| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a CSIRO Land and Water, Private Bag, P.O., Wembley 6014, WA, Australia
b USDA Plant, Soil & Nutrition Lab., Tower Rd., Ithaca, NY 14853 USA
c Dep. of Soil, Crop & Atmospheric Sciences, Bradfield Hall, Cornell Univ., Ithaca, NY 14853 USA
matthew.adams{at}per.clw.csiro.au
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
Abbreviations: Chlt, total chlorophyll (a + b) • EDTA, ethylenediaminetetraacetate • F, chlorophyll fluorescence (subscripts m, o, v represent maximum, minimal, variable fluorescence • 5 min, fluorescence after 5 min illumination) • ICP-ES, inductively coupled argon plasma emission spectrometry • MES, 2-(N-morpholino)-ethanesulfonic acid • RMSE, root mean square error • YI, yellowness index
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
We hypothesize that deficiencies of micronutrient metals that are required in the biosynthetic pathway of chlorophyll or required for electron transport will be most easily detectable by spectral changes in fluorescence or visible and near-infrared reflectance. Of the six micronutrient metals known to be required for all higher plants (Welch, 1995); Mn, Fe, and Cu are the most likely to cause spectral changes. Like Mn, Fe is a good candidate because it is involved directly in electron transport reactions and is essential for the synthesis of chlorophyll (Pushnik et al., 1984; Spiller et al., 1982). Copper, too, is also directly involved in electron transport reactions as it is an essential component of plastocyanin (Droppa and Horváth, 1990).
The remaining micronutrient metals; Zn, Mo, Co, and Ni are not as promising candidates for stress detection or establishment of critical levels by spectral means. In deficiencies of Zn, it is likely that the development of leaf symptoms is influenced primarily by light intensity rather than leaf Zn concentration (Marschner and Cakmak, 1989). Therefore, there may not be a consistent relationship between leaf Zn deficiency and chlorophyll or spectral measures. Molybdenum, Co, and Ni are involved in N metabolism. Their effect on chlorophyll concentration or function would most likely be indirect and related to inhibited protein synthesis. Close linkage to spectral characteristics of leaves seems unlikely.
We compared selected spectral characteristics with nutrient concentrations of trifoliolate leaves of Bragg soybean. Plants were grown under varying degrees of deficiency of Mn, Fe, Cu, or Zn. Although not as promising as Mn, Fe, and Cu, Zn was included in the study because it is easy to induce Zn deficiency by the methods used to grow the plants and because Zn deficiency occurs in soybean in the field, which could interfere with detection of Mn, Fe, or Cu deficiency. The youngest fully mature leaf blade was chosen as the index leaf, as is common for diagnosis of deficiencies in soybean and other crops by plant tissue analysis (see Reuter and Robinson, 1997, for a fuller treatment of concepts and principles of plant analysis). Our results are compared with critical tissue concentrations measured by others.
|
Materials and methods |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
resistivity). Seed imbibition represented Day 0 of an experiment. After 24 h, seeds were transferred to moist filter paper and kept in low light at room temperature for 48 h to allow the radicles to extend. On Day 3, seedlings were transferred to cups with the radicles protruding through a plastic screen into a 5-L black polyethylene pot containing the basal nutrient solution described below. Seedlings were planted one seedling per cup, and four cups were inserted into the lid of each pot.
The plants were grown in an environmentally controlled growth chamber under a mixture of fluorescent and incandescent lighting on a 16 h/8 h day/night cycle. Light intensity at the surface of plant tops varied among experiments from 550 and 700 µmol m-2 s-1, depending on the age of the bulbs. The nutrient solutions were aerated, and air temperature was maintained at 24°C during the day and 20°C at night.
Nutrient Solution
The basal nutrient solution contained (µM): K, 2200; Ca, 1000; NH4, 50; NO3, 3100; P, 10; Mg, 500; S, 500; Cl, 50; B, 12.5; Mo, 0.1; and the pH buffer MES [2-(N-morpholino)-ethanesulfonic acid] at 2000 µM, adjusted to pH 6.1.
The micronutrient metals Fe, Zn, Cu, Mn, Ni, and Co were added on Day 8 as their EDTA (ethylenediaminetetraacetate) chelates. Solutions providing adequate levels of micronutrient metals to soybean contained at least 20 µM Fe, 5 µM Zn, 3 µM Cu, 0.3 µM Mn, 0.1 µM Ni, and 0.1 µM Co. Each of the solutions contained 20 µM EDTA in excess of that needed to chelate the six micronutrient metals. The excess chelate served to buffer the activities of all the micronutrient cations, and to depress the activity of any contaminant metals in solutions (Parker et al., 1995; Norvell, 1991). The initial concentrations of Ni and Co were the same in all experiments. Concentrations of Mn, Fe, Zn, or Cu varied from 0.01 to 100 µM, depending on the metal being studied. GEOCHEM-PC (Parker et al., 1995) was used to estimate free metal concentrations in solution. Total metal concentrations and calculated free metal concentrations are tabulated by treatment for each of the experiments in Table 1 . There were three replicate pots per treatment.
|
Harvest and Nutrient Analysis
One plant from each pot was harvested at 20, 23, and 25 d after seed imbibition. Just prior to harvest, a small (1.71 cm diam.) leaf disk was cut from the middle leaflet of the youngest fully mature leaf at a point on the center-line, roughly three-quarters of the way from the base to the tip. These disks were placed into light-proof cups for use in the fluorescence measurement and chlorophyll analysis described below. The remaining leaf tissue was dry-ashed and analyzed for nutrient content via inductively coupled argon plasma emission spectrometry (ICP-ES) (except ICP mass spectrometry for Cu from leaves of plants grown in the low Cu treatments).
Leaf Reflectance
The reflectance spectra between wavelengths of 0.4 and 1.1 µm were taken from attached leaves prior to harvest, using a Spectron SE-590 spectroradiometer (Spectron Engineering,1
Denver, CO), which has a bandwidth of 0.008 µm. The raw spectra were referenced against an 18% Kodak Gray Card. Reflectance of the center leaflet of the youngest fully mature trifoliolate from each replicate pot was measured without removing the leaf from the plant.
The leaf to be measured was laid on black aluminum foil (less than 2% reflectance in the range of wavelengths measured by the Spectron) mounted 9 cm away from the optics of the spectroradiometer. The field of view of the spectroradiometer at 9 cm was 4.41 cm2. The reflectance spectra were downloaded to a PC-compatible computer immediately after measurement, and smoothed with a five-point moving average calculated as the mean of the reflectance value of the central diode and the reflectance values of the two diodes above and below. A 5 x 1 moving average filter was chosen because it was narrow enough to minimize damping of features of possible interest in the spectra and wide enough to remove a good deal of the noise, particularly in the infrared region of the spectrum.
Four reflectance measures were derived from the reflectance spectrum: R750/R550 (maximum reflectance near 0.750 µm/maximum reflectance near 0.550 µm); R750/R650 (maximum reflectance near 0.750 µm/minimum reflectance near 0.650 µm); the normalized difference vegetation index (NDVI), (R750 - R650)/(R750 + R650); and the yellowness index (YI), defined below. The first three of these, R750/R550, R750/R650, and NDVI, are commonly used as measures of vegetative cover or chlorophyll (Tucker, 1979).
Yellowness index is a new spectral measure that estimates the degree of leaf chlorosis from the concavity–convexity of the reflectance spectrum at a wavelength near the midpoint between the maximum at 0.550 µm and the minimum at 0.670 µm. The mathematical description of YI is the finite difference approximation of the second derivative of the reflectance spectrum between 0.570 and 0.670 µm. YI was calculated as (Adams et al., 1999)
 | (1) |
0 is the central wavelength, -1 and +1 are the lower wavelength and higher wavelength, and 
is the spectral distance between wavelengths
(Philpot, 1991). For these experiments, 0 was centered at 0.624 µm, -1 was centered at 0.580 µm, and +1 was centered at 0.668 µm. Thus, was 0.044 µm. The magnitude of YI is a function of the units used for . When micrometers were used in the denominator of the YI equation, it was convenient to multiply the raw YI values by 10-1 to obtain values that varied between -5 and 20.
Leaf Fluorescence Induction
A Heinz–Walz PAM 101/103 fluorimeter (Heinz Walz GmbH, Effeltrich, Germany) was used to determine Fo (minimal fluorescence), Fm (maximum yield fluorescence), and F5min (fluorescence yield after 5 min illumination) (see the review of Karukstis, 1991, for further detail on fluorescence induction). For the measurement of Fo and Fm, a pulse width of 800 ms was used for illumination of the sample. The light source used to generate Fm was a 1000 W tungsten–halogen source. Intensity of the light source on the sample was in excess of 5000 µmol m-2 s-1. The light source used for measuring F5min was a 100 W tungsten–halogen source with a light intensity of approximately 625 µmol m-2 s-1. Fo was measured with a 1.6 kHz measuring beam. Fm and F5min were measured with a 100 kHz measuring beam. The instrument measured fluorescence from a circular area 1 cm in diameter.
The leaf disks, cut from each trifoliolate just prior to harvesting the plant, were placed in a film cup with a moist piece of tissue paper for approximately 45 min prior to measurement of Fo and Fm to allow dark adaptation. Fo was measured as the average of three 800-ms pulses in which no actinic light was applied to the leaf disk. After this, Fm was measured similarly, except that saturating light was applied to the leaf disk. There was a period of 60 s between saturating pulses.
Roughly 120 s after Fm was measured, longer term fluorescence was measured on selected leaf disks (generally leaf disks from plants grown in a low, marginal, and adequate treatment were selected, so fewer leaf disks were measured for F5min than the other fluorescence quantities). Fluorescence was sampled every 0.5 s for a period of 300 s (5 min) during this time period. At the end of 300 s, F5min was measured; Fv, (Fm - Fo), and the ratios Fo/Fv and Fo/F5min were calculated.
Chlorophyll Determination
Total chlorophyll (Chlt) was determined by methods modified from Moran and Porath (1980) and Inskeep and Bloom (1985) using dimethylformamide (DMF) as the extracting solvent.
Fitting Nutrient Response Curves
The graphical method of Cate and Nelson (1965) was used to fit dry matter yield and spectral measures to leaf metal concentrations. The graphical method is appropriate for the small data sets in this study. However, in addition to the graphical method the dry matter yield was fitted to leaf metal concentrations by least squares fit to a nonrectangular hyperbolic function:
 | (2) |
We found the more common Mitscherlich equation fit the data poorly due to the slope–plateau trend in the data, while the nonrectangular hyperbolic function fit the data well. The nonrectangular hyperbolic function is a three parameter equation, like the Mitscherlich equation. See Thornley and Johnson (1990) for additional information on the properties of the nonrectangular hyperbolic function and its utility.
|
Results |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
Detailed responses in yield and plant composition to the different metal treatments in these chelator-buffered solutions will be presented elsewhere (manuscript in preparation). However, the relationship between shoot yield and metal concentration in the index leaf on Day 23 is shown here (Fig. 1) to provide an estimate of the concentration of each metal required for normal growth. These critical concentrations were estimated from the results by the graphical method of Cate and Nelson (1965) as 9 to 10 mg kg-1 for Mn, 12 to 18 mg kg-1 for Zn, 1.0 to 1.5 mg kg-1 for Cu, and 40 to 50 mg kg-1 for Fe. Estimates of critical concentrations based on a least squares fit of a nonrectangular hyperbolic function of dry matter yield to leaf metal concentration were 8.9 mg kg-1 for Mn, 14.2 mg kg-1 for Zn, 1.5 mg kg-1 for Cu, and 49.7 mg kg-1 for Fe (Table 2) .
|
|
|
Responses of Chlt, YI, Fo/Fv, and Fo/F5min to leaf micronutrient metal concentrations for index leaves harvested on Day 23 of each experiment are shown in Fig. 3a–3d , respectively. Responses of the other reflectance measures are not shown because of their close correlation to chlorophyll and YI. Results for index leaves harvested on other days were similar.
|
The yellowness index (Fig. 3c) responded strongly to a deficiency of Mn, had a moderate response to deficiencies of Zn and Fe, and had no response to a deficiency of Cu.
Similar to YI, the Fo/Fv ratio (Fig. 3b) increased markedly in response to Mn concentrations less than 8 to 10 mg kg-1 dry weight. There was a smaller but abrupt increase in the Fo/Fv ratio in response to deficiencies of Zn at approximately 10 to 12 mg kg-1, but the increase occurred only in leaves displaying moderate to severe deficiency. Fo/Fv increased slightly, but more gradually, when Fe concentrations were less than 35 mg kg-1. There was little, if any, response of Fo/Fv to low Cu concentrations in the index leaves on Day 23, even though the index leaves from the lowest two Cu treatments appeared slightly pale green visually.
Fo/F5min (Fig. 3d) increased slightly in response to deficiencies of Mn, Zn, or Fe; and decreased in response to deficiencies of Cu. Fewer points are present on the figure relative to the other figures, because F5min could be measured on only two or three treatments due to the long data acquisition time for F5min.
Critical concentration ranges for index leaves derived from leaf chlorophyll and spectral measures from 3- to 4-wk-old plants were estimated by the graphical method of Cate and Nelson (1965) from enlarged versions of Fig. 2 and Fig. 3a–3d. The critical concentration ranges estimated were 9 to 10 mg kg-1 for Mn, 12 to 18 mg kg-1 for Zn, 45 to 50 mg kg-1 for Fe, and 1.0 to 1.5 mg kg-1 for Cu.
|
Discussion |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
The increases of Fo/Fv in response to Zn and Fe deficiencies in these experiments were due primarily to decreases in Fv alone, except in severely stressed leaves in which Fo increased slightly (data not shown). The changes in Fo/Fv for deficiencies of Zn and Fe were of lesser magnitude than for Mn, and occurred only in leaves that were severely stressed at time of sampling, suggesting that Fo/Fv is a specific and sensitive indicator of marginal Mn deficiency.
The response of Fo/F5min to Cu deficiency in these experiments is also consistent with previous experiments (Kriedemann and Anderson, 1988; Casimiro et al., 1990). This ratio decreased due to increased F5min, a change that is thought to be caused by thylakoid disorganization and the lack of development of a pH gradient across the thylakoid membrane in Cu-deficient chloroplasts (Casimiro et al., 1990). Further, Cu deficiency was the only deficiency to result in a decrease of Fo/F5min, suggesting that this measure is a sensitive and specific indicator of Cu deficiency. However, because the long term fluorescence induction kinetic is sensitive to many factors such as ambient light intensity (Dau, 1994), it is important to compare plants suspected of Cu deficiency with plants known to be Cu adequate in field situations. This may pose a serious limitation to the use of Fo/F5min, but requires further work to determine if this is the case.
Regardless of the metal deficiency imposed in these experiments, there was good agreement between the visual color of the leaf, total chlorophyll concentration, and the corresponding YI when leaves were chlorotic. Yellowness index was sensitive to total chlorophyll concentrations less than about 25 µg cm-2, but was insensitive to total chlorophyll concentrations greater than about 25 µg cm-2. This effect is related to the lack of change in the shape of the reflectance spectra between 0.57 and 0.65 µm when total chlorophyll concentrations are greater than about 25 µg cm-2 (Adams et al., 1999).
R750/R650 and NDVI are measures of the red-edge feature, the steep rise in reflectance of vegetation between 0.65 and 0.75 µm, and are correlated with total chlorophyll concentration (Tucker, 1979). They are also two of the most common reflectance measures used in remote sensing of vegetation. NDVI is more sensitive than R750/R650 at lower chlorophyll concentrations, while R750/R650 is more sensitive at higher chlorophyll concentrations (e.g., Fig. 2). This makes R750/R650 somewhat less attractive as a measure for detecting deficiencies, because R750/R650 does not plateau when a plant is adequately supplied with nutrients, while NDVI does plateau.
Comparison of Critical Concentrations
Prior studies in nutrient solution, greenhouse, and field have estimated critical concentrations of approximately 10 mg kg-1 for Mn (Ohki, 1976; Reuter and Robinson, 1997), 50 mg kg-1 for Fe (Brown and Jones, 1977), 15 mg kg-1 for Zn (Ohki, 1977), and 4 mg kg-1 for Cu (Makarim and Cox, 1983) in the index leaf for dry matter yield in Bragg soybean. Our estimates of critical concentrations, derived from spectral characteristics from index leaves, suggest similar critical values for Mn, Zn, and Fe but slightly lower values for Cu (1.0–1.5 mg kg-1).
These results are, in part, related to the degree of expression of symptoms versus sensitivity of symptoms to metal availability. In the case of Fe and Mn, which are required for chlorophyll synthesis and Photosystem II function, respectively, slight decreases in metal concentration below that required for these functions result in a rapid and large response in the spectral measures that are sensitive to these functions. Since these functions are essential for plant growth, plant growth is affected at the same time. Therefore, we find good agreement of critical concentrations obtained by tissue analyses and spectral measurements.
In the case of Cu, however, spectral measures underestimated critical concentrations developed from tissue analyses. This is most likely caused by the greater degree of deficiency needed to affect the plant physiology sampled by these spectral measures. For example, Fo/F5min did not decline significantly until leaf Cu concentrations fell below approximately 2.0 mg kg-1, and other spectral measures were insensitive. A somewhat similar effect was noted in the case of Zn.
Practicality of the Method
Although equipment limitations and the need to measure chlorophyll on the same tissue from which fluorescence was measured resulted in destructive sampling of the leaf tissue, commercial instruments exist that are capable of measuring reflectance spectra and fluorescence induction simultaneously (e.g., VIRAF II, described in Buschmann et al., 1994).
One concern is whether the method will be more generally applicable for plant nutrients such as N, P, and K. Validation of this method with other nutrients awaits further work. There is the opportunity, however, to develop other spectral measures that are more sensitive to the physiological processes that are altered by nutrient stresses. For example, modifications to the fluorimeter used in these experiments could be implemented to monitor changes in absorbance at 0.830 µm, the peak absorption of the Photosystem I cation-radical, P700+. Differences in absorbance at 0.830 µm could be used to accurately and specifically detect Fe deficiency, because Fe deficiency impairs assembly of functional Photosystem I complexes (Pushnik and Miller, 1989). This would be a significant advantage, because the generalized chlorosis response to Fe deficiency is difficult to distinguish from deficiencies of other nutrients. This is an useful approach for future work.
Perspective
Figures associated with economic losses due to micronutrient deficiencies are difficult to locate, because of the predominance of macronutrient nutrition over micronutrient nutrition. However, Chaney (1985) estimated a yield loss of 135000 t of grain per year for the potentially Fe-deficient soils in the Clarion–Nicollet–Webster soil association in Iowa, particularly in Calciaquolls. A great deal of research effort has been applied to breeding and screening for more Fe-efficient soybean varieties to address this problem (Chaney, 1985).
Scott and Aldrich (1983) have reported Mn deficiency as the most widespread micronutrient deficiency in the United States with Fe deficiency as the second most, suggesting that potential yield losses associated with Mn deficiency are greater than those for Fe deficiency. Scott and Aldrich (1983) also suggest that Cu deficiencies in the US are relatively rare (limited to Florida and perhaps the South Atlantic Coastal Plain), and Zn deficiency is only a problem for soybean where it is also a problem for corn. However, in other parts of the world (e.g., many parts of Australia), where micronutrient levels in soils are generally low, Zn and Cu deficiencies are also important (Welch et al., 1991).
|
Conclusions |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
1 Mention of a trademark, vendor, or proprietary product does not constitute a guarantee or warranty of the product by USDA-ARS and does not imply its approval to the exclusion of other products or vendors that may also be suitable. 
Received for publication September 4, 1998.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResultsDiscussionConclusionsREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. M. O'Neill, J. F. Shanahan, and J. S. Schepers Use of Chlorophyll Fluorescence Assessments to Differentiate Corn Hybrid Response To Variable Water Conditions Crop Sci., February 1, 2006; 46(2): 681 - 687. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. IZAGUIRRE-MAYORAL and T. R. SINCLAIR Soybean Genotypic Difference in Growth, Nutrient Accumulation and Ultrastructure in Response to Manganese and Iron Supply in Solution Culture Ann. Bot., July 1, 2005; 96(1): 149 - 158. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Read, L. Tarpley, J. M. McKinion, and K. R. Reddy Narrow-Waveband Reflectance Ratios for Remote Estimation of Nitrogen Status in Cotton J. Environ. Qual., September 1, 2002; 31(5): 1442 - 1452. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Adams, W. A. Norvell, W. D. Philpot, and J. H. Peverly Toward the Discrimination of Manganese, Zinc, Copper, and Iron Deficiency in `Bragg' Soybean Using Spectral Detection Methods Agron. J., March 1, 2000; 92(2): 268 - 274. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
