A Soil Plant Analysis Development (SPAD) chlorophyll meter can be used to screen for leaf nitrogen (N) concentration in breeding programs. Lentil (
The majority of leaf N is accumulated in the chloroplast, where photosynthesis takes place, resulting in a strong association between plant photosynthesis and leaf N status [
Despite its extensive application, association of SCMRleaf
With an assumption of independency of input variables, SCMR (independent variable) is generally fitted against leaf
Nonlinear regression continually adjusts the estimated values of the parameters and improves the fit to a minimum satisfactory level [
Lentil (
Lentil cultivars were grown in Saskatoon (52.05° N and 106.43° W) and Indian Head (50.55° N and 103.65° W), Saskatchewan. In Saskatoon, three N fertility treatments of 50 kg N ha^{−1}, 5.6 kg ha^{−1} granular rhizobia (Nodulator, Becker Underwood, Saskatoon, SK), and a nontreated control were applied on eight lentil cultivars in 2006 and 2007 (N fertility trial). This trial was conducted in two different fields in 2006 and 2007. Compared to 2006, the 2007 field in the N fertility trial had low soil available N and no recorded history of legume crop cultivation and rhizobia inoculation. In the N fertility trial, CDC Greenland, CDC Plato and CDC Sedley (latematuring group), CDC Milestone and CDC Viceroy (mediummaturing group), and CDC Blaze, CDC Red Rider, and CDC Rouleau (earlymaturing group) were grown in both years [
Leaf chlorophyll content was estimated using a SPAD Chlorophyll Meter (Model 502 Konica Minolta Sensing, Inc., Japan) at three growth stages of vegetative (up to node 12), firstpod (at least one pod per plant), and latepod (when the canopy started turning yellow). To eliminate daily variations of light quality and leaf starch concentration, SCMR readings were limited to the uppermost leaves during 10:00 to 12:00 h of day. Three plants per plot were randomly selected and SCMR was recorded from the three fully expanded uppermost leaves of each plant. The average of nine SCMR values in each plot was considered as the plot SCMR value. The leaves were immediately detached and transferred on ice into a refrigerator for further measurements the next day. In the laboratory, leaf surface area was measured using a leaf area meter (Model LI3100C, LiCor, Lincoln, NE), and leaves were dried at 60°C for 24 hrs, weighed, and ground. Leaf N concentration was measurement by the combustion method, using a Leco carbonnitrogen determinator (LECO CNS 2000, St. Joseph, MI, USA). Specific leaf weight (SLW), which represents leaf thickness, was the ratio of leaf dry weight (g) to leaf area (m^{−2}), SLN was the ratio of g leaf N (leaf weight × LNC) to leaf area (m^{−2}), and adjusted SCMR for leaf thickness was the ratio of SCMR to SLW. In addition, grain yield and N content of the entire plant biomass (from both trials) and average N content of leaf (referred to entire leaf biomass), stem, and pod from 5 plant plot^{−1} at the three given growth stages (from the N fertility trial) were available.
In each trial, the three leaf characteristics (LNC, SLW, and SLN) were analyzed for the effects of treatment and cultivar. In the analysis of variance, the main factor was N fertility treatment (in the N fertility trial) and notill duration (in the NT trial), and the subfactor was cultivar. Data were analyzed as a yearcombined RCBD for each trialgrowth stage, with year, treatment, cultivar, and their interactions as fixed variables and block and interaction of block with the fixed factors as random variables [
Following the analysis of variance, the leaf characteristics data from two trials, three growth stages, and two years were pooled to compute the correlation of SCMR and adjusted SCMR with LNC, using the CORR procedure of SAS. This data set consisted of 740 data points for each of the leaf characteristics (LNC, SCMR, SLW, and SLN). The pooled data set was randomly divided into two groups, a training set of 630 data points (85% of the data) and a test set of 110 data points (15% of the data), to develop linear and nonlinear models by the GLM procedure in SAS and the Newrb function in Matlab, respectively. The training set was used in model development and the test data to validate the models. The three growth stages of vegetative, firstpod, and latepod were arbitrary considered 1, 2, and 3, respectively.
For the nonlinear approach, we tested an RBF neural networks model. This model is linear combinations of radial basis that produces linear outputs based on nonlinear inputs. Using RBF requires specification of the number of hidden unit activation function, the number of processing units, a criterion for modeling a given task, and a training algorithm for finding the parameters of the network. Weight of the model is found through the training process, where network parameters are optimized to fit the network outputs to the given inputs [
Averaged over the years, trials, treatments, and cultivars, lentil LNC decreased from 4.5% at vegetative to 3.8% at firstpod and 2.7% at latepod growth stage (Table
Average leaf N concentration (LNC), specific leaf N (SLN), and specific leaf weight (SLW) of lentil at three growth stages of vegetative growth (VG), firstpod (FP), and latepod (LP) under three N fertility treatments (top) and two NT duration treatments (bottom) in 2006 and 2007.
Year  Treatments  LNC  SLW  SLN  

(%)  (g DW m^{−2} leaf)  (g N m^{−2} leaf)  
VG  FP  LP  VG  FP  LP  VG  FP  LP  
Fertility study  
2006  Control  4.5a  4.0c  3.8a  41b  50b  50bc  1.6b  1.6d  1.9a 
N fertilizer  4.6a  4.3b  4.1a  39b  47b  48c  1.6b  1.8c  1.9a  
Inoculant  4.5a  4.1c  3.9a  42b  49b  46c  1.6b  1.7c  1.8a  











2007  Control  3.6c  4.7a  3.1b  45a  56a  56a  1.6b  2.7a  1.7ab 
N fertilizer  4.0b  4.0c  2.8bc  41b  55a  52ab  1.6b  2.4b  1.4b  
Inoculant  4.2b  4.5a  2.9c  44a  56a  49b  1.8a  2.6a  1.4b  













NT study  
2006  LT  4.7b  3.6ab  1.8b  41ab  46a  56a  1.9b  1.7b  1.1ab 
ST  4.9a  4.0a  2.3a  45a  48a  51ab  2.3a  1.8b  1.3a  











2007  LT  4.5b  3.4b  1.8b  39b  47a  49b  1.7c  2.1a  1.0a 
ST  4.4b  3.0c  1.9ab  38b  43b  54ab  1.7c  2.2a  1.1a  










Means followed by different small letters within columns within each study indicate significant effects of the fertility treatments (top) and the NT treatments (bottom) in the two years of the studies (
Means followed by different capital letters within columns within each study indicate significant effects of the year in each study (
The leaf characteristics differed between 2006 and 2007 in the N fertility trial (Table
Correlation coefficients of LNC and N concentration of the entire leaf biomass with lentil performance in the N fertility trial in three stages of vegetative (VG), first pod (FP), and latepod (LP).
Yield  HI  DTM  %Ndfa  Above ground biomass  Leaf N  Stem  Pod  

Year^{†}  Stage  g m^{−2}  %  days  %  DM  N%  mg N plant^{−1}  mg N plant^{−1}  N%  mg N plant^{−1}  N%  mg N plant^{−1}  
LNC (%)  2 years  VG  0.49  —  0.45  0.29  —  0.52  —  —  —  —  na^{‡}  na 
FP  —  —  —  —  —  —  —  —  —  —  —  —  
LP  0.24  —  0.35  —  0.38  —  0.20  0.39  0.47  0.32  0.23  —  
2006  VG  —  —  0.43  —  0.29  —  0.33  0.37  —  —  na  na  
FP  —  —  —  —  —  0.37  —  —  —  —  —  —  
LP  0.25  −0.49  0.35  −0.30  0.61  0.58  0.25  0.45  0.49  0.39  —  —  
2007  VG  0.48  0.22  —  —  —  0.46  —  —  0.26  —  na  na  
FP  —  —  —  0.29  −0.39  0.32  —  —  0.30  —  —  —  
LP  —  0.25  0.29  0.28  −0.22  —  —  —  0.32  —  —  —  


Total leaf biomass N%  2 years  VG  —  —  —  —  —  0.28  0.42  0.42  0.64  0.29  na  na 
FP  0.37  —  0.43  0.31  −0.23  0.37  —  0.30  0.78  0.21  0.65  —  
LP  —  —  0.22  —  —  —  —  0.47  0.53  0.27  0.30  —  
2006  VG  —  −0.36  —  —  —  —  0.63  0.63  0.37  0.40  na  na  
FP  0.30  −0.31  0.32  0.41  —  —  0.23  0.55  0.66  0.45  0.34  −0.40  
LP  —  −0.42  —  −0.44  0.41  0.55  —  0.54  0.67  0.49  0.38  —  
2007  VG  0.33  —  —  —  —  0.81  0.36  0.38  0.83  —  na  na  
FP  —  —  0.32  0.31  —  0.48  0.33  0.45  0.85  0.33  0.83  —  
LP  —  —  0.45  0.35  —  —  —  0.40  0.58  —  —  −0.27 
^{‡}Only significant correlation coefficients are presented (
In the NT trial, LNC was greater in 2006 than in 2007 at two final growth stages, SLW was similar during the entire seasons of both years, and SLN was greater in 2006 than in 2007 at vegetative, less in 2006 than in 2007 at firstpod, and similar in both years at latepod. In this trial, lentil yield was similar between the years, but lentil accumulated more biomass and N due to more rainfall in 2007 than in the drier year of 2006 (Table
Averaged over the treatments and years, LNC in the N fertility trial, where one third of the plots received 50 kg N fertilizer ha^{−1} at seeding, was 0.8% (at firstpod) and 1.5% (at latepod) greater than in NT (Table
In the NT trial, where the spring soil N was greater in the long than the shortterm NT in both years, response of lentil yield to the NT duration treatment was limited to 2006 only (Table
Variation of the cultivars for the three leaf characteristics appeared in the N fertility trial in 2007, only. In this case, cultivars CDC Red Rider and CDC Plato had the greatest and CDC Sedley had the smallest LNC at the vegetative, and CDC Red Rider and CDC Viceroy had greater LNC than the other cultivars at the firstpod (Figure
Variations of eight lentil cultivars for specific leaf weight (SLW), specific leaf N (SLN), and leaf N concentration (LNC) at three growth stages of vegetative (VG), firstpod (FP), and latepod (LP) stages in the N fertility trial in 2007. Error bars are standard deviation.
Association of leaf N with lentil performance was calculated for the N fertility trial only. In this experiment, correlation of LNC with N concentration of the entire leaf biomass was strongly positive (
As a result of these variations, correlations of plant growth parameters and grain yield with N concentration of the top leaves (LNC) and N concentration of the entire leaf biomass were not consistent. For example, both LNC and entire leaf biomass N concentration were positively correlated to plant N concentration at latepod stage in 2006 and at vegetative and firstpod stages in 2007 (Table
Averaged over the treatments and cultivars, SCMR values were similar between the two trials (N fertility and NT trials) at the vegetative growth stage (30), greater in the N fertility trial (37) than the NT trial (26) at firstpod growth stage, and greater in the N fertility trial in 2006 (40) than in the other yearstrials at the latepod growth stage (24). In response to the N fertility treatments, inoculated lentil had greater SCMR than lentil in the control treatment at the latepod stage in 2006 (Table
Because of significant treatment and year effects in the N fertility trial compared to the NT trial, correlations of SCMR and adjusted SCMR with LNC, plant N, and SLN were conducted for the N fertility trial only (Table
Correlation coefficients of SCMR and adjusted SCMR (SCMR/SLW) with leaf N concentration (LNC), plant N concentration (plant N), and specific leaf N (SLN) for each growth stage of vegetative, firstpod, and latepod and for the pooled data over the growth stages.
Growth stage  Variable  SCMR  LNC  Plant N%  SLN 

Vegetative  SCMR  1.00  ns^{†}  ns  0.26 
Adjusted SCMR  0.58  0.35  0.36  −0.38  
Firstpod  SCMR  1.00  0.35  −0.19  0.37 
Adjusted SCMR  0.54  0.42  0.26  −0.37  
Latepod  SCMR  1.00  0.63  −0.43  0.65 
Adjusted SCMR  0.84  0.77  −0.40  0.40  


Pooled  SCMR  1.00  0.50  ns  0.51 
Adjusted SCMR  0.70  0.75  0.40  ns 
Equations are developed by linear regression to estimate leaf N concentration, using different combinations of SPAD chlorophyll meter reading (SCMR) values, growth stage (Stage), and specific leaf weight (SLW).
Linear model (independent variable)  Coefficient of determination ( 


0.49 

0.54 

0.64 

0.69 
Four linear models were developed to estimate LNC based on various combinations of SCMR, growth stage, and leaf thickness as independent variables using the pooled data set from both trials (Table
In both the linear and nonlinear models, SLW and, to some extent, growth stage had strong influence on the accuracy of LNC prediction models (Figure
Association of estimated leaf N concentration (LNC) by linear regression (left) and nonlinear redial basis function (RBF) neural networks models (right) against actual LNC values. Data from two trials, two years, and three growth stages. Variables in the right side represent the combination of independent variables that generated each model.
For model validation (reliability), the models were fed by the test data set and then correlation coefficient of the models outcomes was calculated with actual LNC in the test data set. These correlation coefficients represent the models reliability for any new data. Results showed that reliability of the nonlinear models was always greater than the linear models. Correlation coefficients of the estimated LNC and actual LNC from the test set for the linear and nonlinear models were, respectively, 0.41 and 0.60 (
Leaf N concentration represented overall variations of plant N and grain yield of lentil due to the treatments and experimental conditions. In the fertility trial, varied plant N and grain yield between 2006 and 2007 and significant treatment effects in 2007 were all in agreement with the LNC variations in the given situations (Table
Despite similar responses of LNC and lentil performance under the conditions of the experiment (Table
The results demonstrated that leaf position, plant growth stage, and number of sampled leaves (top leaves versus the entire leaf biomass) must be considered when LNC is used to estimate plant N status. Although plant age generally affects LNC amongst the field crops [
The treatments affected SCMR in both trials, but the treatments effects on SCMR were not always similar to their effects on LNC (Table
Adjusted SCMR (SCMR/SLW) showed a stronger correlation with LNC than SCMR (Table
Lentil LNC and SCMR variations due to the experiment were in agreement with the plant N and grain yield changes. This similarity suggests that lentil SCMR from any field can be compared with a well fertilized plot in the same field to diagnose lentil N status. In two cases of treatment effects on plant N in the NT trial in 2006 and in the N fertility trial in 2007, plant N deficiency due to the treatments was detected from the SCMR values (Table
Leaf nitrogen concentration
Crop Development Centre, University of Saskatchewan
No tillage
Soil Plant Analysis Development
SPAD Chlorophyll Meter reading
Specific leaf weight
Specific leaf nitrogen.
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors would like to remember significant contribution of Dr. Guy Lafond, who passed away on October 2013. Part of the experiment was conducted at the Indian Head Research Farm, where Guy and his team planted and maintained the experimental plots and helped in data collection. Funding for this project was provided by NSERC and Saskatchewan pulse growers. Plot maintenance was done by the Crop Development Center field crew and by the Agriculture Canada crew at Indian Head Research Farm. Substantial help in data collection and sample processing by Janet Pritchard is appreciated.