Conference PaperPDF Available

OPTIMUM SUGARCANE GROWTH STAGE FOR CANOPY REFLECTANCE SENSOR TO PREDICT BIOMASS AND NITROGEN UPTAKE

Authors:

Abstract and Figures

The recent technology of plant canopy reflectance sensors can provide the status of biomass and nitrogen nutrition of sugarcane spatially and in real time, but it is necessary to know the right moment to use this technology aiming the best predictions of the crop parameters by the sensor. A study involving eight commercial fields located in the state of São Paulo, Brazil, varying from 16 to 21 ha, planted with four varieties, was conducted during two growing seasons (2009/10 -2010/11). Conditions varied from sandy to heavy soils and the previous harvesting occurred in May and October (early and late season), including first to fourth ratoon stages. Fields were scanned with the reflectance canopy sensor (N-Sensor TM ALS, Yara International ASA) three times in the first season (approximately at 0.2, 0.4, and 0.6 m of stem height) and two on the second season (0.3 and 0.5 m), followed by tissue sampling for biomass, crop height and nitrogen uptake on ten spots inside the area, guided by the different values shown by the canopy sensor. At 0.2 m of field average stem height, sugarcane biomass is low for a good sensor prediction of the parameters; at 0.6 m height starts the saturation, where the ability of the sensor to predict biomass and nitrogen begins to be affected. Between 0.3 and 0.5 m of stem height results show the best correlation between real and sensor predicted biomass and nitrogen uptake for sugarcane crop, indicating that this is the right period for using the sensor to guide variable rate nitrogen application.
Content may be subject to copyright.
OPTIMUM SUGARCANE GROWTH STAGE FOR CANOPY
REFLECTANCE SENSOR TO PREDICT BIOMASS
AND NITROGEN UPTAKE
G. Portz, L. R. Amaral and J. P. Molin
Biosystems Engineering Department, "Luiz de Queiroz" College of Agriculture
University of São Paulo
Piracicaba, São Paulo, Brazil
J. Jasper
Research Centre Hanninghof
Yara International ASA
Duelmen, Germany
ABSTRACT
The recent technology of plant canopy reflectance sensors can provide the status
of biomass and nitrogen nutrition of sugarcane spatially and in real time, but it is
necessary to know the right moment to use this technology aiming the best
predictions of the crop parameters by the sensor. A study involving eight
commercial fields located in the state of São Paulo, Brazil, varying from 16 to 21
ha, planted with four varieties, was conducted during two growing seasons
(2009/10 - 2010/11). Conditions varied from sandy to heavy soils and the
previous harvesting occurred in May and October (early and late season),
including first to fourth ratoon stages. Fields were scanned with the reflectance
canopy sensor (N-SensorTM ALS, Yara International ASA) three times in the first
season (approximately at 0.2, 0.4, and 0.6 m of stem height) and two on the
second season (0.3 and 0.5 m), followed by tissue sampling for biomass, crop
height and nitrogen uptake on ten spots inside the area, guided by the different
values shown by the canopy sensor. At 0.2 m of field average stem height,
sugarcane biomass is low for a good sensor prediction of the parameters; at 0.6 m
height starts the saturation, where the ability of the sensor to predict biomass and
nitrogen begins to be affected. Between 0.3 and 0.5 m of stem height results show
the best correlation between real and sensor predicted biomass and nitrogen
uptake for sugarcane crop, indicating that this is the right period for using the
sensor to guide variable rate nitrogen application.
Key words: nitrogen management, proximal sensing, N-Sensor.
INTRODUCTION
Sugarcane (Saccharum ssp.) is the main crop that supplies sugar, and the
second for ethanol production, growing in tropical and subtropical areas, which
provides around 80% of the sugar world production and 35% of the ethanol.
Brazil is the worldwide main sugarcane producer (FAO 2012).
Sugarcane producers, despite research on the nitrogen nutrition
contributions, continue with the challenge of making better use of the input,
especially due to the spatial variability of the nutrient and soils found in
production areas, often in short distances (Solie et. al., 1999).
The recent technology of canopy sensors using vegetation reflectance at
certain wavelengths can provide georeferenced information about biomass and
nitrogen nutrition of the crop in real time, which can guide the implementation of
variable N application. The sugar-ethanol industry in São Paulo state, which
produces 60% of the commercial sugarcane of Brazil, indicated that precision
agriculture technologies can provide improvements, higher yield, lower costs,
minimize the environmental impacts and bring improvements in sugar cane
quality, suggest a research made by Silva et. al., (2010), that also says that 96% of
the sector wants to expand the use of precision agriculture practices.
One of the existing canopy sensors for nitrogen management is the N-
Sensor (N-SensorTM ALS, Yara International ASA). According to Jasper (2006)
apud Reuch (2005), it uses an optimum waveband selection to generate a
vegetation index (VI) to determining the nitrogen uptake from crops by active
remote sensing, being superior to “classical” reflectance ratios, with one
waveband in the visible and one waveband in the near infrared region of the
electromagnetic spectrum. In particular, the resulting relationship seemed to be
largely independent of growth stage and variety, and showed less saturation at
high N uptake levels.
According to Singh (2006), there is a great scope for use the N-Sensor for
optimize nitrogen application in sugarcane cultivation, but the sensor needs to be
tested and validated for sugarcane cropping systems. This research activity is
already being done in Brazil since 2009 (Portz et. al., 2012), showing that the
sensor is capable to predict biomass and nitrogen uptake with accuracy
independent of soil, variety and year season during a long period of the initial
development of the crop, also showing the first data set capable to provide an
algorithm to guide variable rate application of N over commercial sugarcane
fields.
This optimized VI used by the sensor uses the beginning of the near
infrared (NIR) at 760 nn and the slope of reflectance between the red and the NIR
named REIP (Red Edge Inflection Point), at 730 nn.
Also Mutanga and Skidmore (2004), Heege et, al., (2008) and Mokhele
and Ahmed (2010) showed that the red edge area contains more information on
biomass quantity as compared to other parts of the electromagnetic spectrum and
that narrow wavelengths located in the red edge slope contain information at full
canopy cover, having the highest correlation coefficients with biomass IV
obtained with waveband located in the shorter red edge portion (706 nm) and a
band located in the longer red edge portion (755 nm) for a better estimation of
biomass at high canopy density. However Portz et al. (2012) indicate that in
sugarcane, a high biomass crop, at 0.6 m average of stem height, saturation starts
to appear on the sensor signal using the IV from red edge.
This paper shows an improvement and validation of the results presented
by Portz et al. (2012) by proposing the right moment to use the N-Sensor aiming
to indicate biomass accumulation and nitrogen application demands based on the
N-uptake on commercial sugarcane fields.
MATERIALS AND METHODS
During the 2009/10 and 2010/11 growing seasons eight commercial fields
of sugarcane located around the São Martinho Sugar Mill (21°19’11”S,
48°07’23”W), in the state of São Paulo, Brazil, were evaluated. Conditions varied
from sandy to clay soils, with all crops being mechanically green harvested (with
no burn). On four fields, harvesting of the previous crop occurred at the beginning
of the season (May/Jun) corresponding to the dry time of the year, and on the
other four fields, in late season (Oct/ Nov), corresponding to the wet time of the
year. The crops under investigation included first, second, and third ratoon stages
at 2009/10 and second, third and fourth at 2010/11. The first four fields were
planted with the varieties CTC 9 over sandy soil and RB 855453 over clay soil
and all were harvested in the dry season. The last four fields were planted with the
varieties CTC 2 on sandy soil and SP 80–3280 on clay soil, and harvested during
the wet season (Table 1).
Table 1: Variables of the studied areas
Shortly after harvesting all fields were fertilized with a uniform dose of
100 kg ha-1 of nitrogen using ammonium nitrate (30 % N) as the N source, spread
over the sugarcane rows surface.
The sugarcane fields were scanned using the N-Sensor TM ALS (Yara
International ASA, Duelmen, Germany) (Jasper et al., 2009). The sensor was
mounted behind the cabin of a high clearance vehicle.
The target parameter for the agronomic calibration of the sensor readings
is the N-uptake of the above-ground biomass of the crop (Link, 2005). As the
relationship between sensor readings and crop N uptake might be growth stage
specific, each of the eight fields was scanned with the sensor three times in the
2009/10 growing season (at 0.2, 0.4, and 0.6 m average stem height) and two
times during the 2010/11 (at 0.3, and 0.5 m average stem height) (Fig. 1).
Figure 1: Grow status of sugarcane at the measurement moments
The sensor was connected to a GPS receiver and the vehicle was driven
through the whole field spaced by 10 rows of 1.5 m. After the scanning, the
sensor data was processed generating sensor VI index maps of the fields, over this
maps 10 sample plots were located guided by the different values shown by the
canopy sensor and followed by tissue sampling for biomass, crop height and
nitrogen uptake as explain Portz et al. (2012).
Sensor readings of the respective sample plots were related to the crop
parameters, specific calibration functions were derived, and the capacity of the
sensor measurements to predict the actual crop biomass and N-uptake was
investigated.
An exploratory analysis of the data was done running box plot test using
Sigma plot 10. Sensor data of each field were correlated with biomass and
nitrogen uptake from the respective sample points. Also simple linear regression
models were used to compare sugarcane N-uptake collected data against sensor
predicted N-uptake for each of the field stem height average evaluated.
RESULTS AND DISCUSSION
In order to observe the individual behavior of the variables in each of the
eight fields evaluated in five crop heights during two years, the data from the
studied fields were compared first independently side by side by box plot analyses
for biomass (Fig. 2 and 3) and for N-uptake (Fig. 4 and 5)
Figure 2: Sugarcane real measured biomass compared to sensor predicted
biomass for the four fields of the early season (dry season).
Observations: Sat = saturation point, NA = Not available data.
Figure 3: Sugarcane real measured biomass compared to sensor predicted
biomass for the four fields of the late season (wet season).
Observations: Sat = saturation point, NA = Not available data.
Analyzing the biomass data it is possible to see that the 2010/11 data (0.3
and 0.5 m) fitted right in the 2009/10 (0.2, 0.4 and 0.6 m), even with climate
differences between years (data not show) and one ratoon older fields.
The first measurement (0.2 m) shows low and concentrated values, usually
under 1000 kg ha-1 of dry matter, especially on the early season that is on the dry
and colder period of the year.
The real biomass measured in field (left graph) for the early season (Fig.
2) and for the late season (Fig. 3) have reach around 8000 kg ha-1 of dry matter,
especially on the late season that is on the rainy and warmer season of the year.
However when we analyze the sensor predicted values for the same field points
(right graph), around 6000 kg ha-1 of dry matter a superior limit is achieved,
indicating that the phenomenon of sensor saturation begins (red line).
The sensor saturation happens when the biomass increases but the values
of the sensor for the same biomass increase in a lower rate or stop to increase. The
saturation is related to the IV index used by the sensor and also to the narrow
bands involved on it. The major limitation of using vegetation indices based on
the red and NIR portion of the electromagnetic spectrum is that they
asymptotically approach a saturation level after a certain biomass density (Tucker
1977, Todd et al. 1998, Thenkabail et al. 2000). The results are indicating that the
red edge sensor used is accurate working until biomass covers the entire surface,
as happens when the sugarcane is with 0.6 m of stem height, (Fig. 1).
For N-uptake the behavior is similar, as shown on Figures 4 and 5.
Figure 4: Sugarcane real measured N-uptake compared to sensor predicted N-
uptake for the four fields of the early season (dry season).
Observations: Sat = saturation point, NA = Not available data.
Figure 5: Sugarcane real measured N-uptake compared to sensor predicted N-
uptake for the four fields of the late season (wet season).
Observations: Sat = saturation point, NA = Not available data.
The N-uptake values follow the biomass trend as they are the dry matter
multiplied by the N concentration of the field samples. But the sensor saturation is
not so clear on the N-uptake, part explained because when the plant grows the
biomass increases but the N concentration decreases (not showed data). In this
way sensor saturation does not show so early when the intention is to predict
nitrogen and not biomass, what is good because the main target is to predict N use
by the crop.
Even not so clear like in biomass, it is possible to see that the sensor can
predict N-uptake until around 70 kg N ha-1 whit high accuracy, but having
deviations in some fields and heights. There is an explanation for the second field
(Fig. 4) not showing saturation at the 0.6 m height, like the others; the sensor
scanning was made in a very warm day in the afternoon (2:00 pm) during a very
drought period and the crop was presenting closed lives to preserve water. This
plant reaction led to a decrease in canopy cover reducing the values read by the
sensor without biomass decrease.
On the wet season (Fig. 5), fields 5 and 7 had predicted N-uptake values
over the proposed sensor saturation line of 70 kg of N ha-1 on the 0.5 m
measurement, but at the 0.6 m stem average height both field presented lower
values indicating that the saturation phenomenon appears. At the 0,5 m
measurement of the field 7 it is also possible to see that the real N-uptake was
superior, this is because at this year (2010/11) there was a delay on the application
of the 100 kg N ha-1 of ammonium nitrate that was applied few days before the
evaluation on this field.
At 0.5 m stem height the sugarcane canopy is almost close (Fig. 1), don’t
increasing much the number of leaves anymore on higher heights, but just the
elongation of the stem that increases biomass. This is a crop characteristic that
influences negatively the sensor, and not just a sensor limitation to saturation.
Aiming to solve all doubts about a the N-uptake during the growth stages
and the performance of the sensor, a second analyses was done comparing
directly real N-uptake and sensor predicted N-uptake by correlation curves as
show in Figure 6.
Figure 6: Sugarcane real measured N-uptake compared to sensor predicted N-
uptake for the five measurement heights.
y = 0.856x + 1.5535
R2 = 0.68
0
20
40
60
80
100
120
0 20 40 60 80 100 120
Real N-uptake (Kg ha-1)
Predicted N-uptake (kg ha-1)
0.2 m
Linear (0.2 m)
Linear ("1/1")
y = 1.024 6x - 0.653
R2 = 0.82
0
20
40
60
80
100
120
0 20 40 60 80 100 120
Real N-uptake (Kg ha-1)
Predicted N-uptake (kg ha-1)
0.3 m
Linear (0.3 m)
Linear ("1/1" )
y = 0.92 59x + 5.73 77
R2 = 0.67
0
20
40
60
80
100
120
0 20 40 60 80 100 120
Real N-uptake (Kg ha-1)
Predicted N-uptake (kg ha-1)
0.4 m
Line ar (0.4 m)
Line ar ("1/1")
y = 0.766 4x + 11.21 3
R2 = 0.81
0
20
40
60
80
100
120
0 20 40 60 80 100 120
Real N-uptake (Kg ha-1)
Predicted N-uptake (kg ha-1)
0.5 m
Line ar (0.5 m)
Line ar ("1/1")
y = 0.55 55x + 21.7 47
R2 = 0.66
0
20
40
60
80
100
120
0 20 40 60 80 100 120
Real N-uptake (Kg ha-1)
Predicted N-uptake (kg ha-1)
0.6 m
Linear (0.6 m)
Linear ("1/1")
As noted by Portz, et al. (2012), pooling the data of different fields on a
single set, no differences coming from soil, variety and season is possible to see,
having the sensor values a unique tendency, which is very important when
looking for an agronomic algorithms to guide nitrogen application, because it
simplifies as a unique data set can guide the sensor in different situations. At 0.2
m of stem height the values are low and concentrated and most under 20 kg N ha-1
being this field average height two early to obtain good information from the
sensor. The 0.3 m crop height data present better values showing a good
prediction from 10 to 60 kg N ha-1. Also the 0.4 m crop height data that can
predict the N-uptake from 15 to 70 kg N ha-1, the 0.5 m have the longer range of
N-uptake prediction (20 to almost 100 kg N ha-1), but the saturation phenomenon
starts to appear. At 0.6 m average crop height, data have many sampled spots with
higher crop, with stems having 0.7 to 1.0 m, what causes saturation on the sensor
decreasing the sensor accuracy on these places. The three best heights combining
in a single data set are presented in Figure 7.
y = 0.8732x + 5.326
R2 = 0.83
0
20
40
60
80
100
120
0 20 40 60 80 100 120
Real N-uptake (Kg ha-1)
Predicted N-uptake (kg ha-1)
0.3 m
0.4 m
0.5 m
Linear ("1/1")
Linear (0.3 - 0.5 m)
Figure 7: Integration of sugarcane real measured N-uptake compared to sensor
predicted N-uptake for the 0.3 to 0.5 meter heights.
Using only the crop heights of 0.3 to 0.5 m there is a significant
improvement in the prediction of nitrogen initially presented by Portz et al.
(2012). Not by a higher coefficient of determination (R²), but because it shows a
best slope of the prediction (closer to the 1 to 1 line) that comes to reduce the
error between the real and predicted values.
It is also noted that even decreasing the amplitude of the average heights
sampled, there was no decrease in the amplitude of N-uptake predicted by the
sensor.
CONCLUSIONS
At 0.2 m of field average stem height, sugarcane biomass is low for a good
sensor prediction of the parameters. At 0.6 m height starts the phenomenon of
saturation, where the ability of the sensor to predict biomass and nitrogen begins
to be affected.
Between 0.3 and 0.5 m of stem average height results show the best
correlation between real and sensor predicted biomass and nitrogen uptake for
sugarcane crop, indicating that this is the right period for using the sensor to guide
variable rate nitrogen application.
ACKNOWLEDGMENTS
All this work would not be possible without the collaboration of São
Martinho’s Mill team and Máquinas Agrícolas Jacto support. We also
acknowledge the National Council for Scientific and Technological Development
(CNPq) and the São Paulo Research Foundation (FAPESP) for providing
doctorate scholarship to the first and second authors.
REFERENCES
FAO. (2012). Food and Agriculture Organization, Faostat. Accessed on April 05,
2012.: Available on-line at <http://faostat.fao.org/>.
Heege, H.J.; Reusch, S.; Thiessen, E. (2008). Prospects and results for optical
systems for site-specific on-the-go control of nitrogen-top-dressing in
Germany. Precision Agriculture, Dordrecht, v. 9, n. 3, p. 115-131.
Jasper, J., Reusch, S., and Link, A. (2006). N-Sensor ALS-Active sensing of crop
N status using optimized wavelength combination. In D. J. Mulla (Ed.),
Proceedings of the 8th International Conference on Precision Agriculture and
Other Precision Resources Management (July 23–26). St. Paul, MN, USA:
Precision Agriculture Center, University of Minnesota.
Jasper, J., Reusch, S. and Link, A. (2009). Active sensing of the N status of
wheat using optimized wavelength combination impact of seed rate, variety
and growth stage. In: Van Henten, E.J., D. Goense and C. Lokhorst (eds.),
Precision agriculture ’09. Proceedings of the 7th European Conference on
Precision Agriculture. July 5-8, Wageningen, The Netherlands: 23-30.
Link, A., J. Jasper, and S. Reusch. (2005). Suitability of different crop parameters
for the determination of site-specific nitrogen fertilizer demand. In: Stafford, J.
V. (ed.), Precision Agriculture ’05. Proceedings of the 5th European
Conference on Precision Agriculture. June 9-12, Uppsala, Sweden.
Mokhele, T.A.; Ahmed, F.B. (2010). Estimation of leaf nitrogen and silicon using
hyperspectral remote sensing. Journal of Applied Remote Sensing,
Amsterdam, v. 4, p. 18, Nov.
Mutanga, O. & A. K. Skidmore (2004) Narrow band vegetation indices overcome
the saturation problem in biomass estimation. International Journal of Remote
Sensing, 25, 3999-4014.
Portz, G.; Molin, J.P.; Jasper, J. (2012) Active crop sensor to detect variability of
nitrogen supply and biomass on sugarcane fields, Precision Agriculture.
Volume 13, 1, Page 33-44. DOI 10.1007/s11119-011-9243-4.
Reusch, S. (2005). Optimum waveband selection for determining the nitrogen
uptake in winter wheat by active remote sensing. P. 261-266. In: Stafford, J.
V. (ed.), Precision Agriculture ’05. Proceedings of the 5th European
Conference on Precision Agriculture. June 9-12, Uppsala, Sweden.
Wageningen Academic Publishers, Wageningen, The Netherlands.
Silva, C. B., M. A. F. D. de Moraes & J. P. Molin (2010) Adoption and use of
precision agriculture technologies in the sugarcane industry of São Paulo state,
Brazil. Precision Agriculture, 1-15.
Singh, I., Srivastava, A. K., Chandna, P., & Gupta, R. K. (2006) Crop sensors for
Efficient Nitrogen Management in Sugarcane: Potential and Constraints. Sugar
Tech.
Solie, J.B.; Raun, W.R.; Stone, M.L. (1999). Submeter spatial variability of
selected soil and Bermuda grass production variables. Soil Science Society of
America Journal, Madison, v. 63, n. 6, p. 1724-1733, Nov./Dec.
Thenkabail, P. S., Smith, R. B., and De Pauw, E., (2000). Hyperspectral
vegetation indices and their relationships with agricultural crop characteristics.
Remote Sensing of Environment, 71, 158–182.
Todd, S. W., Hoffer, R. M., and Milchunas, D. G., (1998). Biomass estimation on
grazed and ungrazed rangelands using spectral indices. International Journal of
Remote sensing, 19, 427–438.
Tucker, C. J., (1977), Asymptotic nature of grass canopy spectral reflectance.
Applied Optics, 16, 1151–1156.
... The canopy sensor has to be utilized when the sugarcane tallness is between 40 and 70 cm to get estimation affectability to sugarcane vigor fluctuation [165,166]. At this stage, sugarcane has attained around 10-30% of total biomass with 27-68% N, which is dependent on genotype, soil fertility, climate and developmental stage [167]. ...
... The canopy sensor has to be utilized when the sugarcane tallness is between 40 and 70 cm to get estimation affectability to sugarcane vigor fluctuation [165,166]. At this stage, sugarcane has attained around 10-30% of total biomass with 27-68% N, which is dependent on genotype, soil fertility, climate and developmental stage [167]. ...
Article
Full-text available
Nitrogen is the main limiting nutrient after carbon, hydrogen and oxygen for photosynthetic process, phyto-hormonal , proteomic changes and growth-development of plants to complete its lifecycle. Excessive and inefficient use of N fertilizer results in enhanced crop production costs and atmospheric pollution. Atmospheric nitrogen (71%) in the molecular form is not available for the plants. For world's sustainable food production and atmospheric benefits, there is an urgent need to upgrade nitrogen use efficiency in agricultural farming system. The nitrogen use efficiency is the product of nitrogen uptake efficiency and nitrogen utilization efficiency, it varies from 30.2 to 53.2%. Nitrogen losses are too high, due to excess amount, low plant population, poor application methods etc., which can go up to 70% of total available nitrogen. These losses can be minimized up to 15-30% by adopting improved agronomic approaches such as optimal dosage of nitrogen, application of N by using canopy sensors, maintaining plant population, drip fertigation and legume based intercropping. A few transgenic studies have shown improvement in nitrogen uptake and even increase in biomass. Nitrate reductase, nitrite reductase, glutamine synthetase, glutamine oxoglutarate aminotransferase and asparagine synthetase enzyme have a great role in nitrogen metabolism. However, further studies on carbon-nitrogen metabolism and molecular changes at omic levels are required by using "whole genome sequencing technology" to improve nitrogen use efficiency. This review focus on nitrogen use efficiency that is the major concern of modern days to save economic resources without sacrificing farm yield as well as safety of global environment, i.e. greenhouse gas emissions, ammonium volatilization and nitrate leaching.
... However, these sensors exhibit signal saturation when the sugarcane crop is above the 0.6 m stem height. As shown previously (Portz et al., 2012b), increasing biomass does not reflect an adequate increase of the vegetation index (VI) later in the season when there is still room to conduct nitrogen fertilization inseason. As an alternative, measuring crop canopy using ultrasonic proximity sensors is not new. ...
Article
Full-text available
The use of crop canopy reflectance sensors is being studied intensively in sugarcane crops; however, real-time measurements of plant height provide another possibility for estimating the spatial variability of biomass and nitrogen uptake by the crop during the in-season fertilization period. An ultrasonic proximity sensor system was deployed to measure crop canopy height. The data were combined with commercially available canopy optical reflectance measurements to characterize spatial variability of crop growth in two commercial areas (a sandy soil field and a clay soil field) during two different growing stages. After mapping, ten points in each field were defined to determine plant biomass and N-uptake through manual sampling and traditional measurements. Through the data analysis, a pixel by pixel comparison was performed to relate the interpolated maps obtained using different sensor systems. The ten points in each field were used to relate the actual biomass and N-uptake with the sensor data. It was shown that all correlations were significant and there was a slight indication that canopy reflectance sensing produced a better assessment of crop growth at the earlier growth stage whereas the ultrasonic sensor resulted in more accurate predictions at the later growing stage.
... The canopy sensor should be used when the sugarcane stem height is between 0.4 to 0.7 m to achieve measurement sensitivity to sugarcane vigor variability (Portz et al., 2012;Amaral and Molin, 2014). Thus, at this sensing time (?December-January), sugarcane has accumulated about 10 to 30% of its total biomass and 27 to 68% of its total N, depending on cultivar, soil, climatic conditions, and the period when the crop was previously harvested (information extracted from Oliveira, 2011). ...
Article
Full-text available
Nitrogen fertilization is challenging for sugarcane (Saccharum spp.) producers due to its complex interaction with the crop and soil. Thus, the main goal of this study was to develop a feasible approach to guide variable-rate N application in sugarcane based on canopy sensor readings. This study was conducted for 5 yr. Several plot and strip N-rate experiments were conducted under a wide range of crop conditions in Brazil and evaluated with the Crop Circle active canopy sensor (Holland Scientific Inc.). Because of variability in crop density and growth development within sugarcane fields, the use of an N-rich reference area to estimate the crop response to N application was compromised. Biomass was the main crop parameter influencing canopy sensor readings, allowing yield estimation because biomass typically results in stalk yield. Thus, canopy sensor readings can efficiently predict relative sugarcane yield when working with data that are normalized to the mean for the field. Hence, an algorithm that takes into account this relationship was established. The concept of this algorithm is to apply higher N fertilization rates where the sugarcane yield potential is higher. Such an approach was determined to be useful to guide N application in sugarcane fields. Nevertheless, field validation is needed to confirm this N management strategy. Besides, more information about sugarcane biomass variability within fields may be required to increase algorithm efficiency. Paper can be accessed in: https://dl.sciencesocieties.org/publications/aj/abstracts/107/4/1513
Article
Full-text available
The potential to estimate the nutrient status in important agricultural crops such as maize and sugarcane is of significant interest. In South African sugarcane agriculture, just like in global ecosystem, the estimation of Nitrogen (N) and Silicon (Si) is very important. These nutrients are one of the factors influencing the prevalence of the stalk borer Eldana saccharina Walker (Lepidoptera: Pyralidae). Therefore, the researchers aim at estimating leaf N and Si concentration as well as their ratio in sugarcane using hyperspectral remote sensing (spectroradiometry) for monitoring E. saccharina. A hand-held Analytical Spectral Devices (ASD) Field Spec® 3 spectroradiometer was used to take leaf spectral measurements of sugarcane plants from a potted-plant trial taking place under shade house conditions. In this trial, nitrogen and silicon nutrient applications as well as varieties used were known. In addition, watering regimes and artificial infestation of E. saccharina were carefully controlled. The study results indicate that the Red-edge Index (R740/R720) is linearly related to N concentration (R2 = 0.81, Root Mean Square Error (RMSE) = 0.103) for N37 with the highest correlation coefficient. For Si, the index (R750-R560)/(R750+R560) was linearly related to Si concentration (R2 = 0.53, RMSE = 0.118) for N25. Finally, the N:Si ratio was linearly correlated to the index (R1075-R730)/(R1075+R730) (R2 = 0.67, RMSE = 1.508) for N37, hence this index can be used for early detection of E. saccharina damage or for identifying sugarcane that is prone to attack by E. saccharina. It was concluded that hyperspectral remote sensing has potential for use in estimating the N:Si ratio and E. saccharina potential infestations can be monitored rapidly and nondestructively in sugarcane under controlled conditions. It is recommended that an advanced study be conducted in field conditions using airborne and/or spaceborne hyperspectral sensors.
Article
Aboveground biomass was estimated on the shortgrass steppe of Eastern Colorado using Landsat TM Tasseled Cap green vegetation index (GVI), brightness index (BI), and wetness index (WI), the normalized difference vegetation index (NDVI) and the red waveband (RED), for two grazing treatments (moderately grazed or ungrazed). Field measurements of standing crop were obtained on six sites per grazing treatment. Ordinary least squares regression models of biomass as a function of one or more indices were tested for grazed, ungrazed, and combined grazed and ungrazed data. Biomass from grazed sites was linearly related to GVI, NDVI, WI, and RED indices (R2 0.62-0.67). Ungrazed sites produced no significant relations. With combined ungrazed and grazed data, biomass was not significantly related to GVI, NDVI, WI, or BI, and was poorly related to the RED index (R2 0.35). When grazing treatments were treated as dummy variables for the combined data, the RED index was moderately related to biomass (R2 0.70). These results suggest that information about grazing utilization is useful for estimating aboveground biomass in rangelands. The RED index appears to be sensitive to biomass variations for green vegetation and to a lesser extent dry or senescent vegetation when relatively bright soil backgrounds are present which is often the case for semi-arid environments such as the shortgrass steppe.
Article
The optimal resolution at which soil and plant variables should be sensed and treated is not well defined. This study was conducted to determine the semivariance range where soil test and plant variables were related, and to estimate the minimum spatial scale at which variable rate applications of nutrients should be made. Soil and plant analyses were performed in 490 0.3- by 0.3-m plots from bermudagrass (Cynodon dactylon L.) sod at two locations. Eight soil cores (0-15 cm deep) were collected and composited from each 0.3- by 0.3-m plot. Semivariance analysis was used to estimate the range over which samples of the five soil variables (total N, extractable P, and K, organic C, and pH) and two plant variables (forage total N and biomass) were related. Semivariance statistics including the nugget, sill, correlation range, and integral scale were calculated. Correlation ranges were between 1.9 and 11.4 m with corresponding integral scales between 0.5 and 2.1 m. At one location, P exhibited nested sills with multiple ranges. Results indicate that the fundamental field-element dimensions (the area over which variable rate fertilizer applicators should sense and apply materials) is likely to be 1.0 by 1.0 m or smaller. To describe the variability encountered in these experiments, soil and plant measurements should be made at the meter or submeter level.
Article
Nitrogen management has been intensively studied on several crops and recently associated with variable rate on-the-go application based on crop sensors. Such studies are scarce for sugarcane and as a biofuel crop the energy input matters, seeking high positive energy balance production and low carbon emission on the whole production system. This article presents the procedure and shows the first results obtained using a nitrogen and biomass sensor (N-Sensor™ ALS, Yara International ASA) to indicate the nitrogen application demands of commercial sugarcane fields. Eight commercial fields from one sugar mill in the state of São Paulo, Brazil, varying from 15 to 25 ha in size, were monitored. Conditions varied from sandy to heavy soils and the previous harvesting occurred in May and October 2009, including first, second, and third ratoon stages. Each field was scanned with the sensor three times during the season (at 0.2, 0.4, and 0.6 m stem height), followed by tissue sampling for biomass and nitrogen uptake at ten spots inside the area, guided by the different values shown by the sensor. The results showed a high correlation between sensor values and sugarcane biomass and nitrogen uptake, thereby supporting the potential use of this technology to develop algorithms to manage variable rate application of nitrogen for sugarcane.
Article
The objective of this paper is to determine spectral bands that are best suited for characterizing agricultural crop biophysical variables. The data for this study comes from ground-level hyperspectral reflectance measurements of cotton, potato, soybeans, corn, and sunflower. Reflectance was measured in 490 discrete narrow bands between 350 and 1,050 nm. Observed crop characteristics included wet biomass, leaf area index, plant height, and (for cotton only) yield. Three types of hyperspectral predictors were tested: optimum multiple narrow band reflectance (OMNBR), narrow band normalized difference vegetation index (NDVI) involving all possible two-band combinations of 490 channels, and the soil-adjusted vegetation indices. A critical problem with OMNBR models was that of “over fitting” (i.e., using more spectral channels than experimental samples to obtain a highly maximum R2 value). This problem was addressed by comparing the R2 values of crop variables with the R2 values computed for random data of a large sample size. The combinations of two to four narrow bands in OMNBR models explained most (64% to 92%) of the variability in crop biophysical variables. The second part of the paper describes a rigorous search procedure to identify the best narrow band NDVI predictors of crop biophysical variables. Special narrow band lambda (λ1) versus lambda (λ2) plots of R2 values illustrate the most effective wavelength combinations (λ1 and λ2) and bandwidths (Δλ1 and Δλ2) for predicting the biophysical quantities of each crop. The best of these two-band indices were further tested to see if soil adjustment or nonlinear fitting could improve their predictive accuracy. The best of the narrow band NDVI models explained 64% to 88% variability in different crop biophysical variables. A strong relationship with crop characteristics is located in specific narrow bands in the longer wavelength portion of the red (650 nm to 700 nm), with secondary clusters in the shorter wavelength portion of green (500 nm to 550 nm), in one particular section of the near-infrared (900 nm to 940 nm), and in the moisture sensitive near-infrared (centered at 982 nm). This study recommends a 12 narrow band sensor, in the 350 nm to 1,050 nm range of the spectrum, for optimum estimation of agricultural crop biophysical information.
Article
Remotely sensed vegetation indices such as NDVI, computed using the red and near infrared bands have been used to estimate pasture biomass. These indices are of limited value since they saturate in dense vegetation. In this study, we evaluated the potential of narrow band vegetation indices for characterizing the biomass of Cenchrus ciliaris grass measured at high canopy density. Three indices were tested: Modified Normalized Difference Vegetation Index (MNDVI), Simple Ratio (SR) and Transformed Vegetation Index (TVI) involving all possible two band combinations between 350 nm and 2500 nm. In addition, we evaluated the potential of the red edge position in estimating biomass at full canopy cover. Results indicated that the standard NDVI involving a strong chlorophyll absorption band in the red region and a near infrared band performed poorly in estimating biomass (R2=0.26). The MNDVIs involving a combination of narrow bands in the shorter wavelengths of the red edge (700-750 nm) and longer wavelengths of the red edge (750-780 nm), yielded higher correlations with biomass (mean R2=0.77 for the highest 20 narrow band NDVIs). When the three vegetation indices were compared, SR yielded the highest correlation coefficients with biomass as compared to narrow band NDVI and TVI (average R2=0.80, 0.77 and 0.77 for the first 20 ranked SR, NDVI and TVI respectively). The red edge position yielded comparable results to the narrow band vegetation indices involving the red edge bands. These results indicate that at high canopy density, pasture biomass may be more accurately estimated by vegetation indices based on wavelengths located in the red edge than the standard NDVI.
Article
Signals for site-specific nitrogen-top-dressing can be obtained by a sensor mounted on a tractor. The plant appearance can serve as a criterion. The question is which plant criteria provide pertinent information and how this can be indicated. Increasing the nitrogen-supply changes leaf colour from yellow-green to blue-green via the chlorophyll-concentration in the leaves and leads to growth of plants. Present sensing systems measure either chlorophyll concentration in the leaves, total area of the leaves or crop resistance against bending. The aim and purpose of this study is to outline prospects for application. Therefore, the emphasis is on results and not on experimental methods, to which references are given. Mainly optical sensing systems relying on reflectance or fluorescence are dealt with. Good signals of the nitrogen-supply can be obtained from the red edge plus the near infrared range of the reflectance. The results with some new spectral indices were better than those with standard spectral indices. Fluorescence sensing instead of reflectance sensing eliminates erroneous signals from bare soil. However, only low supply rates were clearly indicated. The biomass of the crop or the total area of its leaves is a very important criterion. Reflectance indices can take this into account. Fluorescence signals are barely influenced by this parameter.
Article
A real-time crop sensor for site specific input application is the new innovation in the field of precision agriculture. At least three different type of crop sensors,viz.. Soil Doctor, N-Senior and Green Seeker have been used in different field crops. The key advantage of all these systems is that they do not need recommendation maps. However, no published data is available on Soil Doctor adoption by farmers due to company’s aggressiveness for protective patent rights. The N-Sensor is being used mainly for wheat and other small grain crops. However, one of the key limitations of N-Sensor is ambient light source. Handheld Green Seeker sensor is the latest addition to the list of crops sensors. The active light source is a major advantage of the Green Seeker sensor. Our preliminary observations on NDVI in relation to canopy development and crop growth in sugarcane are very encouraging and we envisage a potential scope of Green Seeker optical sensor for monitoring crop growth in order to adjust timing and dose of N application for maximizing cane and sugar productivity.
Article
The asymptotic nature of grass canopy spectral reflectance has been evaluated from field experimental data collected over the wavelength region of 0.500-1.000 microm at 0.005-microm intervals. The spectral reflectance of green vegetation against a soil background decreases in regions of absorption and increases in regions of minimal or no absorption as the vegetational density increases until a stable or unchanging spectral reflectance, called the asymptotic spectral reflectance, is reached. Results indicated spectral reflectance asymptotes occurred at significantly lower levels of total wet biomass, total dry biomass, dry green biomass, chlorophyll content, and leaf water content in regions of strong pigment absorption (low detectability threshold) than in the photographic ir region where absorption was at a minimum (high detectability threshold). These findings suggested that photographic ir sensors were more suited to remote sensing of moderate to high biomass levels or vegetational density in a grass canopy than were sensors operating in regions of the spectrum where strong absorption occurred.