|Collection Mémoires et thèses électroniques|
|AccueilÀ proposNous joindre|
Table des matières
BÉLANGER, M.-C 1., VIAU, A.A. 1, SAMSON, G. 2, CHAMBERLAND, M. 3, soumis à Agronomy journal Determination of a Multivariate Indicator of Nitrogen Imbalance (MINI) in potato using reflectance and fluorescence spectroscopy, 1 Laboratoire de Géomatique Agricole et d’Agriculture de précision, Local 3731-A, Pavillon Casault, Université Laval, Qc, G1K 7P4. 2 Université du Québec à Trois-Rivières, Case postale 500, Trois-Rivières, Qc, G9A 5H7. 3 Telops, 100-2600, avenue St-Jean-Baptiste, Québec, Québec, G2E 6J5. Ce chapitre est le fruit du travail de la candidate. À titre de directeurs et superviseur, messieurs Viau, Samson et Chamberland ont apporté des suggestions et commentaires quant au design et à la préparation de ce chapitre.
Dans le cadre de cette étude, nous avons évalué le potentiel de la réflectance et de la fluorescence pour la détection d’un déséquilibre nutritif engendré par un faible apport en nutriments. Des carences nutritives (N, K et Mg) ont été induites sur des plants de pomme de terre cultivés en serre. Leur état de déséquilibre nutritif a été déterminé en utilisant le diagnostic de composition nutritionnelle (CND). Il a été possible de discriminer les plants en déséquilibre des plants sains en utilisant les mesures de réflectance et de fluorescence. De plus, leur combinaison dans une variable canonique a permis une meilleure détection des plants en déséquilibre azoté. En utilisant la variable canonique, un indicateur de déséquilibre azoté a été développé. Il a permis de détecter près de 70 % des plants en déséquilibre et plus de 90% des plants sains. Cet indicateur, de par sa rapidité d’acquisition et de traitement, détecte hâtivement les carences en azote et pourrait permettre une fertilisation efficace.
In this study we evaluated the potential of reflectance and fluorescence for the detection of nutrient imbalance generated by a nutrient shortage. Potato plants were grown in a greenhouse and three different nutrient deficiencies were induced (K, Mg and N). Their global nutrient imbalance was determined using the compositional nutrient diagnosis (CND). Using reflectance and fluorescence it was possible to detect nutrient imbalance. The combination of reflectance and fluorescence parameters provided the best detection of unbalanced plants using canonical discriminant analysis. A new multivariate indicator of nitrogen imbalance (MINI) was developed using the canonical variable computed from reflectance and fluorescence parameters. MINI can detect almost 70% of the N-deficient plants and more than 90 % of the N-sufficient plants. This indicator allows a rapid data acquisition and handling, and provides deficiency detection within the time-window necessary for plant response to recovery fertilization.
Potato crop is cultivated in more than 120 countries worldwide with an annual production of 310 million metric tons (FAO, 2004). Potato crops are generally grown on sandy soils and can suffer from nutrient stresses resulting from nutrient leaching or unbalanced fertilization. In order to assure nutrient balance and improve nutrient management during the growing season, nitrogen and other nutrient deficiencies must be diagnosed precisely. Critical nutrient levels can assist in nutrient diagnosis, but nutrient interactions may interfere with the interpretation of the results (Munson and Nelson, 1990). The Diagnosis and Recommendation Integrated System (DRIS) (Beaufils, 1973) computes plant nutritional status using many dual ratios of leaf nutrient concentration, the minimum expression for nutrient interactions. Compositional Nutrient Diagnosis (CND) (Parent & Dafir, 1992), based on compositional data analysis (Aitchison, 1986), defines a simplex of nutrient composition bounded at 100 %. Relative CND nutrient balance is computed using the geometrical mean of all nutrients including a filling value, and provides information on nutrient interactions and imbalances. The CND nutrient imbalance index has a chi-squared distribution (Khiari et al., 2001a).
Up to now, nutrient diagnosis techniques have been based on destructive analyses of leaves’ mineral contents. Remote sensing techniques such as fluorescence and reflectance might lower the costs and accelerate the acquisition, handling, and analysis of plant tissues and do it in a non-destructive manner. Those two techniques respond to plants’ structural and biochemical properties which are sensitive to plant nutrient status, notably the content in photosynthetic pigments and phenolic compounds (Cerovic et al., 1999; Read et al., 2002; Schuerger et al., 2003). Indeed, several studies have demonstrated the possibility to detect plant mineral deficiencies, especially for nitrogen, by reflectance (Adams et al., 1993; Gamon et al., 1997; Haboudane et al., 2002; Vouillot et al., 1998; Zhao et al., 2003) and by UV-induced fluorescence (Apostol et al., 2003; Chappelle et al., 1984; Corp et al., 2003; McMurthey III et al., 1994; Mercure et al., 2004; Tartachnyk & Rademacher, 2003). To evaluate the efficiency of reflectance and/or fluorescence to detect plant nutrient deficiencies, most studies consider the leaves’ pigment content (chlorophyll), some take into account the treatment induced and a few use biomass or nutrient content. As mentioned in Bélanger et al. (submitted), it is important to measure growth parameters to evaluate the changes induced on fluorescence or reflectance indices. Moreover, as only one element cannot reflect the overall status of a plant it is important to consider many nutrients and their potential interactions by determining the relation between the CND and the remote sensing parameters. Because of their complementary nature, McMurthey III et al., (1994) suggested combining reflectance and fluorescence to improve the effectiveness to detect nutrient deficiencies; this was confirmed in a previous experiment conducted on N-, K-, and Mg-deficient potato plants (Bélanger et al., submitted). By using a dimension-reduction technique, the canonical discriminant analysis, on reflectance and fluorescence indices, an indicator of nutrient imbalance in potato has been developed.
The aims of this work are 1) to determine if there is a relation between reflectance, fluorescence and nutrient imbalance (CND_r2) in potato, and 2) to develop a multivariate indicator of the nitrogen imbalance (MINI) in potato plants combining reflectance and fluorescence parameters using canonical discriminant analysis.
The processes of data acquisition and treatment are schematized in fig. 1 and detailed subsequently. Optical measurements were taken once a week between the 15th and the 44th days after emergence (DAE) and were used to compute a canonical discriminant variable. Chemical analyses of leaf tissues were realized to calculate the nutrient imbalance using compositional nutrient diagnosis (CND). A multivariate indicator of nitrogen imbalance (MINI) was computed using the canonical variable and tested on a validation dataset. Its efficiency was evaluated by estimating alpha and beta error as well as its predictive value (yield).
Three experiments (A, B, and C) took place in 2002 (A, B) and 2003 (C) in a greenhouse located on Laval University Campus (Quebec City, Canada). The experimental plan was set as a nested effects design including five blocks and three treatments (N, K, Mg) applied at four levels (15, 30, 60, 100). A total of 170 potato plants (Solanum tuberosum, L. cv. Superior), were grown in 3L containers, at 22°C day, 16°C night controlled temperatures and under a 16-hour photoperiod. The growing medium was a mix of 80 % vermiculite (Fafard et Frères inc, QC, Canada) and 20 % quartz sand (Unimin Corporation, CT, USA), washed with demineralized water prior to planting (Tukaki & Mahler, 1990). From emergence to the end of experiment, control plants were fertilized using a complete mineral solution (Tukaki & Mahler, 1990) (N 160 ppm; P 29 ppm; K 234 ppm; Ca 160 ppm; Mg 48 ppm; S 62 ppm; Fe 1.83 ppm; Mn 0.5 ppm; B 0.5 ppm; Zn 0.05 ppm; Cu 0.02 ppm; Mn 0.01 ppm). To induce mineral deficiencies, other plants received modified mineral solutions providing 15, 30 or 60 % of N, K or Mg concentrations received by the control treatment.
Foliar analyses realized for the CND computation were taken at the blooming stage because it corresponds to a period of high nutrient uptake during which potato crops are sensitive to nutrient imbalance (Parent & Dafir, 1992). Thus, on every plant, foliar analyses were conducted on the 4th fully expanded leaf, at full bloom corresponding to the 37th day after emergence (DAE) for experiments A and to the 44th DAE for experiment B and C. Plant tissues were dried (for 48h at 65°C), and ground and digested as described by Parkinson & Allen, (1975). Nutrient concentrations (P, K, Ca, and Mg) were determined using an Inductively Coupled Plasma (ICP), OPTIMA model 4300DV (Perkin-Elmer, Boston, MA). Nitrogen concentration was determined using the Quikchem method (Zellweger Analytic inc.) on a Flow Injection Analyser (FIA), model Quikchem4000 (Lachat Instruments, Milwaukee, WI).
Our population of potato plants was divided into high- and low-yield subpopulations. The yield cutoff was set according to the biomass produced and transposed into the induced nutrient treatment. The potato plants included in the high-yield subpopulation are considered to be ‘balanced’. CND computation has been done following Khiari et al., (2001b) and is presented here in four different steps. All four steps were computed for each potato plant grown under our experimental conditions.
Step 1: Determining the ‘filling’ value
Leaf composition is bounded to 100 % by summing values for individual elements and a ‘filling’ value. The compositional simplex ( ) composed of five elements may be defined as follows (Khiari et al., 2001b):
where N, P, K ... are the nutrient proportions (%) and
R5 the filling value which represents the undetermined elemental composition as follows:
Step 2: Computing log-centered ratio:
Each nutrient concentration is divided by the geometrical mean to account for all nutrient interactions simultaneously and the natural log is computed.
d=number of analyzed nutrients (here 5)
Step 3: Computing the nutrient imbalance index
The Vx mean and standard deviation of the high-yield subpopulation are the CND norms to compute nutrient imbalance indices as follows:
where is the VN mean for the high-yield subpopulation,
is the VN std deviation for the high-yield subpopulation,
is the nitrogen imbalance index,
is the phosphorus imbalance index, and so on....
Step 4: Computing the global nutrient imbalance index
The global nutrient imbalance index is the sum of squared nutrient indices as follows:
where is the nitrogen imbalance index and is the phosphorus imbalance index, and so on....
The sum of d+1 squared independent and unit-normal variables produces a new variable having a chi-squared distribution with d+1 degrees of freedom (Ross, 1987). A threshold nutrient imbalance index has to be determined to divide the population into unbalanced and balanced potato plants. For a chi-squared distribution, the threshold global nutrient imbalance index (CND_r2) should be 12.6 for a having six degrees of freedom and α=0.05. The threshold global nutrient imbalance index is verified using the Cate-Nelson partitioning procedure (Nelson & Anderson, 1977). An ANOVA is first computed to reach the highest sum of squares, and the threshold value is confirmed graphically by maximizing the number of points in diagonal quadrants (fig. 4). Summing critical squared nutrient imbalance indices independently determined by the Cate-Nelson partitioning procedure also validates the threshold global nutrient imbalance index.
In 2002, fluorescence measurements were taken at 15, 23, 30, and 37 DAE (experiment A) and at 44 DAE (experiment B) and were made using an Xe-PAM (Walz, Effeltrich, Germany). In 2003, fluorescence measurements were taken the 37 and 44 DAE (experiment C) using the FLUTE prototype (GAAP-TELOPS, Qc, Canada), presented in fig. 2. Both fluorometers can achieve UV-induced fluorescence measurements.
The Xe-Pam measurements were taken on a leaf disk placed on the instrument’s sample holder. The FLUTE measurements were taken 0.5 m from the top of the whole plant under ambient light. The two fluorometers used a Xenon (Xe) flash lamp to induce excitation. UV and blue excitations were obtained by respectively placing in front of the Xe-excitation flash lamp, DG11+UG11 and BG39+UV blocking filters (Schott Glass Technologies, PA, USA) for the Xe-PAM, and 360±40 nm, 436± 20 nm (Chroma Technology Corp., VT, USA), for the FLUTE. For the Xe-Pam, the detection was made at 440, 520, 690 and 750nm ± 10 nm using Oriel filters of 2.54 cm diameter (Spectra-Physics, CT, USA). For the FLUTE, detection was made at 440, 520, 690 and 740 nm ± 10 nm using 5.08 cm diameter filters (CVI Laser, NM, USA). Intensities of UV-induced fluorescence were measured at the four emission bands whereas blue-induced fluorescence intensities were measured at 690 and 740 nm. Fluorescence intensities were calibrated against photodiode sensitivity and transmittance of the filters at the different wavelengths. Various ratios of fluorescence intensities were computed for data analysis (Tab. 1).
Reflectance spectra ranging from 350 to 1500 nm (sampling interval 1.4 nm) were collected using a Field Spec Pro spectroradiometer (Analytical Spectral Devices inc., CO, USA). All reflectance spectra were taken between 1000h and 1400h, 0.3 m directly above the plants, under ambient light in the greenhouse. Three spectra for each plant were collected at 15, 23, 30, 37 DAE (experiement A), at 44 DAE (experiment B) in 2002, at 37 and 44 DAE (experiment C) in 2003. The instrument was regularly calibrated against the ambient light using a Spectralon plate (Labsphere, inc., NH, USA). Raw reflectance data were extracted from the spectroradiometer using the FSVNIR software package (Analytical Spectral Devices inc., CO, USA) as an ASCII file, which was imported into Matlab (The Mathworks, V.6.5, 2000 MA, USA). For each plant, an average from three spectra (30 acquisitions) was computed; the averaged spectrum was smoothed by a Savitsky-Golay filter (order 5, window 21), and 17 reflectance indices were calculated (Table 2).
In order to identify outliers a cluster analysis was performed using the CLUSTER procedure from SAS software package (SAS institute, V8.2). An ANOVA was conducted on dry shoot biomass using the GLM procedure and the LSMEANS option for simple means comparison. The CANDISC procedure was used across reflectance and fluorescence parameters to generate new canonical variables (linear combinations of quantitative variables) for inclusion into a discriminant function. The sum of raw canonical coefficients multiplied by the centered reflectance and fluorescence indices produces a candisc value for each potato plant (equation ). Several discriminant analyses (depending on the classification variable) were conducted using the DISCRIM procedure on the canonical variables created, in order to classify each plant into its corresponding treatment resulting in a classification percentage.
A= fluorescence or reflectance ratios
Ax=centered fluorescence or reflectance ratios
and i= raw canonical coefficient for parameter I
We developed the Multivariate Indicator of Nitrogen Imbalance (MINI) using reflectance and fluorescence observations (total n=313) taken on potato plants from experiment A at 15 (n=46), 23 (n=59), 30 (n=59) and 37 (n=55) DAE and from experiment C taken at 37 (n=46) and 44 (n=48) DAE. Once the canonical coefficients were computed, MINI was validated using reflectance and fluorescence observations (n=60) taken at 44th DAE on potato plants from experiment B. The nitrogen imbalance index for experiment B was computed following the steps presented earlier (Eq. [1 to 6]).
To evaluate the efficiency of MINI, we estimated error α, error β and the predictive value. In terms of nutrient stress detection and fertilization, error α represents the probability of incorrectly concluding that plants should be fertilized and error β corresponds to the probability of incorrectly concluding that plants should not be fertilized. Increasing error α will increase fertilization cost, enhance leaching and environmental contamination risks, and reduce profits. By convention, error α is usually set at 5 % whereas sometimes it can be set at a higher level (10 or 20%) depending on the experience, the variables tested and on the researchers’ preference (Guénette, 2003; Irvine et al., 1999). Increasing error β will reduce yield potential and hence profits. The power of a test defined as 1-β, should be approximately 80 % (Cohen, 1988). Both errors have been estimated using a Gaussian distribution of probabilities for a population X (equation [8-9]).
which probability distribution function is
where and 
The predictive value gives the probability of having a real positive subject when tested positive (Schork & Remington, 2000). The frequencies are denoted by the variables a, b, c, and d, and a, for example, corresponds to the number of subjects (here plants) tested positive when the reference is positive (Tab. 3). The predictive value can be computed using equation . It is useful to evaluate the odds of identifying a real positive subject, for example.
Predictive value 
The Results section is divided into four sections presenting respectively the effects of the induced nutrient deficiencies on the dry shoot biomass, the compositional nutrient diagnosis, the discriminant analyses and the development and computation of MINI.
The effects of the induced nutrient deficiencies on the dry shoot biomass are presented in Tab. 4. The dry shoot biomass was significantly reduced by the nitrogen deficiency in experiment A and C (respectively by 39.1 % and 40.9 %). Because there were, no significant effects observed on the dry shoot biomass for magnesium and potassium treatments, the low yield subpopulation for the computation of the compositional nutrient diagnosis was only composed of plants receiving 15% or 30% of nitrogen.
The CND nutrient norms were computed as the mean and standard deviation of the high yield subpopulation (Tab. 5). The nutrient norms are similar to those obtained by Khiari et al., (2001c) and Parent et al., (1994). Nutrient sufficiency indices were computed using the CND norms presented in Tab. 5. The fig. 3 confirms that the CND_r2 is distributed as a chi-squared variable.
Tab. 5 Compositional nutrient diagnosis (CND) norms (Vx*) corresponding to the Vx mean and standard deviation for the high yield subpopulation.
The CND_r2 obtained by the Cate-Nelson partitioning procedure was 11.5 meaning that plants having a CND_r2 lower than 11.5 will be considered balanced. fig. 4 shows the relationship between dry shoot biomass and the global nutrient imbalance index (CND_r2) on a Cate-Nelson graph to validate graphically the previous result.
The six critical nutrient imbalance indices (Ix 2) were independently determined using the Cate-Nelson partionning procedure and are presented in Tab. 6. Sufficiency ranges of nutrient indices were computed as the square root of critical values as follows: ±2.13 for IN; ±1.30 for IP; ±1.9 for IK; ±1.1 for ICa; ±1.2 for IMg;±0.7 for IR5. The sum of the critical squared nutrient indices was 11.2, hence validating the critical CND_r2 obtained previously (11.5). For a chi-square distribution having six degrees of freedom, the CND_r2 of 11.2 correspond to a probability level (α) of 0.08.
An ANOVA was run to identify if nutrient treatments induced significant differences in the global nutrient imbalance index. The results of three simple mean comparisons are presented here: CND_r2 from nitrogen deficient potato plants (N15) was significantly different than CND_r2 from control plants (p = 0.0110). Potassium and magnesium deficient plants did not produce any significant effect on CND_r2 compared to control plants (p = 0.4055 and p = 0.4456, respectively). Hence, N was the driving variable for diagnosing nutrient imbalance.
The analyses were conducted on fluorescence and reflectance parameters (Table 1–2) taken at 15, 23, 30, and 37 DAE from experiment A in 2002, and at 37 and 44 DAE from experiment C in 2003. Five different discriminant analyses (each including CANDISC and DISCRIM procedures) were conducted on the same dataset, using different classes: 1) nutrient treatments considering three treatments classes (N15, N30 and N_over30) corresponding to the amount of N applied.; 2) leaf N content (g·kg-1) based on Hochmuth et al. (2004) and Walworth and Muniz (1993) and considering three classes (deficient, sufficient, excessive) depending on position (below, between, over) in comparison to its sufficient level included between 0.30 and 0.45 g·kg-1; 3) VN, divided into three classes (deficient, sufficient, excessive) based on its confidence interval (α=0.05); 4) CND_r2 considering two classes (balanced or imbalanced), depending on position (below or over) versus critical value of 11.2; and 5) IN including two classes (deficient, sufficient) depending on position (below or over) in comparison to critical value of -2.13. Although an excessive class could have been added, no plants in our dataset had an IN value exceeding 2.13.
The proportion of potato plants correctly classified to their specific class is presented in Tab. 7. The discriminant analysis conducted over IN classes provided the highest averaged classification percentage (94.9 %) followed by leaf nitrogen content (88.6 %) and CND_r2 (87.5 %). In general, the proportions of plants correctly classified according to discriminant analyses based on either fluorescence (84.5 %) or reflectance (84.2 %) parameters were similar. The combination of fluorescence and reflectance parameters slightly enhanced the treatments’ discrimination (86.8 %).
Tab. 7 Reclassification percentage using the cross-validation option for each discriminatory analysis.
These results support the development of a multivariate indicator able to correctly diagnose nutrient imbalance. As the classes of the nitrogen imbalance index are the most easily detected over our dataset, the indicator development will be based on the canonical discriminant variable calculated using reflectance and fluorescence parameters over IN classes.
Coefficients to compute the canonical variable across reflectance and fluorescence parameters for nitrogen imbalance (IN) detection are presented in Tab. 8. The sum of raw coefficients multiplied by the centered reflectance and fluorescence parameters produces a candisc value for each potato plant, as presented in equation .
As shown in Tab. 9 by the squared canonical correlation coefficient (0.315), the model using both reflectance and fluorescence parameters estimated the between-class variance more precisely. The Multivariate Indicator of Nitrogen Imbalance (MINI) was thus computed as a canonical variable across both reflectance and fluorescence parameters, using the raw coefficients presented in Tab. 8.
Tab. 9 Squared canonical correlation coefficients for canonical variable 1, using reflectance, fluorescence, and a combination of parameters from both methods.
The graphical distribution of IN values within the 2-D space of canonical variables 1 and 2 computed across reflectance and fluorescence parameters is presented in fig. 5. The important discrimination to be made relates to deficient IN, indicating higher N requirements. Sufficient IN values indicate adequate growth conditions. The MINI was developed using Can1: the deficient IN zone was delimited using the lower boundary from the confidence interval (α=0.001) of the Can1 averaged value over all deficient IN. The cutoff was set at Can 1 > 0.94 across reflectance and fluorescence parameters in order to minimize observed α and β.
fig. 5 Plot of the first two canonical variables for IN grouping using reflectance and fluorescence variables.
Tab. 10 presents the frequencies of classification using the MINI. The proportion of N-sufficient potato plants identified as N-sufficient by MINI is 90.5 %. The predictive value representing the probability that a plant be N-sufficient when it is declared as N-sufficient by MINI is 97.5 %. Even though the proportion of N-deficient plants represented only 6.9 % of the total number of observations, there were 69.2 % of N-deficient plants declared as N-deficient by our indicator.
The errors occurring when the MINI is computed are presented in Tab. 11. For the computation dataset (experiment A and C, n=313), the significance level (α) is 9.7 % and the power of the indicator (1-β ) is 76.9 %. Both errors are in an acceptable range for remote sensing (α < 0.20 and 1-β ≈ 80 %). For the validation dataset (experiment B, n=60), the α is 15.2 % and the power is 62.7 %. The power would be higher if the dataset included more samples.
Considering its current format, the MINI can discriminate N-sufficiency from N-deficiency in potato plants. It may also be important to detect N-excess in order to avoid fertilizer application on over-fertilized plants. Reflectance and fluorescence measurements can detect N-excess on plants (Railyan et al., 1990; Romanova et al., 1987) as well as the nitrogen imbalance index (IN > 2.13). To adapt the indicator to N-excess detection, additional data involving plants suffering from N-excess should be aggregated to the database and the N-excess zone be determined by setting the indicator threshold. By using kriging, for instance, a MINI map could be drawn over a complete field using data acquired only over a sampling grid, thus reducing acquisition time.
The objectives of our study were 1) to evaluate whether reflectance and fluorescence parameters can detect nutrient imbalance (CND_r2) in potato and 2) to develop an indicator of potato nitrogen imbalance combining reflectance and fluorescence parameters.
Using greenhouse grown plants and discriminant analysis, our results show that the detection of nutrient imbalance (CND_r2) by reflectance or fluorescence parameters was possible although similar, one to another. According to discriminant analyses, the proportion of plants correctly classified to their nutrient imbalance status has enhanced from 85.5 to 88.9 % when combining fluorescence and reflectance parameters using a canonical variable (Tab. 7). For nitrogen imbalance (IN) the canonical variable combining fluorescence and reflectance parameters provided a proportion of plants correctly classified of 96.6 %. Using this canonical variable it was possible to combine reflectance and fluorescence indices and still reflect their complementary nature and the between-class variance.
The Multivariate Indicator of Nitrogen Imbalance (MINI) was derived using the canonical variable across reflectance and fluorescence parameters. This indicator can correctly detect almost 70 % of the N-deficient plants and more than 90 % of the N-sufficient plants. Contrary to the standard methods of deficiency detection (foliar analyses), this indicator could allow a rapid data acquisition and handling, and provide deficiency detection within the time-window necessary for plant response to a recovery fertilization. It is a first step towards the automation of diagnosing nutrient deficiencies using reflectance and fluorescence parameters. As UV-exposed plants might have a different fluorescence response than greenhouse grown plants, the indicator should now be tested in the field for N diagnosis and site-specific N fertilization using a tractor-mounted instrument (Belzile et al., 2003). Future work should also include testing for growth stage specificity and particularly for the development of the N-excess indicator.
Adams, M., W. Norvell, J. Peverly, and W. Philpot. 1993. Fluorescence and Reflectance Characteristics of Manganese Deficient Soybean Leaves - Effects of Leaf Age and Choice of Leaflet. Plant Soil 156:235-238.
Aitchison, J. 1986. Statistical analysis of compositional data Chapman and Hall, New York.
Apostol, S., A.A. Viau, N. Tremblay, J.-M. Briantais, S. Prasher, L.-E. Parent, and I. Moya. 2003. Laser induced fluorescence signatures as a tool for remote monitoring of water and nitrogen stresses in plants. Canadian journal of remote sensing 29:57-65.
Beaufils, E.R. 1973. Diagnosis and recommendation integrated system (DRIS) Soil Science Bull., Vol. 1. University of Natal, S. Africa.
Bélanger, M.-C., A.A. Viau, G. Samson, and M. Chamberland. in press. Near-Field Reflectance and Fluorescence Measurements for Nutrient Deficiency Detections in Potato Crop (Solanum tuberosum, L.), In ASA-CSSA-SSSA, (ed.) 7th International conference on Precision Agriculture and other resources management. ASA-CSSA-SSSA, Minneapolis, Mn, July 24-28, 2004.
Bélanger, M.-C., A.A. Viau, G. Samson, and M. Chamberland. submitted. Comparison of reflectance and fluorescence spectroscopy for the detection of mineral deficiencies in potato plants. Canadian journal of remote sensing.
Belzile, C., M.-C. Bélanger, A.A. Viau, M. Chamberland, and S. Roy. 2003. An operational system for crop assessment, p. 244-252, In B. S. Bennedsen, et al., eds. Proceedings of Photonic East conference: Monitoring Food Safety, Agriculture, and Plant Health, Vol. 5271. SPIE, Providence, RI.
Bilger, W., M. Veit, L. Schreiber, and U. Schreiber. 1997. Measurement of leaf epidermal transmittance of UV radiation by chlorophyll fluorescence. Physiologia plantarum 101:754-763.
Brach, E.J., A.R. Mack, and G.T. St-Amour. 1981. Mobile field laboratory instrumentation to measure spectral characteristics of agricultural crops 626. Engineering and statistical research institute -Institut de recherche technique et statistique.
Cerovic, Z.G., G. Samson, F. Morales, N. Tremblay, and I. Moya. 1999. Ultraviolet-induced fluorescence for plant monitoring: present state and prospects. Agronomie 19:543-578.
Chappelle, E.W., J.E. McMurtrey III, F.M.J. Wood, and W.W. Newcomb. 1984. Laser-induced fluorescence of green plants. 2: LIF caused by nutrient deficiencies in corn. Applied Optics 23:139-142.
Cohen, J. 1988. Statistical power analysis for the behavioral sciences. 2nd ed. Lawrence Erlbaum Associates, inc., Hillsdale, New Jersey.
Corp, L.A., J.E. McMurthey III, E.M. Middleton, C.L. Mulchi, E.W. Chappelle, and C.S.T. Daughtry. 2003. Fluorescence sensing systems: In-vivo detection of biophysical variations in field corn due to nitrogen supply. Remote Sensing of Environment 86:470-479.
FAO. 2004. FAOStat [Online]. Available by United nations organisation http://faostat.fao.org/default.jsp (posted 2004-05-24; verified 2004-08-20).
Gamon, J.A., L. Serrano, and J.S. Surfus. 1997. The photochemical reflectance index: an optical indicator of photosynthetic radiation use efficiency across species, functional types, and nutrient levels. Oecologia 112:492-501.
Gitelson, A.A., and M.N. Merzlyak. 1994. Spectral reflectance changes associate with autumn senescence of Aesculus hippocastanum L. and Acer platanoides L. leaves. Spectral features and relation to chlorophyll estimation. Journal of Plant Physiology 143:286-292.
Guénette, C. 2003. Évaluation du potentiel de l'infrarouge spectral pour la détection hâtive du stress nutritifs chez les végétaux de grandes cultures. -Application à la culture de la pomme de terre. Mémoire de Maîtrise, Université Laval, Québec.
Haboudane, D., J.R. Miller, N. Tremblay, P.J. Zarco-Tejada, and L. Dextrase. 2002. Integrated narrow-band vegetation indices for prediction of crop chlorophyll content for application to precision agriculture. Remote Sensing of Environment 81:416-426.
Heisel, F., M. Sowinska, J.A. Miehé, M. Lang, and H.K. Lichtenthaler. 1996. Detection of nutrient deficiencies of maize by laser induced fluorescence imaging. Journal of Plant Physiology 148:622-631.
Hochmuth, G., D. Maynard, C. Vavrina, and E. Simonne. 2004. Plant tissue analysis and interpretation for vegetable crops in Florida [Online]. Available by University of Florida, IFAS extension http://edis.ifas.ufl.edu/EP081 (posted February 2004; verified 2004-08-27).
Irvine, J., A. Basinski, B. Baker, S. Jandciu, M. Paquette, J. Cairns, S. Connolly, R. Roberts, M. Gent, and P. Dorian. 1999. Depression and Risk of Sudden Cardiac Death After Acute Myocardial Infarction:Testing for the Confounding Effects of Fatigue. Psychosomatic Medecine 61:729-737.
Khiari, L., L.-E. Parent, and N. Tremblay. 2001a. The phosphorus nutrient diagnosis range for potato. Agronomy Journal 93:815-819.
Khiari, L., L.-E. Parent, and N. Tremblay. 2001c. Critical compositional nutrient indexes for sweet corn at early growth stage. Agronomy Journal 93:809-814.
Khiari, L., L.-E. Parent, and N. Tremblay. 2001b. Selecting the high-yield subpopulation for diagnosing nutrient imbalance in crops. Agronomy Journal 93:802-808.
McMurthey III, J.E., E.W. Chappelle, M.S. Kim, J.J. Meisinger, and L.A. Corp. 1994. Distinguishing Nitrogen fertilization levels in field corn (Zea mays L.) with actively induced fluorescence and passive reflectance measurements. Remote Sensing of Environment 47:36-44.
Mercure, S.-A., B. Daoust, and G. Samson. 2004. Causal relationship between growth inhibition, accumulation of phenolic metabolites, and changes of UV-induced fluorescence in nitrogen-deficient barley plants. Canadian journal of botany 82:815-821.
Merzlyak, M.N., A.A. Gitelson, O.B. Chivkunova, and Y. Rakitin. 1999. Non-destructive optical detection of pigment changes during leaf senescence and fruit ripening. Physiologia plantarum 106:135-141.
Méthy, M., A. Olioso, and L. Trabaud. 1994. Chlorophyll fluorescence as a tool for management of plant ressources. Remote Sensing of Environment 47:2-9.
Munson, R.D., and W.L. Nelson. 1990. Principles and practices in plant analysis, p. 359-387, In R. L. Westerman, ed. Soil testing and plant analysis, Vol. SSSA Book Ser. No. 3. SSSA, Madison, WI.
Nelson, L.A., and R.L. Anderson. 1977. Partitioning of soil test-crop response probability, p. 19-38, In M. Stelly, ed. Soil Testing: Correlating and interpreting the analytical results., Vol. 29. ASA, Madison, WI.
Parent, L.-E., A.N. Cambouris, and A. Muhawenimana. 1994. Multivariate diagnosis of nutrient imbalance in potato crops. Soil Science Society of America Journal 58:1432-1438.
Parent, L.E., and M. Dafir. 1992. A theoretical concept of compositional nutrient diagnosis. Journal of the American Society for Horticultural Science 117:239-242.
Parkinson, J.A., and S.E. Allen. 1975. A wet oxidation procedure suitable for the determination of nitrogen and mineral nutrients in biological material. Communications in Soil Science and Plant Analysis 6:1-11.
Penuelas, J., F. Baret, and I. Filella. 1995. Semi-empirical indices to assess carotenoids/chlorophyll a ratio from leaf spectral reflectance. Photosynthetica 31:221-230.
Railyan, V.Y., R.M. Korobov, O.I. Voinova, N.A. Zhiro, and N.G. Cheban. 1990. Studies on remote spectrometry for detecting stress symptoms in tomatoes. (In Russian, with English abstract). Izvestiya Akademii Nauk Moldavskoi SSR, Biologicheskikh i Khimicheskikh Nauk. 1:60-65.
Read, J.J., L. Tarpley, J.M. Mckinion, and K.R. Reddy. 2002. Narrow-waveband reflectance ratios for remote estimation of nitrogen status in cotton. Journal of environmental quality 31:1442-1452.
Romanova, A.K., L.G. Kuznetsova, E.V. Golovina, N.S. Novinchkova, I.F. Karpilova, and B.N. Ivanov. 1987. The excess of nitrate (nitrogen stress) and photosynthesis in higher plants. Proceedings of the Indian National Science Academy, B Biological Sciences. 53:505-512.
Ross, S.M. 1987. Introduction to probability and statistics for engineers and scientists. John Wiley & Sons, New York.
Schork, M.A., and R.D. Remington. 2000. Statistics with applications to the biological and health sciences. 3rd ed. Prentice Hall, Upper Saddle River, NJ.
Schuerger, A.C., G.A. Capelle, J.A. Di Benedetto, C. Mao, C.N. Thai, M.D. Evans, J.T. Richards, T.A. Blank, and E.C. Stryjewski. 2003. Comparaison of two hyperspectral imaging and two laser-induced fluorescence instruments for the detection of zinc stress and chlorophyll concentration in bahia grass (Paspalum natatum Flugge.). Remote Sensing of Environment 84:572-588.
Sims, D.A., and J.A. Gamon. 2002. Relationships between leaf pigment content and spectral reflectance across a wide range of species, leaf structures and developmental stages. Remote Sensing of Environment 81:337-354.
Subbash, N., and C.N. Mohanan. 1994. Laser-induced red chlorophyll fluorescence signatures as nutrient stress indicator in rice plants. Remote Sensing of Environment 47:45-50.
Tartachnyk, I., and I. Rademacher. 2003. Estimation of Nitrogen Deficiency of Sugar beet and Wheat using Parameters of Laser Induced and Pulse Amplitude Modulated Chlorophyll Fluorescence. Journal of applied botany 77:61-67.
Tukaki, J.L., and R.L. Mahler. 1990. Evaluation of nutrient solution phosphorus concentration on potato plantlet tuber production under greenhouse conditions. Journal of plant nutrition 13:149-168.
Vouillot, M., P. Huet, and P. Boissard. 1998. Early detection of N deficiency in a wheat crop using physiological and radiometric methods. Agronomie 18:117 - 130.
Walworth, J.L., and J.E. Muniz. 1993. A compendium of tissue nutrient concentrations for field-grown potatoes. American potato journal 70:579-597.
Zarco-Tejada, P.J., J.R. Miller, G.H. Mohammed, T.L. Noland, and P.H. Sampson. 2002. Vegetation stress detection through Chlorophyll a + b estimation and fluorescence effects on hyperspectral imagery. Journal of environmental quality 31:1433-1441.
Zellweger Analytic inc. Determination of total Kjedahl nitrogen in soils and plants by flow injection analysis Quickchem Method 13-107-06-02-D. Lachat Instrument Division, Milwaukee, WI.
Zhao, D., K.R. Reddy, V.G. Kakani, J.J. Read, and G.A. Carter. 2003. Corn (Zea mays L.) growth, leaf pigment concentration, photosynthesis and leaf hyperspectral reflectance properties as affected by nitrogen supply. Plant and Soil 257:205-217.
© Marie-Christine Bélanger, 2005