Biometry, Modeling & Statistics
Two-Year Growth Cycle Sugarcane Crop Parameter Attributes
and Their Application in Modeling
Manyowa N. Meki,* Jim R. Kiniry, Adel H. Youkhana, Susan E. Crow, Richard M. Ogoshi, Mae H.
Nakahata, Rebecca Tirado-Corbalá, Ray G. Anderson, Javier Osorio, and Jaehak Jeong
ABSTRACT
The renewed interest in the use of sugarcane (Saccharin officinarum L.) for biofuel could provide a viable market for potential Hawaiian sugarcane feedstock producers. In Hawaii, sugarcane is grown as an irrigated 2-yr cycle crop. There is however little information
on crop parameter attributes of 2-yr cycle sugarcane. This field study on Maui, Hawaii, analyzed the relationship between sugarcane
biomass accumulation and specific crop parameters. Overall, the high dry biomass yield (80.20 Mg ha–1) was the result of a high leaf
area index (LAI, 7.50) and radiation use efficiency (RUE, 2.06 g MJ–1. The crop growth rate was highly correlated to LAI (R 2 , 0.86),
and a light extinction coefficient (k) of 0.53 was estimated. Stalk density was estimated at 18 stalks m–2 , with a maximum plant height
of 3.6 m, and a rooting depth exceeding 2.0 m. When the crop parameters were incorporated into a biological model of Agricultural
Land Management Alternatives with Numerical Assessment Criteria (ALMANAC) the model accurately simulated sugarcane yields
across seven different soil types and multiple management scenarios of applied irrigation water, N and P fertilizer inputs and various
planting and harvest dates. The mean simulation percent (%) errors ranged from –6.4% to 1.8%, while the calculated Fisher’s paired
t test of 1.41 with 39 degrees of freedom, showed no significant differences (P ≥ 0.05) between measured and simulated yields. The
ALMANAC model should be useful as a decision support tool for evaluating sugarcane management alternatives that maximize yields
while optimizing water, N and P inputs.
Sugarcane production in Hawaii has declined since the
1970s due to a number of factors that include low prices, high
labor costs, competition from artificial sweeteners, and low
cost production from such countries as Mexico, Brazil, India,
and China. Despite these challenges, the Hawaii Commercial
& Sugar Company (HC&S) continues to produce approximately 181,400 Mg yr–1 of raw sugar on 14,164 ha of land. This
company is the last remaining sugarcane plantation in Hawaii
and is based on Maui Island. Sugarcane production plays a
major role in the island’s economy, employing more than 800
people. In addition, the company generates its own electricity
and enough energy to meet up to 8% of Maui’s total electricity
needs through the burning of the residual biomass (bagasse),
which is used internally or sold to the county’s electricity
grid, (HC&S, 2011). Recently, competition for scarce water
resources coupled with declining precipitation have become the
major constraints to sugarcane productivity and profitability
at HC&S. Sugarcane needs 1500 to 2500 mm of water evenly
distributed over the growing season (FAO, 2013).
Faced with these multiple challenges, HC&S in collaboration with USDA-ARS implemented a joint project plan to
explore the feasibility and sustainability of producing advanced
Published in Agron. J. 107:1310–1320 (2015)
doi:10.2134/agronj14.0588
Copyright © 2015 by the American Society of Agronomy, 5585 Guilford
Road, Madison, WI 53711. All rights reserved. No part of this periodical
may be reproduced or transmitted in any form or by any means, electronic or
mechanical, including photocopying, recording, or any information storage
and retrieval system, without permission in writing from the publisher.
1310
biofuel feedstocks as alternatives to raw sugar production in
Hawaii. Among other candidate biofuel crops, sugarcane is
an ideal feedstock crop for ethanol production in Hawaii due
to its rapid growth, moisture stress tolerance, and high yields
(Kinoshita and Zhou, 1999; Grantz and Vu, 2009; Laclau
and Laclau, 2009). Furthermore, sugarcane has advantages
M.N. Meki, J. Osorio, and J. Jeong,Texas A&M AgriLife Research,
Blackland Research and Extension Center, 720 E. Blackland Rd,
Temple, TX 76502; J.R. Kiniry, USDA, Agricultural Research Service,
Grassland Soil and Water Research Laboratory, 808 E. Blackland
Rd, Temple, TX 76502; A.H. Youkhana, and S.E. Crow, Dep. of
Natural Resources and Environmental Management, Univ. of Hawaii
at Manoa, 1910 East-West Rd., Honolulu, HI 96822; R.M. Ogoshi,
Dep. of Tropical Plants and Soil Sciences, Univ. of Hawaii at Manoa,
563 Kaumakani Street, Honolulu, HI 96825; M.H. Nakahata, Hawaii
Commercial & Sugar Company, 1 Hansen Rd, Puunene, HI 96784; R.
Tirado-Corbalá, Crops and Agro-Environmental Science Dep., Univ.
of Puerto Rico, Mayagüez, Puerto Rico; and R.G. Anderson, USDA,
Agricultural Research Service, U.S. Salinity Laboratory, Contaminant
Fate and Transport Unit, George E. Brown Jr. Salinity Lab., 450 W.
Big Springs Rd., Riverside, CA 92507-4617. Received 18 Nov. 2014.
Accepted 4 Feb. 2015. *Corresponding author (nmeki@brc.tamus.edu).
Abbreviations: ALMANAC, Agricultural Land Management
Alternatives with Numerical Assessment Criteria; APEX, The
Agricultural Policy/Environmental eXtender; DM, dry matter;
EPIC, Environmental Policy Integrated Climate; FIPAR, fraction
of intercepted photosynthetically active radiation; HC&S, Hawaii
Commercial & Sugar Company; k, light extinction coefficient; LAI,
leaf area index; PAR, photosynthetically active radiation; PHU,
potential heat units; RUE, radiation use efficiency; SWAT, Soil and
Water Assessment Tool.
A g ro n o my J o u r n a l
•
Vo l u m e 107, I s s u e 4
•
2 015
over other potential biofuel feedstock crops because (i) it is
a well-established crop known by both current and previous
producers, (ii) it already has an existing sound production and
agricultural marketing infrastructure, and (iii) it has a wellestablished greenhouse gas reduction potential (Goldemberg,
2008; Wang et al., 2012). This renewed interest in the use of
sugarcane for biofuel could provide a viable market for potential Hawaiian sugarcane feedstock producers.
At HC&S, and previously elsewhere in Hawaii, sugarcane is
grown as an irrigated 2-yr cycle crop. The 2-yr cycle cane and
sugar yields (160 Mg ha–1 cane yield and 16% sugar yield) are
above the national average (79 Mg ha–1 cane yield and 12%
sugar yield) (NASS, 2013). A recent study by Anderson et al.
(2015) also confirmed that Hawaiian sugarcane has a higher
net productivity and overall biomass yield than sugarcane
grown in other regions of the world. There is however little
quantitative information describing the relationship between
2-yr cycle sugarcane biomass yield to important crop parameter
attributes that can be utilized in dynamic crop modeling decision support tools to optimize crop management practices and
to explore more complex issues of sustainability and long-term
environmental impacts of management alternatives. Most studies have tended to focus mainly on the relationship between
sugarcane biomass yield and LAI, light interception and radiation use efficiency (RUE) (Bull and Tovey, 1974; Donaldson
et al., 2008; Muchow et al., 1994; Robertson et al., 1996;
Anderson et al., 2015).
With this in mind, we conducted a field study at the HC&S
plantation to (i) collect data for analyzing the relationship
between sugarcane biomass accumulation and specific crop
parameter attributes, such as LAI, light extinction coefficient
(k), fraction of intercepted photosynthetically active radiation
(FIPAR), RUE, maximum plant height and rooting depth,
and stalk density, and (ii) evaluate the accuracy of the developed crop parameters in calibrating the biological model of
ALMANAC (Kiniry et al., 1992) for simulating 2-yr cycle
sugarcane management practices and yields. The model has
been extensively validated and applied to analyze crop grain
and bioenergy feedstock yields, plant community dynamics,
phenology, water, nutrient and RUEs, hydrology, erosion, soil
organic carbon, and nutrient cycling (Behrman et al., 2013;
Engel et al., 2010; King et al., 1998; Kiniry et al., 1996, 2005,
2008, 2012; Meki et al., 2013; Persson et al., 2011).
MATERIALS AND METHODS
Study Site and Cultural Practices
Sugarcane crop data was collected from a field experiment
conducted at HC&S plantation, Field no. 609 (20.89° N,
–156.41° W; elevation ~ 100 m asl). The field soil belongs to
the Molokai series, and is a kaolinitic, isohyperthermic Typic
Eutrotorox (USDA NRCS, 2011). The soil is very-fine, well
drained and has deep, well-defined horizons below the plow
layer. Field plots were established on 26 June 2011 for the
24 mo cropping cycle ending May 2013. The field had been
continuously cropped with sugarcane for well over 100 yr.
A repeated measures design with three replicates was used.
Overall plot sizes were approximately 15 by 10 m. The space
between plots was 3 m. The sugarcane plots were planted with
seed cane variety H65-7052 (Heinz et al., 1981). Seed billets
Agronomy Journal
•
Volume 107, Issue 4
•
2015
were planted by a machine that in one pass digs the furrow,
drops the cane pieces, injects the irrigation tubing, and covers the furrows. Each plot was made up of six rows with a row
planting scheme consisting of two narrow rows at 0.91 m apart
and an interval row spacing between each two narrow rows
of 1.82 m. Drip irrigation lines were laid out in the middle of
each of the two narrow rows (Fig. 1). All plots were adequately
fertigated with urea (46–0–0) at a rate of 345 kg N ha–1.
At HC&S fertilizers are applied only during the first year of
growth. In total 3520 mm of water was drip-applied during the
2-yr growth cycle. A pre-emergence herbicide mix containing
atrazine (1-Chloro-3-ethylamino-5-isopropylamino-2, 4, 6-triazine), 2, 4-D (2, 4-Dichlorophenoxyacetic acid), Prowl ((N-1ethylpropyl)-3, 4-dimethyl-2, 6-dinitrobenzenamine), Rifle (3,
6-dichloro-2-methoxybenzoic acid), and Velpar (3-cyclohexyl6-dimethylamino-1-methyl-1, 3, 5-triazine-2, 4(1H,3H)-dione)
was applied to control weeds during the early establishment
stage. Crop parameter data was collected from the two middle
narrow rows with the two narrow outer rows acting as borders.
No water or N stress conditions were observed throughout the
crop growth cycle and hence the plants were able to express
their full genetic potential. Soil moisture was monitored with a
Stevens Hydra Probe II soil sensor (Stevens Water Monitoring
Systems, Inc., Portland, OR).Weather data was collected at the
field site by a portable automatic weather station (HOBO logger model H-21, Onset Computer, Bourne, MA).
For model simulations the length of the growing season at
the study site was calculated in accumulated potential heat
units (PHU) (sometimes referred to as growing degree days)
using the equation:
PHU = S [(Tmax. + Tmin.)/2 – T base]
[1]
where Tmax. and Tmin. are the averages of the daily maximum
and minimum temperatures compared to a base temperature,
T base, in °C (10°C for sugarcane and many other tropical crops).
Field Measurements
Sugarcane cultivars grown in Hawaii over a 24 mo growth
cycle have two distinct growth patterns. During the first 12 mo
the cultivars exhibit a growth pattern similar to that of 12 to
18 mo cultivars grown on the U.S. mainland and elsewhere in
the world. Plants grow tall and straight with little or no lodging
(Fig. 2a). After 12 mo the plants start to lodge due to the high
biomass accumulation and in some cases lodging is exacerbated
by high winds (Fig. 2b).
Lodging seems to have the positive benefit of exposing the
sugarcane plants to more sunlight and hence allowing for more
biomass and sugar accumulation (Mae Nakahata, HC&S
agronomist, personal communication, 2011). Biomass sampling of the lodged crop does however present some challenges
when matching harvested biomass to corresponding LAI and
intercepted light measurements. To overcome this challenge we
devised two protocols for biomass sampling, LAI determination and taking light interception measurements during the first
12 mo (Protocol I), and thereafter from 12 to 24 mo (Protocol II).
Protocol I (0–12 mo). Monthly biomass samplings were made
during the first 12 mo of the growth cycle. We used a rectangular PVC pipe frame open on one end (1.25 by 0.8 m = 1.0 m2)
1311
canopy PAR measurements were taken 10 cm above the ground
while moving the ceptometer across the plant row and along the
1.25 m length of the frame. Sampled row areas were chosen at
random but making sure they were not next to any previously
sampled area. The FIPAR was calculated with the mean of at
least five PAR measurements above and below the canopy.
A representative subsample of three to five cane stalks (based
on visual observations of plant height) was harvested from
within the frame for LAI determination in the laboratory.
The rest of the plant material within the frame was harvested
and weighed in the field to get the total weight. In the laboratory, the subsample was weighed and then separated into green
leaves, dead leaves, green photosynthetic stalks and brown
non-photosynthetic stalks. The leaf area was determined with
a LiCor LI-3100 leaf area meter (LiCor Inc., Lincoln, NE).
The LAI was calculated as the area of the subsample multiplied
by the ratio of the total fresh weight/subsample fresh weight
divided by sampled ground area. An important model parameter associated to LAI is the rate of LAI decline between the
maximum LAI and harvest time–between 15 to 24 mo. This
parameter was determined by systematically adjusting model
pre-set values; 1.0 is linear; >1 accelerates decline; <1 retards
the decline rate, until the simulated decline rate matched the
measured LAI data values. The k values for each biomass harvest stage were calculated by changing Beer’s law, as originally
described by Monsi and Saeki’s (1953):
Fig. 1. Plot layout showing sugarcane row spacing in relation to drip
line placement.
(Fig. 1 and also see pointed arrow in Fig. 2a) to demarcate the
sampling area. The frame was placed on the ground with the
long edge (1.25 m) onto and parallel to the drip line and plant
rows as shown in Fig. 1 and 2. The plant height was measured
from the base of the stool (i.e., sugarcane shoots growing out of
buds on the joints of the billets) to the top of the longest leaf. The
stalks within the frame area that were greater than one-fourth
of the maximum plant height were recorded for estimation of
stalk density. The PAR interception was measured with a 0.8
m AccuPAR LP-80 Ceptometer (Decagon, Pullman, WA)
between 1000 and 1300 h–the times when incident solar radiation was usually relatively stable. Two ceptometers were used
to take concurrent above and below canopy PAR readings after
“test matching” PAR readings from both ceptometers on an
open space. These readings should match or have a difference of
<100 µmol m–2 s–1. When they differed by more, we recalibrated
the sensors and took the measurements again. The open space
PAR readings represent the above canopy PAR readings. Below
Fig. 2(a) Eight-month sugarcane showing no signs of lodging, and
(b) Lodged sugarcane at 23 mo lodging can be exacerbated by
high winds.
1312
FIPAR = 1– exp(–k ×LAI)
[2]
Into:
k = [Ln (1.0 – FIPAR)]/LAI
[3]
Subsample total dry matter was measured after drying all
harvested plant materials in a forced-air drying oven at 70°C to
constant weight.
Protocol II (12–24 mo). In 2-yr growth cycle sugarcane
biomass accumulation slows down (reduced growth phenomena) in the second year even though growth conditions such
as water availability, nutrient status, and temperature are all
favorable (Muchow et al., 1997; Park et al., 2005). Hence biomass samplings after the first year of growth were only made
every 2 mo. As pointed out earlier, this growth period presents
many challenges in accounting for all the trash (crop residues
composed of plant tops and dry leaves that are left above the
ground; these residues form a mat known in the sugarcane
industry as trash) biomass from senesced leaves and stalks for
inclusion in the calculation of the total aboveground biomass
and accurately estimating RUE. Furthermore, lodging of the
cane (Fig. 2b) makes relating this biomass to the corresponding LAI and FIPAR difficult. To this end, we applied the
following modification to Protocol I; The rectangular frame
was used to demarcate the sampling area as in Protocol I. The
plant height was measured and below canopy PAR interception measurements taken at a height just below the green leaves
(~1 m above the ground). A representative subsample of three
to five cane stalks was harvested from within the frame for LAI
determination as detailed in Protocol 1. However, and unlike
in Protocol I, the corners of the frame were pinned off using
Agronomy Journal
•
Volume 107, Issue 4
•
2015
four 30-cm metal pins. To accurately estimate stalk density,
sugarcane stalks originating from outside the pinned area were
slashed and cleared to enable easy recording of only those stalks
originating from within the pinned area. All the plant material
within the frame–stalks and cane trash mat was cut, collected,
and weighed in the field. A representative subsample of the
stalks and cane trash mat were collected, weighed in the lab,
and oven-dried to constant weight at 70°C. This subsample dry
matter weight was used to estimate the total dry biomass yield.
The RUE was calculated as the slope of the regression equation of the sugarcane aboveground dry matter (including the
trash cane mat) and accumulated intercepted PAR.
ALMANAC Model Description
For a more detailed description of the ALMANAC model,
refer to Kiniry et al. (1992). The model was developed for use in
field management; several fields may be simulated to comprise
a whole farm up to about 100 ha. The model can be used to
compare management systems and their effects on yields, N,
P, C, pesticides, and sediment. The management components
that can be changed include irrigation scheduling, crop rotations, tillage operations etc. In summary, the model uses a daily
time step to simulate various biophysical processes including
plant growth. Radiation interception and RUE are the drivers of plant growth or biomass accumulation which are in
turn a function of the LAI and k for FIPAR as described by
Monsi and Saeki’s (1953) Beer’s law in Eq. [2] above. RUE is
a function of the vapor pressure deficit and atmospheric CO2
(Kemanian et al., 2004; Stöckle and Kiniry, 1990), while LAI
evolution is simulated with a daily heat unit system that correlates plant growth with temperature.
To accurately estimate total plant biomass the model takes
into account the contribution of root biomass in calculations
of RUE. According to Jones (1985) the fraction of total biomass partitioned to the root system of most crops normally
decreases from 0.30 to 0.50 in the seedling to 0.05 to 0.20 at
maturity. The model simulates this partitioning by decreasing
the fraction linearly from emergence to maturity. We applied
the maximum root biomass fraction at maturity of 0.20. Daily
incident PAR values were taken as 45% of the total solar radiation (Meek et al., 1984; Monteith, 1965) which was measured
on-site with the portable automatic weather station.
Other important model parameters included maximum
plant height and rooting depth, percent moisture content in
biomass at harvest, N and P nutrient contents at early, mid- and
late crop growth stages. Various stresses that include temperature, soil moisture, plant nutrients (N and P), aeration, salinity, pH, and soil compaction limit plant growth (Kiniry et al.,
1992; Meki et al., 2013).
Model Calibration and Crop
Parameter Fine-tuning
A new crop for 2-yr growth cycle sugarcane was created in the
ALMANAC Crops database by modifying crop parameter data
for a 1-yr growth cycle sugarcane crop from the Environmental
Policy Integrated Climate (EPIC) model (Williams, 1995) with
our field gathered data. The ALMANAC and EPIC models
have similar crop growth models but each also has subroutines
describing different biophysical processes: both models contain
Agronomy Journal
•
Volume 107, Issue 4
•
2015
Table 1. Actual sugarcane field management operations and inputs
for field 609 that were simulated with the ALMANAC model.
Date of
Field operations
Irrigation Fertilizer†
operation
mm ha–1 kg N ha–1
Subsoiler, deep ripper,
1 June 2011
depth: 40 cm
Harrow, finisher,
5 June 2011
depth: 8–15 cm
–2
Planting, 18 stalks m
26 June 2011
150
22
17 July 2011
210
79
4 Aug. 2011
250
55
23 Sept. 2011
216
45
28 Oct. 2011
128
2 Nov. 2011
148
34
5 Dec. 2011
128
11
4 Jan. 2012
198
11
8 Feb. 2012
149
7 Mar. 2012
170
88
27 Apr. 2012
183
9 May 2012
287
13 June 2012
216
4 July 2012
126
15 Aug. 2012
265
8 Sept. 2012
293
3 Oct. 2012
197
7 Nov. 2012
207
2 Dec. 2012
Harvesting (Harvest
21 May 2013
Index = 0.90)
† Fertilizer was applied as liquid urea (46–00–00) through the drip
irrigation lines. Fertilizer amounts shown in table represent the elemental N equivalent.
detailed functions for water balance, nutrient cycling, and plant
growth, but with ALMANAC having additional detail for light
competition, population density effects, and vapor pressure
deficit effects which enable it to simulate the growth and seed
yield of two competing plant species in a wide range of environments (Kiniry et al., 1992). On the other hand, the EPIC model
has since evolved into a comprehensive agro-ecosystem model
with a soil organic carbon (SOC) module adapted from the
CENTURY model (Parton, 1996) capable of simulating SOC
dynamics in a wide range of plant species, including crops, native
grasses, and trees (Izaurralde et al., 2006).
Using the new 2-yr growth cycle sugarcane crop we then compiled a simulation budget (Table 1) to model sugarcane growth
in the experiment from which we gathered the crop parameters
(Field no. 609). We ran the model simulation using actual 2011
to 2013 weather summarized in Fig. 3 and the NRCS USDA
Soil Survey Geographic (SSURGO) soils data for a Molokai soil
series (Table 2). To refine the crop parameters, we systematically adjusted the new crop parameters iteratively by running
the ALMANAC model, comparing the simulated LAI and
aboveground dry biomass yields to actual data, and then altering crop parameter values until there was a good match between
predicted and measured LAI and dry matter yield values.
1313
Fig. 3. Mean monthly temperature and precipitation for Field no. 609:
June 2011 to May 2013.
Crop Parameter and Model Evaluation
To evaluate the developed crop parameters and test the
calibrated ALMANAC model’s ability to accurately simulate
2-yr cycle sugarcane management practices and yields, we
applied the model to simulate historical sugarcane yields across
7 HC&S fields; F no. 202, 308, 312, 416, 601, 719, and 905
over five to six 2-yr growth cycles. These fields were chosen
primarily based on different soil types and the availability of
the nearest weather station data (Table 3). Before conducting the simulations, historical management information and
data on field preparations, applied N and P fertilizer (N and P
fertigation), and drip irrigation amounts for each of the seven
fields were processed into monthly formats for input into the
ALMANAC management database. The simulated crop budgets for the seven fields are summarized in Table 3. For most
simulated 2-yr growth cycles we noted that the model often
required supplemental P additions to attain actual measured
sugarcane yields. As a result we opted to use the automatic
P fertilizer trigger option in the model, instead of the actual
applied P. When the plant suffers any P stress, the automatic
option triggers P fertilization to achieve an optimum soil P level.
Customized weather data from the HC&S weather station nearest each field and the corresponding SSURGO soils
data were used for the simulations. We applied the Penman–
Monteith method option in the model as it best accounts for
proper water balance and evapotranspiration rates under the
windy conditions of the island and is critical for estimation of
actual ET in such environments (Osorio et al., 2014).
Data Analysis
The REG procedure in SAS version 9.3 (SAS Institute,
2007) was used to conduct regression analyses to describe the
relationship between FIPAR vs. LAI, aboveground dry biomass vs. accumulated intercepted PAR, crop growth rate vs.
LAI, and simulated vs. measured yields. Differences between
measured and simulated dry biomass yields were assessed using
Fisher’s Paired t test. We applied mean separation statistics–
standard errors for the mean at P ≤ 0.05 probability levels
(Snedecor and Cochran, 1989) to determine any significant
differences between measured variables.
RESULTS AND DISCUSSION
Recorded Meteorological Data for Field
Number 609
Mean monthly temperature and precipitation for Field no.
609 showed normal trends for this site for June 2011 to May
2013 (Fig. 3). Mean annual air temperature and precipitation
1314
Table 2. Initial soil conditions for a few selected soil variables
for the simulated Molokai (MUA) silty clay loam, slope: 0 to
3% (Field no. 609). The soil properties were compiled from the
USDA NRCS SSURGO database: http://sdmdataaccess.nrcs.
usda.gov/ (accessed March 2014).
Soil layer number
Soil property
1
2
Depth, m
0.00–0.32 0.32–1.58
0.54
0.50
Porosity, m m–1
0.47
0.45
Field capacity, m m–1
0.30
0.29
Wilting point, m m–1
32.40
32.40
Saturated conductivity, mm h–1
1.23
1.33
Bulk density 33kpa, T m–3
1.29
1.29
Bulk density (Oven dry soil), T m–3
Sand, %
6
6
Silt, %
44
44
Clay, %
50
50
pH
7.20
7.80
10.50
10.50
CEC, cmol kg–1
Soil organic carbon, %
1.62
0.77
during the study period were 23.4°C (range 19.7–29.0°C) and
241 mm, respectively. Mean daily relative humidity was 76%
(range 67–91%), while the average daily incident solar radiation
was 21.5 MJ m–2 d–1 (range 10–27 MJ m–2 d–1). Wind speed
ranged from 0 to 37 km h–1 with a daily mean of 6.8 km h–1.
Measured and Derived Crop Parameters
The full suite of measured and derived parameters that were
incorporated into the ALMANAC model crops parameter
database to represent the growth and development of the 2-yr
cycle sugarcane are presented in Table 4.
Leaf Area Index. Measured LAI increased from 0.40 at
2 mo after planting to a peak value of 7.5 at 13 mo (Fig. 4).
This LAI value is similar to the 8.0 reported by Muchow et al.
(1997) on a study conducted in Kunia, HI. The sugarcane crop
maintained this LAI for 3 mo, after which it started to decline.
According to Williams and Izaurralde (2013) the LAI of vegetative crops such as sugarcane and some forage crops reaches a
plateau at which time the rates of senescence and growth of leaf
are approximately equal. The LAI started to decline when the
calculated fraction of the growing season was 0.65
The LAI and FIPAR were used to derive k (Fig. 5). The fitted
exponential curve indicated a value of 0.53 and a coefficient
of determination of 0.88. According to Thornley (1976), k
varies with foliage characteristics, sun angle, row spacing, row
direction, and latitude. A somewhat smaller value (0.40–0.60)
might be appropriate for tropical areas in which average sun
angle is higher and for cropping systems with wide row spacing (Begg et al., 1964; Bonhomme et al., 1982; Muchow et al.,
1982). Our estimated value compares well with the value of
0.58 derived by Inman-Bamber (1994) for irrigated sugarcane.
Dry Matter Accumulation. The dry matter (DM) accumulation closely resembled the LAI evolution. As pointed out
earlier, sugarcane cultivars grown in Hawaii over a 2-yr growth
cycle have two distinct growth patterns (Fig. 6). During the
first 5 to15 mo after planting, the cultivars exhibit a rapid and
almost linear growth pattern. Thereafter, DM plateaus at the
maximum measured yield of 80.20 Mg ha–1and then starts
Agronomy Journal
•
Volume 107, Issue 4
•
2015
Table 3. Summary of historical management and cane yield data for seven Hawaii Commercial & Sugar Company (HC&S) fields used to
test ALMANAC’s capability to accurately simulate sugarcane yields. All lands were subsoiled with a deep ripper (40 cm) and harrowed
(8–15 cm) before mechanized planting.
Field
F no. 202
F no. 308
F no. 312
F no. 416
F no. 601
F no. 719
F no. 905
HC&S Weather station
201
501
201
414
501
711
906
Latitude
20.88
20.87
20.88
20.80
20.87
20.86
20.81
Longitude
–156.35
–156.39
–156.35
–156.41
–156.39
–156.46
–156.50
Elevation, m
255
131
255
189
131
30
30
Soil series
Paia (PcC) Keahua (KnB) Haliimaile (HgB) Waiakoa (Wh) Molokai (MuA) Jaucas (JL)
Pulehu (PrA)
Soil texture
silty clay silty clay loam silty clay loam
silty clay loam silty clay loam loamy sand cobbly silty loam
Slope, %
7–15
3–7
3–7
3–7
0–3
0–15
0–3
Years of simulation
1998–2011 1997–2010
1999–2012
1999–2012
1998–2011
1999–2011
1998–2011
Number of 2-yr
6
6
5†
5†
6
6
6
growth cycles (GC)
GC precipitation, mm
Average
1533
925
1667
559
990
831
721
Range
1125–2173
517–1236
1316–2544
312–763
712–1314
283–1412
342–1481
GC irrigation, mm
Average
1897
2633
2015
2895
2579
3191
2734
Range
1433–2305 1941–3272
1750–2423
1892–4177
1952–3277
2290–3849
1139–3665
GC applied N, kg ha–1
Average
375
424
358
348
344
331
351
Range
327–512
384–492
339–366
197–404
303–377
241–375
323–383
GC applied P, kg ha–1
Average
71
93
75
77
9
0
0
Range
0–308
0–205
0–256
0–212
0–55
0
0
Cane yield, Mg ha–1
Average
84
86
88
83
88
101
85
Range
55–115
57–114
64–108
62–115
61–103
86–118
68–107
† Only five 2-yr growth cycles were simulated for fields F no. 312 and F no. 416 because of errors in data records for the sixth cycle.
Table 4. Determined crop parameters for 2-yr cycle sugarcane used to calibrate the ALMANAC model. The biomass-energy ratio
(WA) value includes cane mat and root dry matter. Nitrogen and P nutrient contents are from Williams et al. (1989).
Category
Crop parameter definition
Symbol
Determined value
WA
3.37
Plant growth
Biomass-energy ratio, g MJ–1 m–2
Max. leaf area index (LAI)
DMLA
7.50
Fraction of season when LAI starts to decline
DLAI
0.65
Leaf area decline rate index
RLAD
0.30
Light extinction coefficient for Beer’s Law
k
0.53
First point on optimal LAI curve
DLAP1
15.10
Second point on optimal LAI curve
DLAP2
55.99
Maximum crop height, m
HMX
3.60
Maximum root depth, m
RDMX
2.00
Potential heat units, °C
PHU
9250
Yield components
Plant population (stalk number), m–2
Dry matter decline rate index
Fraction of water in forage yield
Harvest index
Nutrient content
(kg nutrient/kg of biomass)
Nitrogen content at early establishment
Nitrogen content at mid-season
Nitrogen content at maturity
Phosphorus content at early establishment
Phosphorus content at mid-season
Phosphorus content at maturity
Agronomy Journal
•
Volume 107, Issue 4
•
2015
PLANTPO
RBMD
WCY
HI
BN1
BN2
BN3
BP1
BP2
BP3
18.00
1.00
0.75
0.90
0.0100
0.0040
0.0025
0.0075
0.0030
0.0019
1315
Fig. 4. Measured and ALMANAC model simulated leaf area index
(LAI) evolution of 2-yr cycle sugarcane at Hawaii Commercial &
Sugar Company (HC&S) plantation, Maui Island, Hawaii. The crop
growth and simulation period started on June 2011 to May 2013.
Each data point is the average of three replicates. Vertical bars
represent the standard error of the means of the measured data.
Fig. 6. Measured and ALMANAC model simulated 2-yr cycle
sugarcane dry matter accumulation at the Hawaii Commercial &
Sugar Company (HC&S) plantation, Maui Island, Hawaii. The crop
growth and simulation period started from June 2011 to May 2013.
Each data point is the average of three replicates. Vertical bars
represent the standard error of the means of the measured data.
to decline at about 21 mo. At HC&S no fertilizer is applied
in the last 12 mo before harvesting of the commercial crop,
while irrigation is progressively withdrawn during the last 6
mo. The induced water and nutrient stresses decrease vegetative
growth, thus pushing photosynthates toward sucrose storage
(Humbert, 1968; Gascho, 1985). Furthermore, the induced dry
weather conditions also help concentrate the sucrose through
increased evaporation from the leaf surface. The high sucrose
concentration is attractive to both growers and millers because
of reduced costs for harvesting, hauling, and milling.
According to Rostron (1972), as DM increases with crop
age, the proportion of respiring to photosynthesizing tissue increases, causing a gradual decline in net productivity.
Muchow et al. (1997) also attribute the reduced DM to stalk
death. The estimated harvest index (HI) for total harvested
biomass yield and cane moisture content at the last biomass
harvest sample (23 mo) were 0.90 and 75%, respectively.
Commercial harvesting of sugarcane occurs between 20 and 24 mo.
Compared to sugarcane grown in other regions of the world,
the high DM accumulation in 2-yr cycle sugarcane is a result of
the longer growing season, higher LAI, light interception, and
RUE. Sugarcane growth rate is a function of the LAI (Fig. 7).
For the 2-yr growth cycle sugarcane in this study, the relationship is linear. According to Evans (1993), LAI is a key driving
variable for biomass accumulation in most crops, especially
before full crop canopy closure when light interception by
individual leaves is at its maximum. The tillering plant population was estimated at 18 stalks m–2 (Table 4). Sugarcane has
the capacity to tiller rapidly. Under favorable conditions, stalk
numbers increase exponentially with time until a maximum of
20 to 30 stalks m–2 is reached at 4 to 6 mo (Bull and Glasziou,
1975). Over time many younger tillers begin to die due to
shading by older tillers, and tiller number normally stabilizes at
10 to 20 stalks m–2 . In crops such as sugarcane which produce
a high number of yielding tillers compared to the number of
seeds or shoots planted, the modeled plant population is estimated based on the final yield producing tiller number. The
crop grew to a maximum height (HMX) of 3.6 m. Root studies
by Youkhana et al. (2013) showed sugarcane roots reaching to
more than 2.0 m.
Radiation Use Efficiency. The relationship between DM
and accumulated intercepted PAR from early establishment
to final harvest is sigmoid due to DM leveling off at between
15 and 21 mo after planting (Fig. 6 and 8a). This leveling off of
DM is commonly referred to as the reduced growth phenomena (van Heerden et al., 2010) and is widely reported in the
literature (Donaldson et al., 2008; Muchow et al., 1994; Park
et al., 2005; Robertson et al., 1996; Rostron, 1972; Wood et
al., 1996). The leveling off of DM resulted in an overall crop
cycle RUE of 2.06 g MJ–1. This RUE value is close to values
reported for sugarcane elsewhere in the literature; 1.70 g MJ–1
(Robertson et al., 1996), 1.75 and 2.00 g MJ–1 (Muchow et al.,
1994; 1997) for irrigated sugarcane under tropical conditions.
The relationship between DM and accumulated intercepted PAR during the rapid growth phase can however be
viewed as being linear (Fig. 8b). Sugarcane growth is generally
Fig. 5. Relationship between the fraction of intercepted
photosynthetically active radiation (FIPAR) and leaf area index
(LAI). Each data point is the average of three replicates. The light
extinction coefficient (k) derived from the fitted exponential curve
indicates a value of 0.53 and a coefficient of determination of 0.88.
1316
Agronomy Journal
•
Volume 107, Issue 4
•
2015
characterized by a period of rapid biomass accumulation which
corresponds to the near-linear portion of a sigmoidal growth
curve (Coale et al., 1993). According to Muchow et al. (1994),
the linear relationship is more useful as an estimation of RUE
for crop modeling purposes. The RUE during this rapid growth
phase hereby referred to in the model as the Biomass-energy
ratio (WA-Table 4) is approximately 3.37 g MJ–1. This relatively
high RUE value is most certainly due to the inclusion of trash
cane mat. Failure to account for trash biomass results in an
underestimation of aboveground biomass (Muchow et al., 1997).
Several studies showed that trash cane mat can contribute as
much as 10 to 34 Mg ha–1 yr–1 of total sugarcane DM (Evensen
et al., 1997; Trivelin et al., 1995; Vitti et al., 2011; Fortes et al.,
2012). The computation of RUE can vary with the experimental assumptions and methodologies which can result in a wide
range of RUEs for the same crop (Gallo et al., 1993; Sinclair and
Muchow, 1999). Demetriades-Shah et al. (1992) stressed that
comparisons of RUE should only be made when measurements
are made under similar conditions.
Fig. 7. Sugarcane growth rate is a function of the leaf area index
(LAI). Each data point is the average of three replicates. For the 2-yr
growth cycle sugarcane in this study, the relationship is linear.
Fig. 8. (a) Relationship between aboveground (AB) dry matter accumulation and cumulative intercepted photosynthetically active radiation (PAR)
during the 2-yr growth cycle, and (b) The linear relationship between dry matter accumulation and cumulative intercepted PAR during the
period of rapid growth. Each data point is the average of three replicates.
Table 5. Measured and ALMANAC simulated sugarcane yields of seven HC&S plantation fields, Maui, Hawaii. The Mean Error % =
[(Measured – Simulated)/Mean Measured] × 100.
Two-year growth cycle number
Field
Yield
1
2
3
4
5
6
Mean
Mean error
%
no.
Mg ha–1
202
Measured
111.1
114.6
68.4
73.8
55.4
78.8
83.7
Simulated
110.2
101.3
69.1
78.9
53.2
80.6
82.2
1.8
308
Measured
114.5
89.8
92.7
71.4
56.7
90.2
85.9
Simulated
105.6
88.5
92.8
71.3
58.4
91.5
84.7
1.4
312
Measured
89.0
108.0
96.1
63.7
84.8
†
88.3
Simulated
88.3
106.8
94.1
92.8
70.0
†
90.4
–2.4
416
Measured
90.6
114.6
77.4
61.6
76.9
†
84.2
Simulated
91.2
115.3
102.5
61.8
77.1
†
89.6
–6.4
601
Measured
92.4
103.3
79.2
89.6
61.4
102.2
88.0
Simulated
113.1
105.6
79.2
96.2
60.4
100.8
92.6
–5.2
719
Measured
118.2
110.8
95.2
93.4
86.2
103.4
101.2
Simulated
112.7
109.6
115.3
100.1
85.7
103.4
104.5
–3.2
905
Measured
93.4
107.1
91.3
68.0
75.6
73.9
84.9
Simulated
93.1
106.8
91.0
74.6
77.0
75.4
86.3
–1.7
† Only five 2-yr growth cycles were simulated for fields Fno.312 and Fno.416 because of errors in data records for the sixth cycle.
Agronomy Journal
•
Volume 107, Issue 4
•
2015
1317
Supplementary Parameters. The fraction of the growing
season when the LAI starts to decline (DLAI) was derived
from data in Fig. 4, while the LAI decline rate index (RLAD)
defines the shape of the LAI curve after DLAI (Fig. 4). We
used the data presented in Fig. 4 in conjunction with the
ALMANAC LAI simulation data to estimate DLAP1 and
DLAP2, which are two points on the optimal (non-stress) leaf
area development curve. In both DLAP1 and DLAP2 (Table
4), numbers before the decimal are percent of growing season,
while numbers after the decimal are fractions of maximum
potential LAI (DMLA). The DM decline rate index (RBMD)
functions like the RLAD for LAI and reduces the efficiency of
conversion of intercepted PAR to biomass toward the end of
the growing season. As with RLAD the RBMD decline rate
index was determined by adjusting pre-set values (range 1–10)
until the simulated decline rate approximates that of measured
DM data values. The model estimated PHUs for the 2-yr cycle
sugarcane at approximately 9250°C.
Even though root studies by Youkhana et al. (2013) showed
sugarcane roots exceeding 2 m, we applied the model maximum
default rooting depth of 2.00 m. This depth is adequate for
simulating water and nutrient uptake with the ALMANAC
model in most cases. The N and P nutrient contents decline
from early seedling emergence (BN1; BP1), midseason (BN2;
BP2) and at maturity (BN3; BP3) (Table 4) (Williams et al.,
1989). These crop parameters represent the optimal N and P
concentrations at key growth stages and decline with increasing
growth (Jones, 1983).
Except for the plateauing of DM accumulation in the second
year of growth, the 2-yr cycle sugarcane’s growth pattern is
similar to that of 12 to 18 mo sugarcane systems on the U.S.
mainland and other regions of the world. Modelers applying
the crop parameters determined in this study in these regions
should however be aware of the need to pay attention to specific
parameters whose values are dependent on local conditions,
such as climate, soil type, and rooting depth, crop management
practice etc. For example, the growth cycle PHUs will have to
be adjusted for local climatic conditions, while as pointed out
earlier, k varies with foliage characteristics, sun angle, row spacing, row direction, and latitude.
Overall, our results on DM accumulation and RUE confirm
the recent findings of Anderson et al. (2015) which showed
that 2-yr cycle sugarcane has higher overall biomass and net
productivity compared to sugarcane systems in other regions of
the world.
Table 6. Paired t test statistics to assess differences between
measured and simulated sugarcane yields.
Data variable
Measured yield
Simulated yield
Mean
88.12
90.03
Variance
303.31
289.44
Observations
40
40
Pearson Correlation
0.88
Degrees of freedom
39
t Statistic
1.41
P (T £ t) two-tail
t Critical two-tail
1318
0.17
2.02
ALMANAC Model Evaluation
The calibrated model satisfactorily simulated historical sugarcane yields harvested from seven HC&S fields; F no. 202, 308,
312, 416, 601, 719, and 905, over five to six 2-yr growth cycles
(Tables 5 and 6 and Fig. 9). The mean simulation percent (%)
errors ranged from –6.4 to 1.8%. The small negative mean simulation percent errors indicate that the model slightly overpredicted sugarcane yields (fields 312, 416, 601, 719, and 905), while
the positive errors indicated that sugarcane yields were slightly
underpredicted (fields 202 and 308). The slope and intercept
of the regression line shown in Fig. 9 were not significantly different from 1 and zero, respectively. In addition, the calculated
Paired t test of 1.41 with 39 degrees of freedom, and the Pearson
correlation coefficient of 0.88 (Table 6) show that there were no
significant differences (P ≥ 0.05) between measured and simulated yields. The coefficients of variation of the measured and
simulated yields were 3.05 and 2.99%, respectively.
Although the model managed to capture various sugarcane
production stresses on water, nutrient (N and P), and temperature across the fields and over years, the model simulated N
and irrigation water inputs were in general adequate to meet
historical yields, though additional P fertilizer was required
across fields and growth cycles (Table 3). Field-applied irrigation water and N averaged 2563 mm (1433–4177 mm), and
362 kg N ha–1 (197–512 kg N ha–1), respectively. ALMANAC
additional auto P fertilizer applications averaged 80 kg P ha–1
(0–168 kg P ha–1). The higher P additions by the model could
be attributable to either low P levels in the NRCS SSURGO
soils database used in the model or the ALMANAC model’s
incapability to accurately model P dynamics of various soils.
There was no sufficient soil P analysis data for the modeled
fields that could be used to update the soil available P for each
soil type (field) as impacted by various management practices
over time. As the soil survey for NRSC SSURGO soils characterization was conducted several years before the simulated
years, this discrepancy in soil P and modeled P inputs could
somewhat be expected. While P fixation is predominant in
Fig. 9. Measured and ALMANAC model simulated historical
sugarcane yields for fields 202, 308, 312, 416, 601, 719, and 905
over five to six growth cycles (see Table 3 for the actual years
that were simulated for each field). The dashed line is the 1:1
line through the origin.
Agronomy Journal
•
Volume 107, Issue 4
•
2015
oxidic tropical soils and P deficiency is a major limitation to
crop production (Grierson et al., 2004), measuring plantavailable P in these soils is challenging (Gama-Rodrigues et
al., 2014). According to FAO (2013), sugarcane requires 20 to
90 kg P ha–1 with the actual amounts varying depending on
several factors such as soil type, management history etc.
CONCLUSIONS
When compared to sugarcane grown in other regions of the
world, the results of this study indicate that the high biomass
yield of 2-yr cycle sugarcane is due to multiple factors that
include the long growing season, high LAI, light interception, and RUE. When the crop parameters were incorporated
into the ALMANAC biophysical model, the model was able
to accurately simulate sugarcane yields across five to six 2-yr
growth cycles, different soil types and multiple management
scenarios of applied irrigation water, N fertilizer inputs, and
various planting and harvest dates. Not only did the model
give a good representation of final biomass yields, but also a
good simulation of crop growth and development throughout
the growth cycle. It is apparent that when fully calibrated and
tested, the model can be used by researchers to evaluate growers’ management alternatives that can potentially maximize
biomass yields while optimizing water, N and P nutrient
inputs. Furthermore, the crop parameters developed in this
study can also be used in watershed scale models such as the
Agricultural Policy/Environmental eXtender (APEX) and Soil
and Water Assessment Tool (SWAT) (both of which share the
same crop growth model with ALMANAC) for simulating
crop growth at a much larger scale.
ACKNOWLEDGMENTS
Preparation of this manuscript is supported by the USDA-ARS,
Grassland, Soil and Water Research Laboratory, and Texas A&M
AgriLife Research, Temple, TX, through Specific Cooperative
Agreement: 58-6206-1-053, and was funded by the U.S. Navy, Office
of Naval Research (ONR). This work was also partly supported by
ONR Grant N00014-12-1-0496 to the University of Hawaii. We
thank Neil Abranyi, Jason Drogowski, Daniel Richardson, Meghan
Pawlowski, and HC&S staff for field work. USDA is an equal opportunity provider and employer.
REFERENCES
Anderson, G.R., R. Tirado-Corbala, D. Wang, and J.E. Ayars. 2015. Longrotation sugarcane in Hawaii sustains higher carbon accumulation and
radiation use efficiency in 2nd year of growth. Agric. Ecosyst. Environ.
199:216–224. doi:10.1016/j.agee.2014.09.012
Begg, J.E., J.F. Bierhuizen, E.R. Lemon, D.K. Misra, R.O. Slatyer, and W.R.
Stern. 1964. Diurnal energy and water exchanges in bulrush millet. Agric.
Meteorol. 1:294–312. doi:10.1016/0002-1571(64)90037-8
Behrman, K.D., J.R. Kiniry, M. Winchell, T.E. Juenger, and T.H. Keitt. 2013.
Spatial forecasting of switchgrass yield under current and future climate
change scenarios. Ecol. Appl. 23:73–85. doi:10.1890/12-0436.1
Bonhomme, R., F. Ruget, M. Derieux, and P. Vincourt. 1982. Relationship
between aerial dry matter production and intercepted solar radiation for
various maize genotypes. Comptes Rendus Hebdomadaires Séances
Acad. Sci. III 294:393–398.
Bull, T.A., and K.T. Glasziou. 1975. Sugarcane. In: L.T. Evans, editor, Crop
physiology. Some case histories. Cambridge Univ. Press, Cambridge, UK.
p. 51–72.
Agronomy Journal
•
Volume 107, Issue 4
•
2015
Bull, T.A., and D.A. Tovey. 1974. Aspects of modeling sugarcane growth by
computer simulation. Proc. Int. Soc. Sugar Cane Technol. 15:1021–1032.
Coale, F.J., C.S. Sanchez, F.T. Izuno, and A.B. Bottcher. 1993. Nutrient accumulation and removal by sugarcane grown on Everglades histosols. Agron. J.
85:310–315. doi:10.2134/agronj1993.00021962008500020028x
Demetriades-Shah, T.H., M. Fuchs, E.T. Kanemasu, and I. Flitcroft. 1992. A
note of caution concerning the relationship between cumulated intercepted radiation and crop growth. Agric. For. Meteorol. 58:193–207.
doi:10.1016/0168-1923(92)90061-8
Donaldson, R.A., K.A. Redshaw, R. Rhodes, and R. van Antwerpen. 2008.
Season effects on productivity of some commercial South African sugarcane cultivars, I: Biomass and radiation use efficiency. Proc. SASTA.
81:517–527.
Engel, B., I. Chaubey, M. Thomas, D. Saraswat, P. Murphy, and B. Bhaduri.
2010. Biofuels and water quality: Challenges and opportunities for simulation modeling. Biofuels 1:463–477. doi:10.4155/bfs.10.17
Evans, L.T. 1993. Crop evolution, adaptation and yield. Cambridge Univ. Press,
Cambridge, UK.
Evensen, C.I., R.C. Muchow, S.A. El-Swaify, and R.V. Osgood. 1997. Yield accumulation in irrigated sugarcane: I. Effect of crop age and cultivar. Agron.
J. 89:638–646.
FAO (Food and Agricultural Organization). United Nations. 2013. Water. Natural Resources Management and Environment Department. http://www.
fao.org/nr/water/index.html (accessed March 2014) .
Fortes, C., P.C.O. Trivelin, and A.C. Vitti. 2012. Long-term decomposition of
sugarcane harvest residues in São Paulo state, Brazil. Biomass Bioenergy
42:189–198. doi:10.1016/j.biombioe.2012.03.011
Gallo, K.P., C.S.T. Daughtry, and C.L. Wiegand. 1993. Errors in measuring
absorbed radiation and computing crop radiation use efficiency. Agron. J.
85:1222–1228. doi:10.2134/agronj1993.00021962008500060024x
Gama-Rodrigues, A.C., M.V.S. Sales, P.S.D. Silva, N.B. Comerford, W.P. Cropper, and E.F. Gama-Rodrigues. 2014. An exploratory analysis of phosphorous transformations in tropical soils using structural equation modeling.
Biogeochemistry 118:453–469. doi:10.1007/s10533-013-9946-x
Gascho, G.J. 1985. Water-sugarcane relationships. Sugar J. 48:11–17.
Goldemberg, J. 2008. The Brazilian biofuels industry. Biotechnol. Biofuels 1:6
(1–7). doi:10.1186/1754-6834-1-6.
Grantz, D.A., and H.B. Vu. 2009. Sensitivity in a potential C4 bioenergy
crop: Sugarcane in California. Crop Sci. 49:643–650. doi:10.2135/
cropsci2008.04.0210
Grierson, P.F., P. Smithson, G. Nziguheba, G.S. Radersma, and B. Comerford.
2004. Phosphorus dynamics and mobilization by plants. In: M. van
Noordwisk, G. Cadisch, and C.K. Ong, editors, Below-ground interactions in tropical agroecosystems: Concepts and models with multiple plant
components. CABI Int., Wallingford, UK. p. 127–142.
HC&S. 2011. Hawaiian Commercial & Sugar Co., Maui. http://www.hcsugar.
com/hcs_qa.shtml#about (accessed March 2014).
Heinz, D.J., T.L. Tew, H.K. Meyer, and K.K. Wu. 1981. Registration of H657052 Sugarcane reg. no. 51. Crop Sci. 21:634–635. doi:10.2135/cropsci19
81.0011183X002100040050x
Humbert, R.P. 1968. The growing of sugarcane. Elsevier Publishing Co. Ltd,
Amsterdam.
Inman-Bamber, N.G. 1994. Temperature and seasonal effects on canopy development and light interception of sugarcane. Field Crops Res. 36:41–51.
doi:10.1016/0378-4290(94)90051-5
Izaurralde, R.C., J.R. Williams, W.B. McGill, N.J. Rosenberg, and M.C. Jakas.
2006. Simulating soil C dynamics with EPIC: Model description and
testing against long-term data. Ecol. Modell. 192:362–384. doi:10.1016/j.
ecolmodel.2005.07.010
Jones, C.A. 1983. A survey of the variability in tissue nitrogen and phosphorus
concentrations in maize and grain sorghum. Field Crops Res. 6:133–147.
doi:10.1016/0378-4290(83)90053-9
Jones, C.A. 1985. C-4 Grasses and Cereals. John Wiley & Sons, New York.
Kemanian, A.R., C.O. Stöckle, and D.R. Huggins. 2004. Variability of barley radiation-use efficiency. Crop Sci. 44:1662–1672. doi:10.2135/
cropsci2004.1662
1319
King, J.E., J.M. Hannifan, and R. Nelson. 1998. An assessment of the feasibility
of electric power derived from biomass and waste feedstocks. Rep. KRD9513. Kansas Electric Utilities Research Program and Kansas Corp. Commission, Topeka. p. 1–266.
Kiniry, J.R., K.A. Cassida, M.A. Hussey, J.P. Muird, W.R. Ocumpaugh, J.C.
Read et al. 2005. Switchgrass simulation by the ALMANAC model
at diverse sites in the southern US. Biomass Bioenergy 29:419–425.
doi:10.1016/j.biombioe.2005.06.003
Kiniry, J.R., M.V. Johnson, S.B. Bruckerhoff, J.U. Kaiser, R.L. Cordsiemon,
and R.D. Harmel. 2012. Clash of the Titans: Comparing productivity via
radiation use efficiency for two grass giants of the biofuel field. BioEnergy
Res. 5:41–48. doi:10.1007/s12155-011-9116-8
Kiniry, J.R., M.A. Sanderson, J.R. Williams, C.R. Tischler, M.A. Hussey, W.R.
Ocumpaugh et al. 1996. Simulating Alamo switchgrass with the ALMANAC model. Agron. J. 88:602–606. doi:10.2134/agronj1996.00021962
008800040018x
Kiniry, J.R., M.R. Schmer, K.P. Vogel, and R.B. Mitchell. 2008. Switchgrass
biomass simulation at diverse sites in the northern Great Plains of the US.
BioEnergy Res. 1:259–264. doi:10.1007/s12155-008-9024-8
Kiniry, J.R., J.R. Williams, P.W. Gassman, and P. Debaeke. 1992. A general
process oriented model for two competing plant species. Trans. ASABE
35:801–810. doi:10.13031/2013.28665
Kinoshita, C., and J. Zhou. 1999. Siting evaluation for biomass-ethanol production in Hawaii. Natl. Renewable Energy Lab., Oak Ridge, TN.
Laclau, P.B., and J.P. Laclau. 2009. Growth of the whole root system for a plant
crop of sugarcane under rainfed and irrigated environments in Brazil. Field
Crops Res. 114:351–360. doi:10.1016/j.fcr.2009.09.004
Meek, D.W., J.L. Hatfield, T.A. Howell, S.B. Idso, and R.J. Reginato. 1984. A
generalized relationship between photosynthetically active radiation and
solar radiation. Agron. J. 76:939–945. doi:10.2134/agronj1984.0002196
2007600060018x
Meki, N.M., J.L. Snider, J.R. Kiniry, R.L. Raper, and A.C. Rocateli. 2013. Energy
sorghum biomass harvest thresholds and tillage effects on soil organic
carbon and bulk density. Ind. Crops Prod. 43:172–182. doi:10.1016/j.
indcrop.2012.07.033
Monsi, M., and T. Saeki. 1953. Uber den Lichtfaktor in den Pflanzengesellschaften und sein Bedeutung für die Stoffproduktion. J. Jpn. Bot.
14:22–52.
Monteith, J.L. 1965. Radiation and crops. Exp. Agric. 1:241–251. doi:10.1017/
S0014479700021529
Muchow, R.C., D.B. Coates, G.L. Wilson, and M.A. Foale. 1982. Growth
and productivity of irrigated Sorghum bicolor L. Moench. In: Northern
Australia. I. Plant density and arrangement effects on light interception
and distribution, and grain yield, in the 128 hybrid Texas 610SR in low
and medium latitudes. Aust. J. Agric. Res. 33:773–784. doi:10.1071/
AR9820773
Muchow, R.C., C.I. Evensen, R.V. Osgood, and M.J. Robertson. 1997. Yield
accumulation in irrigated sugarcane: II. Utilization of intercepted radiation. Agron. J. 89:646–652. doi:10.2134/agronj1997.00021962008900
040017x
Muchow, R.C., M.F. Spillman, A.W. Wood, and M.R. Thomas. 1994. Radiation
interception and biomass accumulation in a sugarcane crop grown under
irrigated tropical conditions. Aust. Agric. Res. 45:37–49. doi:10.1071/
AR9940037
NASS. 2013. National statistics for sugarcane. USDA. http://www.nass.usda.
gov (accessed June 2014).
Osorio, J., J. Jeong, K. Bieger, and J. Arnold. 2014. Influence of potential evapotranspiration on the water balance of sugarcane fields in Maui, Hawaii. J.
Water Resour. Protection 6:852–868. doi:10.4236/jwarp.2014.69080
Park, S.E., M. Robertson, and N.G. Inman-Bamber. 2005. Decline in the
growth of a sugarcane crop with age under high input conditions. Field
Crops Res. 92:305–320. doi:10.1016/j.fcr.2005.01.025
1320
Parton, W.J. 1996. The CENTURY model. In: D.S. Powlson, P. Smith, and
J.U. Smith, editors, Evaluation of soil organic matter models using existing
long-term datasets. Springer-Verlag, Berlin, Germany.
Persson, T., B. Ortiz, B.I. Bransby, W. Wu, and G.T. Hoogenboom. 2011. Determining the impact of climate and soil variability on switchgrass (Panicum
virgatum L.) production in the south‐eastern USA; a simulation study.
Biofuels Bioprod. Biorefin. 5:505–518. doi:10.1002/bbb.288
Robertson, M.J., A.W. Wood, and R.C. Muchow. 1996. Growth of sugarcane under high input conditions in tropical Australia. I. Radiation use,
biomass accumulation and partitioning. Field Crops Res. 48:11–25.
doi:10.1016/0378-4290(96)00041-X
Rostron, H. 1972. The effects of age and time of harvest on the productivity of
irrigated sugarcane. Proc. SASTA 46:142–150.
SAS Institute. 2007. SAS system version 9.3. SAS Inst. Cary, NC.
Sinclair, T.R., and R.C. Muchow. 1999. Radiation use efficiency. Adv. Agron.
65:215–265. doi:10.1016/S0065-2113(08)60914-1
Snedecor, G.W., and W.G. Cochran. 1989. Statistical methods. 8th ed. Iowa
State Univ. Press, Ames.
Stöckle, C.O., and J.R. Kiniry. 1990. Variability in crop radiation use efficiency
associated with vapor-pressure deficit. Field Crops Res. 25:171–181.
doi:10.1016/0378-4290(90)90001-R
Thornley, J.H.M. 1976. Photosynthesis. In: J.H.M. Thornley, editor, Mathematical models in plant physiology. Academic Press, London, UK. p. 92–110.
Trivelin, P.C.O., R.L. Victoria, and J.C.S. Rodrigues. 1995. Utilization by
late harvest sugar cane ratoon of nitrogen from 15N-aqua ammonia and
15N-urea applied to the soil as vinasse N-complement. Pesqi. Agropecu.
Bras. 30:1375–1385.
USDA NRCS. Soil Survey Staff. 2011. Spatial and tabular data of the Soil Survey for Maui, Hawaii. USDA. http://soildatamartnrcs.usda.gov/ (accessed
July 2014).
van Heerden, P.D.R., R.A. Donaldson, D.A. Watt, and A. Singels. 2010. Biomass accumulation in sugarcane: Unravelling the factors underpinning
reduced growth phenomena. J. Exp. Bot. 61:2877–2887. doi:10.1093/jxb/
erq144
Vitti, A.C., H.C.J. Franco, P.C.O. Trivelin, D.A. Ferreira, R. Otto, C. Fortes,
and C.E. Faroni. 2011. Nitrogen derived from fertilization and straw for
plant cane nutrition. Pesqi. Agropecu. Bras. 46:287–293. doi:10.1590/
S0100-204X2011000300009
Wang, M., J. Han, H. Dunn, H. Cai, and A. Elgowainy. 2012. Well-to-wheels
energy use and greenhouse gas emissions of ethanol from corn, sugarcane and cellulosic biomass for US use. Environ. Res. Lett. 7:045905.
doi:10.1088/1748-9326/7/4/045905
Williams, J.R. 1995. The EPIC model. In: V.P. Singh, editor, Computer models
of watershed hydrology. Water Resources Publ., Highlands Ranch, CO.
p. 909–1000.
Williams, J.R., and C. Izaurralde. 2013. APEX Theoretical Documentation.
Texas A&M AgriLife Res., Blackland Res. & Ext. Ctr., Temple. http://
epicapex.tamu.edu/files/2013/02/the-apex-theoretical-documentation.
pdf (accessed March 2014).
Williams, J.R., C.A. Jones, J.R. Kiniry, and D.A. Spanel. 1989. The EPIC crop
growth model. Trans. ASAE 32:497–511. doi:10.13031/2013.31032
Wood, A.W., R.C. Muchow, and M.J. Robertson. 1996. Growth of sugarcane
under high input conditions in tropical Australia. III. Accumulation, partitioning and use of nitrogen. Field Crops Res. 48:223–233. doi:10.1016/
S0378-4290(96)00043-3
Youkhana, A., S. Crow, M. Meki, J. Kiniry, R. Ogoshi, and M. Nakahata. 2013.
Belowground biomass and C dynamics in sugarcane and ratooning energycane cultivated for biofuel production in Hawaii. Abstract. ASA-CSSASSSA Annual International Tri-Society Meeting, Tampa, FL. 3-6 Nov.
2013. ASA, CSSA, and SSSA, Madison, WI.
Agronomy Journal
•
Volume 107, Issue 4
•
2015