This study was supported by a three-year grant from the Florida Water Resources Research Center of the University of Florida, Gainesville, Florida. Funding was matched by the School of Forest Resources and Conservation throughout the study period. I very much appreciate the patience of Dr. James Heaney for my struggles wHh the field installations, and the help from Ms. Mary V. Robinson with the administrative and accounting procedures. Special thanks go to Mr. Gary Ashby and Mr. Larry Korhnak for their persistent technical assistance with the project.

PAGE 3

ii ABSTRACT Evapotranspiration data of a young slash pine tree (Pinus e11iottii) is presented. The information was obtained with a weighing 1ysimeter placed in the poorly-drained soil of a flatwoods site. The installation had a sensitivity of about 0.5 mm water. Average seasonal evapotranspiration was 2.4 mm/day for the autumn months, 1.2 mm/day for the winter months, and 5.7 mm/day for the spring months. Equipment failures due to high humidity and lightning damage prevented rel iabl e measurement of evapotranspiration for the summer Potential evaporation was calculated with the Penman equation using data from a nearby weather station. Total potential evaporation was 1440 mm for the year of measurement. The seasonal ratios of measured evapotranspiration to calculated potential evaporation were 0.92 for the autumn months, 0.44 for the winter months, and 0.89 for the spring months. Measurements will be continued for several years until root restriction begins to have an effect.

PAGE 4

ACKNOWLE DGEMENTS. ABSTRACT I. INTRODUCTION II. LITERATURE REVIEW TABLE OF CONTENTS Forest Water Management Forest Water Use Information Methods of Water Use Management III. PROCEDURES Site Description Lysimeter Installation IV. RESULTS Installation of Lysimeter Evapotranspiration Data. V. DISCUSS ION Lysimeter Performance. Pine Tree Evapotranspiration VI. CONCLUSIONS. LITERATURE CITED. APPENDIX Page i i i 1 3 4 5 10 12 14 21 26 26 29 30 37

PAGE 5

CHAPTER I INTRODUCTION Forests are the dominant vegetation type in Florida covering about 15.7 million acres (6.4 million ha) (45% of the total land area) of which about 175,000 acres (70,000 ha) are replanted annually (Bechtold and Sheffield, 1980; U.S.D.A., 1977). Annual income during 1979 was estimated at 132.5 million dollars including sales from farm forests (Greene et al. 1980). Vegetati on cover and its stage o'f development are major determi nants of water fluxes in the hydrologic cycle. Forest regeneration starts on nearly bare soil, where evaporation dominates water loss, and progresses through herbaceous development to full site occupation by the forest trees which expand water loss through intensive evapotranspiration. Water yields from forest lands under a given precipitation regime vary with the stage of forest development (Hibbert, 1967; Anderson et al., 1976). Early successional stages of uplands in North Carolina may initially yield up to 40% more runoff than the mature stage (Swank and Helvey, 1970). Similar studies in central Florida flatwoods suggested 100 to 250% increases in runoff after forest vegetation removal (Swindel et al., 1982). Such water yield changes are controlled mainly by rates of evapotranspiration as modified by changes in vegetation cover (leaf area index) and interception, rooting depth, annual insolation and albedo (Douglass and Swank, 1975; Robins, 1965). Annual insolation varies little over north-central Florida because aspect influences are insignificant. Therefore, manipulation of the distribution and extent of each forest successional stage over time and space represents an important tool for controlling soil moisture storage, aquifer recharge, and runoff from most of Florida's landscape. Water supply and water quality are rapidly becoming crucial issues for water policies in Florida (Lynne and Kiker, 1976; Maloney 1980; Hutchinson, 1981) The objective of this study was to evaluate the relationship of the Penman equation of potential evaporation to actual evapotranspiration from a developing flatwoods pine plantation. Daily estimates of atmospheric demand from weather station information were compared with daily evapotranspiration data from a nearby weighing lysimeter installed in a young pine plantation. This report covers the first three years of construction, calibration and evapotranspiration data of the installation. Monitoring will be continued for another decade until root restriction and crown closure are expected to significantly affect the lysimeter tree.

PAGE 6

CHAPTER II LITERATURE REVIEW Forest Water Management 3 Forest water management in Florida traditionally has been focused on drainage of excess water from wet areas for better equipment access and tree growth (Young and Brendemuehl, 1973; Klawitter and Young, 1965). Water management control by shallow ditches (0.6 m deep) did not affect water table levels, but i.5-m-deep ditches improved drainage to the point of droughty conditions for slightly higher elevations in the flatwoods landscape (Kaufman et al. 1977). A seasonally fluctuating water table was less beneficial for tree growth than a constant-level wa ter_ tab 1 e(Wh i te and Pritchett, 1970). Little attention has been given to non-structural water management control by manipulation of evapotranspiration. Forest evapotranspiration represents the major pathway (60-90%) of precipitation disposal in Florida. Manipulation of evapotranspiration can be achieved by selecting appropriate tree species with respect to crown structure and vigor of growth (Krygier 1971; Verry, 1976; Van Lil1 et al., 1980). Vegetation with deeper and more extensive root and canopy systems tends to use more water (Pritchett, 1980; Duncan and Terry, 1982). Also, it has been shown that coniferous tree canopies increase water use mainly due to more rain interception and evaporation as compared to deciduous broadleaf trees (Swank and Douglass, 1974; Maxwell, 1976). The dense needle-like foliage of Australian pine (Casuarina glauca) of south Florida may be expected to significantly increase rain interception. Major management control of areal evapotranspiration, however, is by judicious harvest patterning (Douglass, 1965). Vegetation removal reduces evapotranspiration in some proportion of the amount of leaf area removed. Consequently soil moisture increases often to the point of saturation bringing the water table to the surface and generating runoff (Williams and Libscomb, 1981). This process begins in the wetlands adjacent to streams and expands the runoff source area uphill with continual rainfall inputs (Hewlett and Hibbert, 1967). Douglass and Swank (1975) estimated for Appalachian highlands the annual water yield increase inch/yr) from percent of basal area cut (xi) and solar radiation (xz microlangleys) by: = 0.0024 (Xl/X2)1.446. For north central Florida the calculation amounts to a 30 cm yield increase after

PAGE 7

4 clearcutting. Clearcut harvesting followed by windrowing for site preparation of a pine flatwoods site in north-central Florida increased water yield by 12 cm (Riekerk et al., 1980). Minimum site preparation by drum chopping quickly regenerated considerable weed and sprout vegetation resulting in an increase of only 5 cmrunoff/year. Hibbert (1967) in a worldwide literature review reported average increases of 25 cm (range 3 to 45 cm) water yield from clearcut upland forests. Experimental manipulation of evapotranspiration by species conversion and vegetation removal generated a range in runoff changes from minus 20 cmto plus 40 cm, respectively (Stone et al. 1978). As an example, areal evapotranspiration and water yield from a large pine flatwoods watershed may be regulated to maintain a sustained .. increase of water yields assuming a 25-year harvesting rotation and a 100% increase of runoff due to commonly used c1earcut harvesting and regeneration practices. A 25-year rotation schedule causes 4 percent of such a large watershed to be harvested each year. Doubling the yield from this c1earcut area results in an annual yield increase of 4 percent from the total watershed. Triple the runoff from the clearcut area would result in an 8 percent annual yield increase from the total watershed (Riekerk, 1982). Recent interest in tree farming for fuelwood with fast-growing exotic species may drastically alter these water yield fluctuat;'ons. Rapidly growing dense stands of Melaleuca quinquenervia in south Florida are thought to transpire substantially more water than the more open native vegetati on (Woodall, 1980). Simil arly, high-density fuelwood plantations could considerably increase transpiration as well as rain interception, reducing watershed yields significantly below those of native forest or wet savanna vegetation (Hibbert, 1969; Ursic, 1974). Forest growth at a high planting density soon occupies the total growing site, necessitating a much shorter harvesting rotation (Wells and Crutchfield, 1974). Therefore, doub li ng of runoff from 20% of the a rea each yea r by a 5-yea r ha rves ti ng rotation would result in a 20% increase of annual water yield from the total fue1wood-farm watershed. Rapid coppice sprouting however would reduce this effect. Forest Water Use Information Some crude evapotranspiration information may be derived from annual precipitation and runoff data, neglecting differences in storage between years, and assuming groundwater loss to be only a few percent of the precipitation input (Lee, 1970). Using these assumptions et al. (1975) estimated a 1220 mm/yr evapotranspiration rate from the Big Cypress Swamp in south Florida. This rate represented 90% of the precipitation input. Speir et al. (1969) reported undisturbed runoff

PAGE 8

5 from a forest and range watershed of Taylor Creek in South Florida as 27% of the 129 cm/yr precipitation input. Subsequently, more detailed accounting for aquifer recharge and changes in storage yielded an estimate of 86,cm evapotranspiration per year for this area (Knisel et a1., 1982). Faulkner (1975) calculated evapotranspiration as 66% of 135 cm/yr rainfall using flow-net analysis of geohydrologic data from uplands in north-central Florida. Riekerk et a1. (1978) reported 39% runoff from a poorly-drained flatwoods forest in north-central Florida after 153 cm precipitation in a wet year. Similar data for drier years (126 cm/yr precipitation) yielded about 10% runoff ,(Riekerk, 1981),. Campbell (1979) reported runoff from a flatwoods forest in north central Florida as 5 and 10 percent with 105 and 88 cm/yr precipitation during dry years. Turner et al. (1977) from a forested-ag'ricu1tural watershed in north Florida measured an average of 12.3% runoff with an average of 127 cm/yr precipit,:1tion. Such data suggest that a significant portion of incoming precipitation is disposed of by evapotranspiration as compared to runoff and deep seepage. These calculations are based on measurements from experiments on research wafershecfs ... A wider appllcation requires a state-wide data base for representative physiographic provinces and forest vegetation types. Direct measurements of actual evapotranspiration from forest vegetation at different stages of successional development could considerably refine forest water management guidelines. Dohrenwend (1977) calculated with the Holdridge method (Holdridge, 1967) annual potential and II actual II evaporation rates for Florida from biotemperature data (Figure 1). The calculated value for Gainesville in north-central Florida underestimated within 6 percent the corresponding value measured by Bartholic and (1976). Agreement of calculated potential evaporation values with lake evaporation data was good for southern Florida but underestimated measurements for north Florida. Burns (1978) measured annual evapotranspiration from Fakahatchee Swamp at 1024 mm which approximates the 1000 mm value calculated by Dohrenwend for the area. Parker et a1. (1955) measured a range of 890 to 1525 mm annual evapotranspiration for a variety of vegetation covers near West Palm Beach. Potential evaporation for that area is about 1250 mm/year and evapotranspiration as calculated by Dohrenwend is about 1000 mm/yr. Preliminary information from tension lysimeters in north F10rida s sand hills suggested a water use rate of more than 90% of 120 cm precipitation during a relatively dry year at the initial stages of pine forest succession (Riekerk et a1. 1981). Methods of Water Use Measurement Measurements of forest evapotranspiration as reported above contain large errors and are poorly replicated in time and space. Predictions

PAGE 9

LEGEND Clay Hill, !2a Uploncts Cl L.owland. Evapotran,piration ___ Wat ... Surplu. Figure 1. Physiographic provinces and annual evapotranspiration .(mm) regimes (Dohrenwend, 1977). 6

PAGE 10

7 of evapotranspiration at any location in Florida for a given successional stage of a forest type are difficult to make. Utilization of climatological data, or more recently aircraft imagery (Jackson et al., 1981), in empirical or theoretical equations crhornthwaite, 1948; Penman, 1948; Monteith, 1965) may provide areal estimates of evapotranspiration such as published by Dohrenwend (1977). Checking these calculated values against actual field measurements for the duration of a forest rotation is still required. Field estimates of evapotranspiration have been made from water balances of experimental watersheds over the entire development of a forest (Swank and Helvey, 1970; Rodda, 1976). These units were also large enough to cal ibrate satell ite imagery (IFAS, 1980). Time resol uti on of such water balances is in the order of days. Calibration for watershed leakage may take a decade. Measurement errors of less than 15% are considered fair (Lee, 1978). Measurements of instantaneous vapor fluxes at a measuring point can be made for short time series with an anemometer and sensitive temperature, humidity and net radiation sensors (Stanhill, 1969; Hicks et al., 1975), but rapid degradation of complex electronic instruments and sensors is yet a serious Transpiration of individual branches or even small trees enclosed in transparent chambers may be measured from vapor analyses of input output fluxes of the ventilation air stream (Lee, 1978; Kaufman, 1981). Temperature control to approximate exterior field conditions is the largest error component of this method. By definition the method excludes evaporation of rain normally intercepted by the foliage. Water use also may be estimated from frequent soil moisture measurements to assess changes in water content. This technique works well in deep soils of dry climates where drainage is minimal, or for a short time interval by excluding rainfall with a soil cover (Metz and Dougl ass, 1959). Daily changes in watertable levels may be used to estimate evapotranspiration (White, 1932). Assuming early-morning rises to represent only recharge, this term can be subtracted from the daytime drawdown to estimate the water use component. Woodall (1980) used this method for a Mela1euca stand in south Florida and found a good correlation with potential evaporation and daily insolation. This simple method is limited to shallow watertables in or near the rooting zone such as occur in Florida's flatwoods. Daily barometric pressure changes may confound the observations and need to be accounted for (Turk, 1975). drainage during low atmospheric pressure was also observed from a saturated subsoil of a San Dimas 1ysimeter under pine (Patric, 1974), and from a sloping forest soil in North Carolina (Hewlett and Hibbert, 1963)

PAGE 11

8 Accurate daily water balance measurements can be made from smaller forested areas with sealed boundaries such as lysimeters (Van Bavel, 1961, and lysimetric plots (Law, 1957) .. Boundary restrictions and the usua 11y disturbed so i 1 profi 1 e are the maj or drawbacks of these methods. However, lysimeter studies of water balances appear to be a practical compromise between long-term accuracy, time resolution and cost to evaluate evapotranspiration during stand development. Walled lysimetric plots require an impermeable substrate (aquiclude) such as compacted basal till of glaciated soils {Law, 1957} or a dense clay layer under the rooting zone (Riekerket al. 1981}. Such plots have undisturbed soils and can be large enough to alleviate root restriction, but require calibration or assumptions regarding leakage. Monolith lysimeters are large containers flush with the surface and backfilled with soil in an approximation of the original horizons. Drainage of earlier units was only by gravity, causing a saturl).ted subsoil, a condition that was absent in the surrounding area (Patric, 1974). Later units had forced drainage through ceramic tubes to simulate subsoil suction. The problem of root restriction for large trees remained unresolved except for the very large lysimeters built and planted with pines near Castricum, Holland (Van Wyk, 1967). The weighing lysimeter is a large container buried in the soil but placed on a weighing mechanism to follow changes in water content continually (Fritschen et al., 1977). For practical purposes a daily re.solution of 0.1 mm sensitivity is sufficient for most stUdies of water relations in trees. Fritschen et al. (1973) built a steel weighing lysimeter around the root ball (3.7 mm diam x 1.2 m depth) of a 28 m Douglas-fir tree. Later lysimeters were filled with the rootballs of small trees transplanted with a large crane (Schiess; 1977). Sensitivity of these units was about 0.1 mm water. Lysimeter water balances may not be very representative for forest stand evapotranspiration because of the above noted limitations, but can be used successfully to evaluate plant water relations and to calibrate other more extensive (meteorological) methods (Lee, 1978; Van Bavel,1961). For example, Mustonen and McGuinness (1967) used data from the grass-covered Coshocton, Ohio, lysimeters to establish correlations with data from watersheds under different levels of forest management. Numerous empirical and theoretical equations have been proposed to describe evapotranspi ration using weather data (Gray., 1973). Several of these estimate "potential evapotranspiration" for a given set of conditions as a reference. Seasonal crop correction factors derived from comparisons with actual evapotranspiration measurements have been published (Van Bavel, 1966). The term "potential evapotranspiration" is misleading because the transpiration component is not only determined by atmospheric

PAGE 12

9 conditions (lee, 1980). Some investigators incorporated soil moisture and plant stomata variables to achieve a closer correlation (Monteith, 1965; Rutter, 1967). A better descriptive term for the predictions from pure meteorological equations is "potential evaporation." The Penman equation (Penman, 1948) has been used in this report because of the sound theoretical foundation, the daily resolution, and the availability of standard weather data. The original Penman equation is as follows: E = H + & Ea mm/day + & where E = potential evaporation or atmospheric demand = slope saturated vapor pressure (mm Hg/F) and where and where & = psychrometric constant = 0.27 mm H9/ F. H = Rs (1 -r) -Rb H = heat budget (mm/day) Rs = incoming shortwave radiation (cal/cm2/day) Rt, = outgoing longwave radiation (cal/cm2/day) r = albedo (percent) Ea = 0.35 (es ea) (1 + 0.24V) Ea = evaporation at vapor deficit e s -ea (mm/day) es = saturated vapor pressure at Ta (mm Hg) ea = actual vapor pressure at Ta (mm Hg) Ta = air temperature (F) V = wind speed (mph)

PAGE 13

CHAPTER II I PROCEDURES Site Description 10 The study site is located within the research area of the Cooperative Research in Forest Fertilization (CRIFF) program at the Austin Cary Forest about 20 km northeast of the University of Florida (Figure 2). The general area is representative of the extensive flatwoods forest type in the Gulf-Atlantic lower coastal plain and Florida. The sandy soils are poorly drained and developed under slash pine (Pinus elliotti;, Engelm.) and longleaf pine (Pinus palustris, Mill.) forest. The climate is characterized by about 1450 mm/yr rainfall (Dohrenwend, 1978). Winter storms are associated with passing cold fronts while summer rains derive primarily from convective storms. Average maximum air temperatures during January and July are 20.5 C and 32.7 C, respectively. Average minimum air temperatures are 7.0 C and 21.6 C, respectively. Evapotranspiration is about 900 mm/yr (Bartholic and Buchanan, 1976). The research area was cleared and planted to 1000 trees/ha of slash pine during 1977 (Burger, 1979). A central weather station was established that records air temperature and humidity, rainfall, water table level, wind speed and direction, and total and net solar radiation. These variab les i ncl ude the parameters necessary for cal cul ati on of daily potential evaporation according to the Penman method (see Appendix). The soil surrounding the weighing lysimeter is a siliceous sandy hyperthermic ultic haplohumod (Electra series) typical of the poorly drained flatwoods in north-central Florida. The ash-colored surface soil (pH = 4.8) ;s about 53 cm deep with a bulk density of 1.7 g/cm3 0.7% organic matter, 0.7 meq/100 9 CEC, 143 ppm total nitrogen, 20 ppm total phosphorus, and 60 ppm extractable calcium (CRIFF, 1978). The underlying spodic horizon is darkly colored by humic-ferric precipitates and is about 36 cm thick with a bulk density of 1.9 g/cm3 Chemical properties of this zone include 1.4% organic matter, 2.0 meq/100 g CEC, 219 ppm total nitrogen, 37 ppm total phosphorus, and 25 ppm extractable calcium. The light colored sandy parent material below grades into clay at 140 cm depth.

PAGE 14

-""-'-.: ":''' if" :... ., Fi oure 2. station in Location the pine of the weiohinq plantation: 11 lysimeter and adjacent weather

PAGE 15

12 Lysimeter Installation A nested double-tank lysimeter was anchored flush with the soil surface (Figure 3) about 60 m east of the existing weather station. Each fiberglass tank had two 5-cm-thick reinforcing rings in the 0.6 cm thick wall to prevent circle deformation. The bottom was 0.5 cm thick fiberglass with reinforcements as explained under RESULTS. The outside tank (3.2 m diam x 1.4 m deep) provided a level and rigid base and a dry operating space in the poorly drained soil. Forced drainage from the inside tank (3.0 m x 1.2 m deep) was from twenty ceramic filter candles (25 cm x 5 cm diam) buried at the bottom. Pumping started when the interior water level exceeded the soil surface. After 1ysimeter installation, soil settlement and system testing, pine tree seedlings were planted in and around the lysimeter to homogenize the plantation area at large. In this fashion the experiment became nested in a typical landscape and subject to the meteorological conditions of a developing pine plantation. Weight changes of the inside tank (resolution about 0.5 mm water) were monitored by a sensitive differential pressure tranducer and recorded both on a millivolt chart recnrder and an electronic datalogger.

PAGE 16

13 PPT 0 1 PPT :::;, ..J 1 :::l 0 0: 0: I-STO Q z >-0 WEIGHT %: 0 RECORDS { SAND / / / / / / / / / WATER BALANCE: PPT -RO + ET j: STO E Figure 3. Diagram of hydraulically weighing lysimeter.

PAGE 17

14 CHAPTER I V.: RESULTS Installation of Lysimeter The installation of the weighing lysimeter in the poorly-drained soil of the area caused several problems. The external tank was buried and anchored through fiberglass extensions at the base during relatively dry conditions in the spring of 1979. Flotation pressure from the rising watertable broke the anchorage after the first rains. A second try during the summer also failed as funding limitations prevented the use of multi pl e-well pumping to keep the working area dry. Ouri ng the dry fall season the tank was successfully anchored by the rim to 3 m deep x 10 cm diameter .i!!. situ grouted pilings. Tenlapstrake butyl-rubber hydraulic tubes (15 m long x 5 cm diam) were filled with de-aired water and tested for pinhole leaks and then coiled on the bottom (Figure 4). Tire valves vulcanized to one end of each tube were connected by thickwall plastic tubing to the adjacent manometer standpipe. Multiple hydraulic tubes are easier to manage, and if one fails the.remaining tubes still remain functional. Also manometer sensitivity can be changed by disconnecting individual tubes. Subsequently, the inside tank (213 kg) was centered on the hydraulic tubes and filled with water for preliminary testing of the system (Figure 5). The addition of 7400 liters of water brought the manometer from 96 cm to 297 cm above the bottom of the outside tank. Sensitivity was 0.4 mm water as read to the nearest mm on the manometer (coefficient = 2.2). After this test the water was pumped out of the inside tank and the bottom covered with 5 cm of coarse silica filter sand containing the 20 ceramic drainage tubes (Figure 6). Soil was backfilled to approximate the original horizonation and allowed to settle for some time in saturated cohdition. Fixed weight additions for calibration showed sensitivity to be 0.5 mm water. Increased hydraulic pressure by the heavier soil was brought down to the 3.0 m manometer level by bleeding excess water out of the hydraulic system, causing the rubber tubes to become more compressed. The reduction in sensitivity was probably due to the greater weight of soil and the resulting larger contact area between the rubber tubes and the bottom of the inner tank. A slash pine tree seedling was planted in the center of the lysimeter as part of the surrounding plantation during the winter of 1980 (Figure 7).

PAGE 18

.... .. :-,.. -.r .... -. "'. -. --;;.-" w_ Figure 4. Hydraulic pressure tubes for weighing 1ysimeter. 15 ..

PAGE 19

... \ .... Figure 5. 16 .. "" ... :" ____ Preliminary testing of water-filled weighing lysimeter.

PAGE 20

-"'" 17 -Jj>'" ''l: Figure 6. Ceramic drainage tubes at the bottom of the inside tank.

PAGE 21

18 Figure 7. Weighing lysimeter after installation in 1980.

PAGE 22

19 A differential pressure transducer (SETRA* Model 228, 24 VDC) was delivered and installed during the spring. The manometer line was connected to one port and a static reference line to the opposite port. The reference line was to compensate for pressure changes due to barometric pressure, and due to changes in water density at different air temperatures. The transducer initially had a 1 inear sensitivity of 7 mV/mm water for a range of + 50 cm differential water pressure. The electrical output was recorded first only on an Esterline 10 mV/mm chart recorder advancing at 2 cm/hr. Response time to fixed weight additions was immediate but stabil ;zation required 2-4 hours. These early chart recordings were very difficult to interpret because of errati c and contradictory patterns. Figure 8 summari zes all manometer recordings uncorrected for re-calibration and manometer coefficients. An electronic data logger (Campbell Scientific* Model CR-21) with 1 mV sensitivity also was delivered, tested, and installed in the weather station adjacent to the lysimeter during the fall of 1980. Datalogger memory storage was about 650 data points, but has been expanded to 10,000 data pOints with an on-line cassette recorder. Weather station sensors and the lysimeter transducer were connected for recordings of hourly summaries. Cassette tapes were read into the main frame computer' by an RS 232 interface (Campbell Scientific* Model A 235). During the wet sunmer months considerable confusion showed up in the chart recordings. The recordings also decreased in sensitivity due to a gradual unbalancing of the inner tank until it finally rested against the outside tank. A lightning storm severely damaged the pressure transducer and datalogger. Analysis of the earlier records revealed a strong influence of out side watertable changes on the manometer readings. Only when the watertable dropped below the 1ysimeter installation did the recordings show some regularity. It was surmised that the bottom of the outside tank was fl exing in response to forces from the watertab1 e as well as from the hydraulic rubber tubes. During the following dry autumn the lysimeter was dissassembled and a 10 cm thick slab of reinforced concrete was constructed in the bottom of the outside tank. The bottom of the inside tank was reinforced with fiberg1assed wood braces (Figure 6). Hydraulic rubber tubes were replaced and the inside tank was centered again and backfilled with soil which now was more mixed than before. Sixteen steel skate wheels bearing against steel strips in the annular space were used to maintain balance. *Use of brand names does not imply endorsement.

PAGE 23

0 ): g 2 ..... :E 5 ...J 4 ...J 5 Z 4; 7 400 380 360 :E U 340 ... 0: ;:, 320 ... II: Q. 0 300 :::i ;:,
PAGE 24

21 The surrounding young pine plantation had been damaged by the heavy backhoe used for lysimeter repair and by soil storage piles. Replacement trees from the 1977 plantation adjacent to the weather station were transpl anted'during the spring, incl uding a 2 m tree centered in the lysimeter (Figure 9). Leaf area of this tree was estimated during November 1981 at 5.8 m2 resulting in a pine leaf area index (LA!) of 0.85 m2/m2 for the lysimeter area. Hydraulic pressure in the manometer rose to 4.0 m by rainfall during the fall settlement period (Figure 8). Calibration showed that the sensitivity remained at 0.5 mm. A severe cold spell during the winter froze the manometer line and the resulting over pressure damaged the transducer. A break in the hydraulic line dropped the pressure, but winter rains brought the level back up. A calibrated lowering of the manometer level by 1.25 m on June 26 reduced the sensitivity to 0.7 mm (coefficient = 1.4). As a result of the bottom rigidity and free-standing balance of the lysimeter general trends associated with periods of drought and rainfall became apparent in the chart recordings. However, contradictory daily fluctuations were still present and were found to be correlated with temperature changes. Such temperature effects have been reported by others for hydraulic weighing lysimeters (Dyllaand Cox, 1973; Schiess, 1977). Apparently the reference line was not representative enough for the temperature conditions of the manometer system. The reference was connected to a larger waterfilled plastic tube located at the bottom of the annular space of the lysimeter. Temperature changes at the 1.3 m soil depth are insignificant and temperature essentially remains constant at 25 C in this soil (Bastos and Smith, 1979). The larger volume of more temperature-stable water improved the thermal stability somewhat. High humidity and a lightning storm early in the summer of 1981 again damaged the pressure transducer and datalogger. Periodic record ings of the manometer and watertable level presented general impressions of the changes in water content while instruments were being repaired. Datalogger sensitivity was diminished to 4 mV/mm water after repair. During December 1981 the reference was removed altogether and the manometer plus connecting hydraulic lines protected by a temperature controlled insula:ted chamber. The differential pressure transducer was moved to within 50 cm of the manometer level with one port open to the atmosphere. This arrangement stabilized the recordings and retained the 1 inear sensitivity range of the pressure transducer .. Evapotranspiration Data A data set of daily observations has been summarized in Figure 10. This information includes air temperature (TEMP), precipitation (PPT)

PAGE 25

Figure 9. Weighing lysimeter installation during the second growing season with the standard weather station in the background. 22

PAGE 26

(l, "W 0 N 0 (l, W Q. Q. (/) s;! 0 C'II II) "5 &0 0 & e-... , ..... --.. ) "I" {. '" ........ --------... i II) 0 II) co co n / _I;? c.., ............ ot:: .............. ... 0 n 0 s;!Z 0 N -u s;!O 0 Q. N III 2(/) 0 N 01 CD en 0 N Q' 0 N 0 N >0-C Q 0 N .. 2 Q. ct 0 N Is 2 0 N .0 III Q IL 0 c: N C 2 0 N U III 2 0

PAGE 27

24 calculated potential evaporation (PE), relative water content (STO) and interior watertable level (WT). The lysimeter water content data (STO) has been corrected for calibration shifts and associated changes in the manometer coefficient. Water content values estimated from rainfall inputs (dashed lines) are supported by the occasional observations as documented in Figure 8. The tentative nature of these estimated patterns unfortunately precludes reliable conclusions for the summer period. It should be noted that the interior watertable level reached the tank bottom at the 1.02 m depth. The estimated values in Figure 10 were reconstructed from rainfall inputs and a specific yield value of 10%. The storm event of November 10 has been deduced from the watertable record. The information of figure 10 has been obtained during an exceptionally dry year. Total precipitation for 1981 was only 88 cm which is 37% below normal. As a consequence no forced-drainage "runoff" was generated from the lysimeter tank. This simplified the water balance calculations considerably. Daily rates of evapotranspiration have been calculated from the slope of changes in water content during dry periods (Table 1). Seasonal averages were weighted by the lengths of periods and came to 2.4 mm/day for the autumn months (October, November, December), 1.2 mm/day for the winter months (January, February, March), and 5.7 mm/day for the spring months (April, May, June). The seasonal potential evaporation was calculated by the Penman method and came to 2.7, 5.8,4.9, and 2.6 mm/day for the winter, spring, summer, and autumn seasons of 1981, respectively. Total annual potential evaporation was 1440 mm which was considerably more than the longterm average of 1100 mm/yr as calculated by Dohrenwend (1978) for the area. The ratio of evapotranspiration to potential evaporation represents a "crop factor". Penman (1963) published agricultural crop factors of 0.8 for May-August, 0.6 for November-February, and 0.7 for the transition months of March-April and September-October. The average seasonal crop factors obtained by this study were 0.92 for autumn, 0.44 for winter, and 0.89 for spring.

PAGE 28

25 Table 1. Comparisons of evapotranspiration (ET) and potential evaporation (PE) for selected peri ods from December 17, 1980 to January 31, 1982. Peri od Dec 17 -Dec 23 Dec 25 Jan 5 Jan 7 Jan 27 Feb 27 -Mar 4 Mar 5 -Mar 12 Mar 24 -Mar 28 Apr 1 -May 1 May 2 -May 26 May 27 -Jun 9 Jun 12 -Jun 18 Sep 18 Oct 20 Oct 28 -Nov 4 Dec 3 -Dec 11 Dec 15 -Dec 22 Jan 18 Jan 31 .. ET mm/day 1.0 2.0 1.2 1.3 1.7 l.5 1.9 2.6 5.7 2.7 2.2 1.3 2.0 1.9 1.3 PE mm/day 1.2 1.3 1.7 3.9 3.7 3.9 5.1 5.8 7.0 6.1 3.7 2.3 2.0 1.8 1.8 ET/PE ratio 0.83 1.54 0.71 0.33 0.46 0.38 0.37 0.45 0.81 0.44 0.60 0.58 1.00 1.00 0.73

PAGE 29

CHAPTER V DISCUSSION Lysimeter Performance 26 The above saga of trials and tribulations represents considerable frustration often to the point of project dismissal. However, the most serious problems were overcome, resulting in some useful information. The summertime failures especially made short thrift of the expectations at the start of the project. Part of this was due to the application of existing technology to untried (poorly-drained) soil conditions, and part was due to the time delays associated with limited resources. Such growing pains are commonly experienced with these installations (Fritschen, pers. comm. 1972). Notwithstanding these problems this large weighing lysimeter is yet the only successful unit of high sensitivity in north-central Florida. Furthermore, the 1ysimeter is unique world-wide in that it measures water-use by a tree in a poorly-drained soil. All other lysimeters contain well-drained soils and associated plants. The information derived from this installation was used to establish seasonal crop factors" of water-use by a developing flatwoods pine plantation. These factors are being verified with data from ongoing watershed studies of similar plantations in the area. Also, the controlling processes of water relations are being elucidated in conjunction with different measurement methods of evaporation, transpiration, stomatal resistance, and soil moisture. An alternative simpler method for evapotranspiration measurements is to build a deep single-tank lysimeter for monitoring a confined water table to assess daily changes in water content. Unsaturated soil conditions during the brief dry season could be assessed periodically with a neutron probe. The tank would prevent deep seepage and would simplify the daily water balance to precipitation, runoff and evapotranspiration. As an example, large excavations from surface mining operations could be 1 ined with heavy-gauge plastic before refilling with spoil and reclamation with forest vegetation. PireTree Evapotranspiration The collected data provided for some information on evapotranspiration by the pine tree planted in the weighing lysimeter. As noted earlier any extrapolation to the surrounding large-scale plantation has to be viewed with caution.

PAGE 30

27 The total potential evaporation of 1440 mm/yr was about 30% more than the longterm average as reported by Dohrenwend (1978). This over estimate may in part have been generated by the exceptionally high solar radiation input because the dry year of study had a reduced cloud cover according to data from a nearby airport. The ET/PE ratios of 0.89 and 0.92 for the spring and autumn, respectively, of 1981, suggest a good prediction of evapotranspiration for these two seasons. However, the low ratio of 0.44 for the winter season reflects a significant over estimate by the Penman equation. This metereological method relies primarily on the solar radiation budget. The relatively high radiation input of the winter period in effect did not seem to significantly increase the vapor pressure deficit of frontal air masses flowing into north-central Florida. The data of Figure 10 and Table 1 show that early winter time evapotranspiration of 1.2 mm/day slowly increased to nearly 2 mm/day with increasing air temperatures during March and April. Accelerated evapotranspiration rates oLabout2.6 mm/day-during May reflected the fully acti vated processes of tree physiology. During the same period the rate of water table subsidence decreased slightly after passing the BO cm depth mark. The rootball of the transplanted tree was about 60 cm deep. The capi" ary fri nge above the water tabl e in these sandy soils is about 20 cm high. These observations combined suggested that after the water table subsided below the reach of the majority of tree roots more of the transpired water was derived from the unsaturated rooting zone. Evapotranspiration rates during the following period of May 27 to June 9 averaged 5.7 mm/day which was close to that of potential evaporation. Air temperatures had reached and exceeded 250 C and a drought breaking rainstorm recharged the moisture supply in the rooting zone (Figure 10). A similar rate of 5.5 mm/day was also obtained from preliminary measurements during the spring of 1980 (Figure 8). A significant reduction in the rate of evapotranspiration down to 2.7 mm/day was measured during the middle of June. A similar reduction down to 2.8 mm/day was obtained for the first two weeks of July, 1980 (Figure 8). This reduction occurred after the above noted decrease of moisture supply from the receding water table. Apparently moisture became drastically limited when the water table in the lysimeter disappeared altogether. The interference of the lysimeter bottom with further soil water table interactions may have depressed the measured evapotranspiration more so than that of the surrounding plantation. I

PAGE 31

28 Nevertheless areal reduction in evapotranspiration occurred during this dry period which could significantly affect other tree physiological processes. Rosenzweig (1968) reported a good correlation between primary productivity at the rate of 2 t/ha for each 10 cm of additional evapotranspiration. Other observations of this nature for slash pine in poorly-drained flatwoods (White and Pritchett, 1970) also brought into question the common assumption that moisture supply is ample throughout the year. At this time it is not known whether or not the following summertime rains kept the surface soil moist enough to provide for a nonl imiting supply of water to the roots. The water table level did not reappear until a large storm recharged the lysimeter during late August (Figure 10). Perusal of the lysimeter observations for the dry period in the fall of 1981 (Figure 10) showed a rate of 2.2 mm/day. This was equal to the average evapotranspiration rate of the early spring months under similar temperature conditions. Some data from early fall of 1980 (Figure 8) approximated a rate of 3.0 mm/day and for early winter about 1.5 mm/day in comparison. Watertable reduction during the dry spring season amounts to 17 mm/day for the month of April, and 8 mm/day for May when the general level had substantially subsided. The averaged rate for April-May was about 13 The average watertable reduction over a similar range of depths during September-October was 19 mm/day. Comparison of these data with the corresponding evapotranspiration withdrawals gives ratios of 0.11,0.31, and 0.12 mm evaporated water per mmwatertable drop for April, May, and September-October, respectively. The general rates of lysimeter weight increase and watertable rise during October-November 1981 show a ratio of about 0.1. The above ratios suggests that the specific yield of the soil in the lysimeter is about 10%. The higher value during May probably reflects a relative increase in uptake of water from the unsaturated rooting zone.

PAGE 32

CHAPTER VI CONCLUSIONS 29 Installation and maintenance of a large weighing lysimeter and its instrumentation proved to be troublesome in the poorly-drained flatwoods soil, especially during the lightning-prone humid summer months. Recordings by sensitive automated equipment have to be supplemented with periodic manual manometer readings. Similarly, state-of-the-art electronic weather sensors and datalogger have to be backed up by less elegant but more reliable hygrotherrnographs and by periodic wind and cumulative precipitation measurements. The available data show that cool season evapotranspiration was about 2 mm/day while that for late spring was about 6 mm/day when soil moisture became replenished by drought breaking rains. ?reaiction of the above seasonal rates bfevapotransplratiOri by the Penman method was good for the autumn and spring seasons. Water use during the spring drought lowered the watertable below the rooting zone. Soil water drawn by roots from the unsaturated zone rapidly became limiting for slash pine transpiration until summer rains raised the water table again. I.

PAGE 33

30 LITERATURE CITED Anderson, H. W., M. D. Hoover, and K.G. Reinhart. 1976. Forests and Water: Effects of Forest Management on Floods, Sedimentation, and Water Supply. USDA Forest Service, Pac. SW Forest and Range Exp. S ta., Re p. PS 18. 11 5 p. Bartho1ic, J. F., and D. W. Buchanan. 1976. Disposition of Water from Fruit Crops and Approaches to Increase Water Use Efficiency. Untversity of Florida Water Resourc. Res. Center Public No. 33. Bastos, T. X., and W. H. Smith. 1979. Influence of Pine Forest Removal on Flatwoods Soil Temperature and Moisture Conditions. IMPAC Report Vol. 4, No.3. 28 p. Bechtold, W. A., and R. M. Sheffield. 1980. Forest Statistics for Florida, 1980. USDA Forest Service, SE For. Expt. Sta. Resource Bull. SE-58. Burger, J. A. 1979. The Effects of Harvest and Site Preparation on the Nutrient Budget of an Inteniive1y Managed Southern Pine Forest Ecosystem. Ph.D. Dissertation, Univ. of Florida, Ga i nesvi 11 e, FL. 185 p. Burns, L. A. 1978. Productivity, Biomass, and Water Relations in a Florida Cypress Forest. Ph.D. Dissertation, Univ. of North Carolina, Chapel Hill. 214 p. Campbell, K. L. 1979. Nutrient Loads in Streamflow from Sandy Soils in Florida. Transactions ASAE 22(5):1115-1120. CRIFF. 1977. Cooperative Research in Forest Fertilization: Progress Report 1976-1977. University of Florida, Gainesville, FL, Institute of Food and Agricultural Sciences, School of Forest Resources and Conservation, CRIFF Report 1977:131-140. Dohrenwend, R. E. 1977. Evapotranspiration Patterns in Florida. Florida Scientist 40(2):184-192. Dohrenwend, R. E. 1978. The Climate of Alachua County, Florida. Institute of Food and Agricultural Sciences, Univ. of Florida, Gainesville, FL. 25 p. I.

PAGE 34

31 Douglass, J. E. 1967. Effects of Species and Arrangement of Forests on Evapotranspiration. In IIForest Hydrologyll eds. W. E. Sopper and H. W. Lull, Pergamon Press, New York:1451-1461. Douglass, J. E., and W. T. Swank. 1975. Effects of Management Practices on Water Quality and Quantity: Coweeta Hydrologic Laboratory, North Carolina. In IIMunicipal Watershed Managementll, USDA Forest Service, Gen. Tech. Rep. NE-13:1-13. Duncan, D. V., and T. A. Terry. 1982. Water Management. In liThe Managed Slash Pine Ecosystemll, Ed. R. C. Biesterfeldt, USFS Southeastern Exp. Sta., Ashville, N.C. (In press.) Dylla, A. S., and L. M. Cox. 1973. An Economical Hydraulic Weighing Evapotranspiration Tank. Transact. ASAE 1973:294-295. Faulkner, G. L. 1975. Flow Analysis of Karst Systems with Well-Developed Underground Circulation. In "Karst Hydrology and Water Resourcesll, Proc. U.S.-Yugoslavian Symposium, Dubrovnik; 6.1-6.28. Fritschen, L. J. ,L. Cox, and R. Kinerson. 1973. A 28-meter Douglasfir in a weighing lysimeter. For. Sci. 19:256-261. Fritschen, L. J., J. Hsia, and R. Doraiswamy. 1977. Evapotranspir ation of a Douglas-fir Determined with a Weighing Lysimter. Water Resour. Res. 13(1):145-149. Gray, D. M. 1973. Handbook on the Principles of Hydrology. Water Information Center, Inc., Huntington,New York. 670 Greene, R. E. L., K. Mathis, L. Polopolus, and J. Holt. 1980. Economic Data for Florida Agriculture; 1975-1980. Univ. Florida, Institute of Food and Agricultural Sciences, Gainesville, FL. 152 p. Hewlett, J. D., and A. R. Hibbert. 1963. Moisture and Energy Conditions Within a Sloping Soil Mass During Drainage. J. Geophys. Res. 68(4):1081-1087. Hewlett, J. D., and A. R. Hibbert. 1967. Factors Affecting the Response of Small Watersheds to Precipitation in Humid Areas. In IIForest Hydrologyll, Eds. W. E. Sopper and H. W. Lull, Pergamon Press, New York. Hibbert, A. R. 1967. Forest Treatment Effects on Water Yield. In IIForest Hydrologyll, Eds. W. E. Sopper and H. W. Lull, Pergamon Press, New York. 527-543.

PAGE 35

32 Hibbert, A. R. 1969. Water Yield Changes after Converting a Forested Catchment to Grass. Water Resour. Res. 5:634-640 .. Hicks, B. B., P. Hyson, and C. J. Moore. 1975. A Study of Eddy Fluxes over a Forest. J. Applied Meterology 14:58-66. Holdridge, L. R. 1967. Life Zone Ecology. Tropical Science Center, San Jose, Costa Rica. 206 p. Hutchinson, B. 1981. Running on Empty. New Florida Special Report, Oct. 1981, Tequesta, FL. 8 p. IFAS. 1980. Florida Water Resources NAS 10-9348: Final Report. Institute of Food and Agricultural Sciences, Univ. Florida. 311 p. Jackson, T. J., T. J. Schmugge, L. H. Allen, and others. 1981. Aircraft Remote Sensing of Soil Moisture and Hydrologic Parameters, Taylor Creek, FL., and Little River, CA; 1979 Data Report. USDA Agric. Research Service, Agric. Res. Results 13. 36 p. Kaufman, C. M., W. L. Pritchett, and R. E. Choate. 1977. Growth of Slash Pine on Drained Flatwoods. Univ. Florida, Florida Agric. Exp. Sta. Bulletin 792. 30 p. Kaufman, M. R. 1981. Automatic Determination of Conductance, Transpiration, and Environmental Conditions in Forest Trees. Forest Science 27(4):817-827. Klawitter, R. A., and C. E. Young, Jr. 1965. Forest Drainage Research in the Coastal Plain. J. Irr. and Drainage. Div. ASCE, Vol. 91, No. IR 3, Proc. Paper 4456. 1-7. Klein, H. J., T. Armbuster, B. F. McPherson, and H. J. Freiberger. 1975. Water and the South Florida Environment. U.S. Geol. Survey Water Resour. Invest. 24-75. 165 p. Knisel, W. G., Jr., P. Yates, J. M. Sheridan, T. K. Woody, III, L. H. Allen, Jr., and L. E. Asmussen. 1982. Hydrology and Hydrogeology of Taylor Creek Watershed, Florida: Data and Analysis. S.E. Watershed Research Program, USDA Agricultural Research Service Technical Bulletin. (In press.) Krygier, J. T. 1971. Comparative Water Loss of Douglas-fir and Oregon White Oak, Oregon State Univ., Water Resour. Res. Instit, and School of For. Resour. Res. A-0001-OREG. Law, F. 1957. Measurement of Rainfall, Interception and Evaporation Losses in a Plantation of Sitka Spruce Trees. I.A.S.H., Gen. Ass., Toronto Proc., Vol. 2. 377-411

PAGE 36

Lee, R. 1970. Theoretical Estimates versus Forest Yield. Water Resour. Res. 6:1327-1334. 33 Lee, R. 1978. Forest Microclimatology. Columbia University Press, New York. 276 p. Lee, R. 1980. Forest Hydrology. Columbia University Press, New York. 349 p. Lynne, G. D., and C. F. Kiker. -An Economic Perspective. Report No. 82. 70 p. 1976. Water Use in South West Florida Univ. Florida, Gainesville, Economics Maloney, F. E. 1980. Legal Principles of Florida Water Management. Business and Economic Dimensions 16(1):16-26. Maxwell, J. 1976. The Effects of Alder Conversion on Streamflow. Sius1aw National Forest, NW Forest Sciences Lab, Corvallis, OR. 4 p. Mustonen, S. E., and J. L. McGuinness. 1967. Lysimeter and Watershed Evapotranspiration. Water Resour. Res. 3(4):989-996. Metz, L. J., and J. E. Douglass. 1959. Soil Moisture Depletion Under Several Piedmont Cover Types. USDA, Tech. Bull. 1207. 23 p. Monteith, J. L. 1965. Evaporation and Environment. In liThe State and Movement of Water in Li vi ng Organi sms. II Ed G. E. Fogg. Academic Press, New York. 205-234. Parker, G. G., G. E. Ferguson, S. K. Love, and others. 1955. Water Resources of Southeastern Florida. U.S. Geol. Survey, Water Supply Paper 1255. 965 p. Patric, J. H. 1974. Water Relations of Some Lysimeter-Grown Wildland Plants in Southern California. USDA Forest Serv, N.E. Forest Exp. Sta., Upper Darby, PA. 134 p. Penman, H. L. 1948. Natural Evaporation from Open Water, Bare Soil, and Grass. Proc. Royal Soc (London) Vol. 193:120-145. Penman, H. L. 1963. Vegetation and Hydrology. Commonwealth Agric. Bur., Farnham Royal, Bucks, England, Commonwealth Bureau of Soils, Tech. Commun. 53.

PAGE 37

34 Pritchett, W. L. 1980. Properties and Management of Forest Soils. John Wiley and Sons, New York. 500 p. Riekerk, H. 1981. Impacts of Silviculture on Flatwoods Runoff, Water Quality, and Nutrient Budgets. In "Progress in Wetlands Utilization and Management" Ed. P. M. McCaffrey. Coord. Council Ki ss immee River Valley and Taylor Creek-Nubbi ns Slough, Tallahassee, FL. (In press.) Riekerk, H., S. A. Jones, L. A. Morris, and D. A. Pratt. 1978. Hydrology and Water Quality of Three Small Lower Coastal Plain Forested Watersheds. Soil and Crop Sci. Soc. Florida 38:105-112. Riekerk, H., B. F. Swindel, and J. A. Replogle. Forestry Practices in Florida Watersheds. ment 180" Am. Soc. Civil Engin., New York. 1980. Effect of In "Watershed Manage706-721. Riekerk, H., L. F. Conde, J. C. Hendrickson, and W. S. Gain. 1981. Research on Environmental and Site Effects of Forest Management Practices in the Lower Coastal Plain. In "First Biennial Southern S11 vicul tural Research Conference II Ed. J. P. Barnett. USDA Forest Service, S. For. Exp. Sta., Gen. Tech. Rep. SO-34:331-339. Riekerk, H., and D. K. Winter. 1982. Forest Water Management Principles in Florida. In "Environmentally Sound Water and Soil Management", Eds. E. G. Kruse and Y. A. Yousef. ASCE, Irrig. and Drain. Div. 255-267. Robins, J. S. 1965. Evapotranspiration. In "Methods of Soil Analysis Part I", Ed. C. A. Black, Agronomy Series No.9, Am. Soc. Agronomy Inc., Madison, WI. 286-298. Rodda, J. C. 1976. Facets of Hydrology. John Wiley and Sons, New York. 368 p. Rosenzweig, M. L. 1968. Net Primary Productivity of Terrestrial Communities Prediction from Climatological Data. Amer. Nat. 102:67-74. Rutter, A. J. 1967. Analysis of Evaporation from a Stand of Scots Pine. In "Forest Hydrology" Eds. W. E. Hopper and H. W. Lull. Pergamon Press, New York. Schiess, P. 1977. Wastewater Utilization in Forest Soil. In "Municipal Wastewater and Sludge Recycling on Forest Land and Disturbed Land." Eds. W. E. Sopper and S. N. Kerr, Pennsylvania State Univ. Press l.

PAGE 38

35 Speir, W. H., W. C. Mills, and J. C. Stephens. 1969. Hydrology of Three Experimental Watersheds in Southern Florida. USDA, Agric. Research Service, ARS-41-l52. 50 p. Stanhill, G. 1969. A Simple Instrument for the Field Measurement of Turbulent Diffusion Flux. J. Applied Meteorology 8(4):509-513. Stone, E. L., W. T. Swank, and J. W. Hornbeck. 1978. Impacts of Timber Harvest and Regeneration Systems on Streamflow and Soil s in the Eastern Deciduous Region. In "Forest Soil s and Land Use" Ed. C. T. Youngberg, Dept. Forest and Wood Sciences, Colorado State Univ. Ft. Collins, CO. 516-536. Swank, W. T., and J. E. Douglass. 1974. Streamflow Greatly Reduced by Converti ng Dec i duous Hardwood Stands to Pi ne. Sci ence 185: 857-859. Swank, W. T., and J. D. Helvey. 1970. Reduction of Streamflow Increases Following Regrowth of Clear cut Hardwood Forests. Intern. Assoc. Sci. Hydrol., Publ. 96:346-360. Swindel, B. F., C. J. Lassiter, and H. Riekerk. 1982. Effects of Clearcutting and Site Preparation on Water Yields from Slash Pine Forests. For. Ecol. and Management. 101-114. Thorntwaite, C. W. 1948. An Approach Toward a Rational Classification of Climate. Am. Geographical Review Vol. 38:55-95. Turk, L. J. 1975. Diurnal Fluctuations of Water Tables Induced by Atmospheric Pressure Changes. Journal Hydrology 26:1-16. Turner, R. R., T. M. and R. C. Harriss. 1977. Descriptive Hydrology of Three North Florida Watersheds in Contrasting Land Use. In "Watershed Research in Eastern North America" Ed. D. L. Correll, Chesapeake Bay Center for Environmental Studies, Smithsonian Institution, Edgewater, Maryland. 211-227. Ursic, S. J. 1974. Pine Management Influences the Southern Water Resource. In "Symposium on Management of Young Pines" Eds. H. L. Williston and W. E. Balmer. USDA Forest Service, SE For. Exp. Sta. Athens, GA. 42-49. Van Bavel, C. H. M. 1961. Lysimetric Measurements of Evapotranspiration Rates in the Eastern United States. Soil Sci. Proc. 25(2) :138-142.

PAGE 39

36 Van Bavel, C. H. M. 1966. Potential Evaporation: The Combination Concept and Its Experimental Verification. Water Resour. Res. 2:455-467. Van Lill, W. S., F. J. Kruger, and D. B. Van Wyk. 1980. The Effect of Aforestation with Eucalyptus grandis and Pinus patula on Streamflow from Experimental Catchments at Mokobulaan, Transvaal. J. Hydrology 48:107-118. Van Wyk, W. R. 1967. Forest Hydrology Research in the Netherlands. In "Forest Hydro10gy" Eds. W. E. Sopper and H. W. Lull, Pergamon Press. 65-67. Verry, E. S. 1976. Estimating Water Yield Differences Between Hardwood and Pine Forests: An Appl ication of Net Precipitation Data. USDA Forest Service, N.C. For. Exp. Sta., St. Paul ,Minn., Res. Paper NC-128. 12 p. Wells, C. G., and D. M. Crutchfield. 1974. Intensive Culture of Young Loblolly Pine. In "Symposium on Management of Young Pines" Eds. H. L. Williston and W. E. Balmer. USDA Forest Service, S&SE For. Exp. Sta., Athens, GA. 212-229. White, E. H., and W. L. Pritchett. 1970. Watertable Control and Fertilization for Pine Production in Flatwoods. Univ. of Florida, Florida Agric. Exp. Sta. Tech. Bull. 743. 41 p. White, W. N. 1932. A Method of Estimating Ground Water Supplies Based on Discharge by Plants and Evaporation from Soil. U.S. Geol. Survey Water Supply Paper 659. 105 p. Williams, T. M., and D. J. Libscomb. Cutting on Coastal Plain Soils. 49. 1981. Watertable Rise After S. Journ. Applied For. 5(1):46Woodall, S. L. 1980. Evapotranspiration and Melaleuca. In "Mela1euca Symposium" Ed. R. K. Geiger. Florida Division of Forestry, Coll ins Bldg. Tallahassee, FL: 117-125. Young, C. E., and R. H. Brendemuehl. 1973. Response of Slash Pine to Drainage and Rainfall. USDA Forest Service SE Forest Exp. Sta., Athens, GA, Res. Note SE-186. I.

PAGE 40

APPENDIX CONTINUOUS SYSTEM MODELING PROGRAM VERSION 1.3 Title-Calculation of pet and potential rainfall deficit Initial 37 / Dimension RSI (365), TEMP (365), DEWP (365), WIND (365), RAIN (365) / Dimension DAY (365) FIXED I INCON Inc = 0.0 CONST PYSCH = 0.27 CONST ALB = 0.88 CONST BOLTZ = 1.9978E-9 FUNCTION VAPCUR = (-20.0, 0.776), (-15.0, 1.436), (-10.0,2.149), ... (-5.0,3.163), (0.0,4.579), (5.0, 6.543), (10.0,9.209), (15.0,12.788), (20.0,17.535), (25.0,23.756), (30.0, 3l.824), :.. .. (35.0,42.175), (40.0, 55.324) FUNCTION SI = (0.0,510.0), (30.0,580.0), (60.0,710.0), (90.0,840.), ... (120.0,925.0), (150.0, 990.0), (180.0, 1005.0), (210.0, 950.0), ... (240.0, 885.0), (270.0, 760.0), (300.0, 630.0), (330.0, 520), (365.0, 500.0) NOSORT READ (5, 1000) (DAY (I), RSI (I), TEMP (I), DEWP (I), WIND (I), RAIN (I), I = (1,365) 1000 FORMAT (F3.0, F6.0, F6.1, F6.1, F5.1, F6.2) DYNAMIC NOSORT I = TIME + 1. 0 DATE = DAY (I) DEWPC = (DEWP (I) -32.))*0.5555 TEMPO = (TEMP (I) -32.0)*0.5555 TEMPA = TEMPO + 0.2 TEMPB = TEMPO 0.2 ABST = TEMPO + 273.0 ETA = NLFGEN (VAPCUR, TEMPO) EST = NlFGEN (VAPCUR, DEWPC) SLPVAP = (NlFGEN (VAPCUR, TEMPA) -NlFGEN (VAPCUR, TEMPB)) /0.72 NAPDEF = 0.35*(ETA EST)*(1.0 + 0.24*WIND(I)) NN = RSI (I)?NLFGEN(SI,DATE)) -0.18)/0.55 SHRTWV = RSI(I) ALB /58.0 LONGWV = BOLTZ*ABST**4.0*(O.56 0.092*EST**0.5)*(0.1 + 0.9*NN) H = SHRTWV -LONGWV PET = SLPVAP/PYSCH)*H + VAPDEF)/ SLPVAP/PYSCH) + 1.0) TERMINAL METHOD RECT PRTPLT PET (DATE, VAPDEF, H) PRTPLT TEMPO (DATE) TIMER DELT = 1.0, FINTIM = 214.0,OUTDEL = 1.0 END