cropscience.ch


Just another CropScience weblog

Spectral reflectance indices

Leaf pigments absorb light strongly in the photosynthetic active radiation region (PAR, 400-700 nm) but not in the near infrared region (NIR), thus reducing the reflection of PAR but not of NIR (700-1200 nm) (Araus et al. 2001). Such a pattern of pigment absorption determines the characteristic reflectance signature of leaves. While leaf reflectance is driven by the chemical composition of the leaves, the reflectance of a canopy is influenced by its geometry as well as the reflectance of single leaves (Linke et al. 2008). The reflectance is used to calculate different reflectance indices, which sum up the large amount of information contained in a reflectance spectrum. Some of them are related to plant biomass, photosynthetic size and radiation use efficiency. Other parameters are related to the physiological status, e.g. water content (Araus et al. 2001).

This means on the opposite the more energy is lost due to reflection, heat dissipation and development of reactive oxygen species, the less energy is available to secure grain filling (Lambers et al. 2008, Maxwell and Johnson 2000).

1.1    Reflectance indices related to photosynthesis parameters
Cumulative absorption of photosynthetic active radiation (PAR) is one of the parameters determining total biomass and thus final yield (Araus et al. 2001). The photosynthetic use efficiency can be indirectly assessed using the photochemical reflectance index (PRI) derived from narrow-band reflectance at 531 and 570 nm (Filella et al. 1996, Peñuelas et al. 1995). The index was originally developed for estimation of xanthophyll cycle pigment changes (Gamon et al. 1997) but is also related to carotenoid/chlorophyll ratios in green leaves (Sims and Gamon 2002), zeaxanthin concentration (Filella et al. 1996) and PSII quantum yield (Peñuelas et al. 1995). PRI is most useful at the scale of leaves and homogenous, closed canopies where structural complexities are minimized (Gamon et al. 1997). If PRI is measured at canopy level the measurements are very sensitive to soil reflectance properties especially in vegetation areas with low green leaf area index (Mänd et al. 2009).
Aside of PRI the red edge inflection point (REIP, Vogelmann et al. 1993, Filella and Peñuelas 1994), the RE3/RE2 and the D715/705 (Vogelmann et al. 1993), the indices reported by Sims and Gamon (2002) were reported to be related to chlorophyll content. The ratio of the reflectance of pine canopies at 701±2 and 820±2 regressed stronger with photosynthetic capacity than the red edge inflection point (Carter 1998).

PRI (Gamon et al. 1997):                       (R531-R570) / (R531+ R570)
REIP (Vogelmann et al, 1993):             Maximum in R680-R780 nm
RE3/RE2 (Vogelmann et al. 1993):     (R734 to R747) / (R715 to R726)
D715/705 (Vogelmann et al. 1993):     (R710 to R720) / (R700 to R710)
SR680 (Sims and Gamon 2002):         R800 / R680
SR705 (Sims and Gamon 2002):         R730/ R705
SR (Sims and Gamon 2002):                (R750 to R900) / (R660 to R720)
mSR (Sims and Gamon 2002):            ((R750 to R900) – R445)/ ((R660 to R720) – R445)
NDVI680 (Sims and Gamon 2002):     (R800 – R680) / (R800 + R680)
NDVI705 (Sims and Gamon 2002):     (R750 – R705) / (R750 + R705)
mND (Sims and Gamon 2002):            (R750 to R900) – (R660 to R720) / (R750 to R900) +(R660 to R720) – 2R445)

PRI: Photochemical reflectance index
REIP: Red Edge Inflection Point
REP: Red Edge Peak
SR: Spectral reflectance index
mSR. Modified simple reflectance
NDVI: Normalized Difference Vegetation Index
mND: Modified normalized difference index

1.2    Reflectance indices related to plant nutrition
Plants with low N usually have a high carotenoid (Car) to Chl ratio, which can also be assessed by reflectance indices (Araus et al. 2001). For example the canopy reflectance at 531 nm is correlated to the xanthophylls pigment epoxidation state (XES, Gamon et al. 1990). In the study of Sims and Gamon (2002) the photochemical reflectance index (PRI, see 1.1) was related to carotenoid/chlorophyll ratios whereas the Structure-Insensitive Pigment Index (SIPI) and Plant Senescence Reflectance Index (PSRI) did not correlate with carotionoids and chlorophyll concentration. The Normalized total Pigment to Chlorophyll a ratio Index (NCPI) was reported to be correlated with chlorophyll content, carotenoid / chlorophyll a, lutein, and neoxanthin / chlorophyll a.  Further NCPI varies with the ratio to total pigments/chl, indicative of plant phenology and physiological status (Peñuelas et al. 1994). The plant senescence reflectance index (PSRI) was found to be sensitive to the Car/Chl ratio and was used in the study of Merzlyak et al. (1999) as a quantitative measure of leaf senescence and fruit ripening.

Low arrow band ration (Carter1998):    (R820-R701) / (R820+R701) or R701/R820
XES (Gamon et al. 1990):                         R531
SIPI (Penuelas et al. 1995):                      (R800 – R445) / (R800 – R680)
NCPI (Penuelas et al. 1993 I):                 (R680 – R430) / (R680 + R430)
PSRI (Merzlyak et al. 1999):                   (R680 – R500) / R750
PSR (Penuelas et al. 1993 I):                   R430 / R680

XES: Xantophyll epoxidation state
SIPI: Structure-Insensitive Pigment Index
PSRI: Plant Senescence Reflectance Index
NCPI: Normalized total Pigment to Chlorophyll a ratio Index


1.3    Reflectance indices related to aboveground biomass

Leaf area index (LAI), green area index (GAI) and green leaf area index (GLAI) can be estimated through their positive correlation with vegetation indices (Baret and Guyot, 1991, Price and Bausch, 1995). Further the leaf area duration (LAD) can be estimated by measuring vegetation indices periodically during the crop growing cycle (Araus et al. 2001).
One of the most widely used spectral reflectance index (SRI) is the Normalized Difference Vegetation Index (NDVI) which relates the difference between near infrared reflectance and red wavelength reflectance with the reflectance of both wavelengths (Marti et al. 2007). NDVI is a sensitive indicator of canopy structure, green biomass, green leaf area index, chlorophyll content and foliar nitrogen content (Gamon et al. 1995). The NDVI is more sensitive to changes in the crop canopy when the leaf area index is low. It becomes saturated when the crop canopy closes (Inman et al. 2008). NDVI can be measured using a GreenSeeker device. The device includes its own radiation source and can thus be used regardless of the athmospheric conditions (Araus et al. 2008).
Different studies showed that the development of shoot biomass (Marti et al. 2007), yield potential (Teal et al. 2006, Inman et al. 2007) and leaf area index (Marti et al. 2007, Mänd et al. 2009) of maize could be accurately predicted with NDVI. Aside of NDVI, REIP was reported to be related to leaf area index (Filella and Peñuelas (1994).
In response to increasing salinity NDVI (formula 3) decreased in parallel with biomass and yield of barley (Penuelas et al. 1997 I). On the other hand, Jiang and Carrow (2007) stated that NDVI ratios (formula 4, 5) were not as drought sensitive as three to five broadband models using a wider range of 660 to 1480 nm. The green normalized difference vegetation index (GNDVI) has a wider dynamic range than the NDVI and is more sensitive to Chlorophyll-a concentration (Gitelson et al. 1996). Further NDVI is sensitive to differences in canopy cover within a green LAI of 0 and 2, but it loses sensitivity in moderate and dense canopies (Gamon et al., 1995, Baret and Guyot 1991). To overcome the bias of soil reflection Huete (1988) developed the Soil-Adjusted Vegetation Index (SAVI), which includes the addition of a constant L in the NDVI formula. In the study of Price and Bausch (1995) L was set 0.5 to measure the SAVI for a maize canopy.
Sims and Gamon (2003) developed a Canopy Structure Index (CSI) that combines the low absorbance water bands with the simple ratio vegetation index (SR) to produce an index with a wider range of sensitivity to photosynthetic tissue area at all canopy thicknesses. CSI can be used for the prediction of total area of photosynthetic tissues (Sims and Gamon 2003).
(1)    NDVIred (Mänd et al. 2009):          (R780-R680) / (R780+R680)
(2)    NDVIgreen (Mänd et al. 2009):     (R780-R570) / (R780+R570)
(3)    NDVI (Penuelas et al. 1997):           (R900-R680) / (R900+R680)
(4)    NDVI (Jiang and Carrow 2007):    (R760-R710) / (R760+R710)
(5)    NDVI (Jiang and Carrow 2007):     (R950-R660) / (R950+R660)
GNDVI (Gitelson et al. 1996):                   (R780-R550) / (R780+R550)
LAI (Jiang and Carrow 2007):                  R950 / R660
SAVI (Huete 1988):                                     [(RNIR - RRED)/( RNIR + RRED + L)] ×(1+L), L=constant, i.e. 0.5
CSI (Sims and Gamon 2003):                   2 sSR – sSR² + sWI²
WI=                                                                 R900 /R970
sWI=                                                               (R900 / R1180 -1) / (R900 / R1180 -1)max
sSR=                                                               (R800 /R680 -1) /(R800 /R680 -1)max
TCARI (Liu et al. 2010):                             3(R700-R670)-0.2(P700-P550)R700/R670
OSAVI (Liu et al. 2010):                             1.16(R800-R670)/(R800+R670+0.16)

R: Reflectance
NDVI: Normalized Difference Vegetation Index
GNDVI: Green Normalized Difference Vegetation Index
LAI: Leaf Area Index
SAVI: Soil-Adjusted Vegetation Index
CSI: Canopy Structure Index
WI: Water Index
SR: Simple Ratio

1.4    Reflectance indices related to plant water status
The results of Peñuelas et al. (1993) indicated that the reflectance in the 950-970 nm regions is an indicator of Gerbera water status. With the ratio of R970 to R900 the authors developed the water balance index (WBI, formula 1-3). WBI was related to water content of sunflower under drought stress (Peñuelas et al. 1994), of barley under salt stress (Penuelas et al. 1997 II), of aquatic vegetation (Peñuelas et al. 1993 I) and of shrubs (Penuelas et al. 1997 I). The WBI was also reported to be linked to water vapor and carbon dioxide fluxes, which suggests that WBI may be useful in modeling CO2 and water fluxes (Claudio et al. 2006).
The normalized difference water index (NDWI), developed by Gao (1996) proved to be sensitive to changes in water content of vegetation canopies at the landscape scale (Gao 1996, Serrano et al. 2000). Further three wavelength regions (950-970 nm, 1150-1260 nm and 1520-1540 nm) were identified by Sims and Gamon (2003) that are related to water content measured at the canopy level of 23 species. Fensholt and Sandholt (2003) developed a short wave infrared water stress index (SIWSI) to indicate canopy water content. The indices listed were measured until now only on the canopy level. Linke et al. (2008) predicted the relative and actual water content of wheat leaves under drought stress with reflectance models. For tracing changes in physiological parameters during phenology and stress periods the use of these indices was not promising (Linke et al. 2008). Yu et al. (2000) developed models to estimate leaf water content (RWC) and relative leaf moisture percentage on fresh weight basis (RMP) in herbaceous and woody plants which showed high correlation. The ratios measured at leaf level have not yet been tested in maize.

(1)    WBI (Peñuelas et al. 1993 II):     R970/ R900
(2)    WBI (Peñuelas et al. 1997 II):    R905 / R980
(3)    WBI (Peñuelas et al. 1994):        R970 / R902
NDWI (Gao 1996):                                (R800-R680) / (R800+R680)
Sims and Gamon (2003):                    ∑R950 to R970
∑R1150 to R1260
∑R1520 to R1540
SIWSI (Fensholt and Sandholt 2003):     (R1628 to R1652) – (R841 to R876) / (R1628 to R1652) + (R841 to R876)
SIWSI (Fensholt and Sandholt 2003):    (R1230 to R1250) – (R841 to R876) / (R1230 to R1250) + (R841 to R876)
RWC (Linke et al. 2008):                           R1483 / R1650
RWC (Yu et al. 2000):                                R1100 / R1430
R1120 / R1430
R1430 / R1650
R1483 / R1650
R1430 / R1830
AWC (Linke et al. 2008):                           R1121 / R1430
RMP (Yu et al. 2000):                                 R2200 / R1430
R1430 / R1650
R1483 / R1430

WBI: Water Band Index
NDWI: Normalized Difference Water Index
SIWSI: Short Wave Infrared Water Stress Index
RWC: Relative Water Content
AWC: Actual water content
RMP: relative leaf moisture percentage on fresh weight basis

1.5    Reflectance indices related to grain yield
In the study of Ferrio et al. (2005) canopies with higher grain yield showed lower reflectance in 500–700 nm than low-yielding plots. The reflectance in the near-infrared (>700 nm) was higher, with the exception of the bands between 950 and 1000 nm. Partial least squares regression (PLSR) was used in the construction of models. The empirical models for the estimation of grain yield showed generally stronger and more robust assessment of grain yield than previously assayed spectral indices. Ferrio et al. (2005) concluded that, although the models did not provide an accurate quantification of grain yield, they could still be used to rank genotypes for breeding purposes. The most reliable ranking of genotypes was attained using measurements made at milk-grain stage. The possibility of ranking genotypes with measurements at leaf level has yet to be tested.

Grain yield (Ferrio et al. 2005):     R500 to R700
R700 to R950
R950 to R1000
Grain yield (Royo et al. 2003):     R680

1.6    Reflectance indices related to drought stress
Linke et al. (2008) researched the spectral reflectance of wheat leaves in the climate chamber during flowering drought stress. The leaf reflectance of plants subjected to drought increased over the entire range of the spectrum. However, five spectral regions (520-530 nm, 570-590 nm, 690-710 nm, 1410-1470 nm and 1880-1940 nm) with relatively high differences were identified. However, for tracing changes in physiological parameters during stress periods the use of these indices was not promising (Linke et al. 2008). The relative reflectance index (PRI) was reported to be sensitive for the determination of early water stress in rape (Morgensen et al. 1996). The stress indices developed by Jiang and Carrow (2007) showed a positive relationship with turf quality under drought stress. The spectral reflectance indices (NDVI, GNDVI, Indices for Chlorophyll content) used by Rodriguez et al. (2004) showed potential for identifying higher-yielding wheat genotypes in a breeding program under dry or irrigated conditions. The application of the indices for maize under drought stress has yet to be researched.
Jiang and Carrow (2007) concluded that the traditional vegetation indices based on two bands within 660 to 950 nm may not be as sensitive as three to five band models using a wider band range of 660 to 1480 nm. Additionally the authors concluded that the wavelength in R900 and R1200 should be considered in drought sensitive models (Jiang and Carrow 2007). Thus, the applicability of models based solely on spectral data is limited (Doughtry et al. 1992). Several approaches have been proposed to use multispectral data in conjunction with meteorological and soil data in models to predict crop yields. For example, spectral estimates of intercepted radiation may be incorporated into the Energy-Crop Growth (ECG) model developed by Coelho and Dale (1980). Within this model the light, temperature and moisture are combined.
In general the application of the listed indices and models for drought tolerance selection maize has to be tested.

The scarce use of spectral reflectance measurements as tools for screening in breeding programs is that a wide range of variability for the trait of interest must occur in order to be detected by the apparatus (Royo et al. 2003, in Araus et al. 2008). Further the devices currently available allow measurements only at canopy level (Araus et al. 2008).

Linke et al. (2008):        ∑520 to 530 nm
∑570 to 590 nm
∑690 to 710 nm
∑1410 to 1470 nm
RI (Morgensen et al. 1996)    RNIR/ RPR
PRI (Morgensen et al. 1996):     PRIDS / PRIWW
SI (Jiang and Carrow 2007):     R710 / R810
SI (Jiang and Carrow 2007):     R710 / R760
ECG (Coelho and Dale 1980):    SR/600 * f(LAI) * (ET / PET) * FT

RI: Spectral Reflectance Index
NIR: Reflectance in the near infrared band
PR: Reflectance in the photosynthetically active band
PRI: Relative Reflectance Index
SI: Stress Index
ECG: Energy Crop Growth Model
SR: daily solar radiation
LAI: Leaf Area Index
f(LAI): 1-exp(-0.79 * LAI)
ET/PET: Ratio of actual and potential evapotranspiration
FT: daily temperature function, as described in Coelho and Dale (1980)

Literature

Araus, J.L., G. a. Slafer, C. Royo, and M.D. Serret. 2008. Breeding for Yield Potential and Stress Adaptation in Cereals. Critical Reviews in Plant Sciences 27(6): 377–412.

Claudio, H., Y. Cheng, D. Fuentes, J. Gamon, H. Luo, W. Oechel, H. Qiu, A. Rahman, and D. Sims. 2006. Monitoring drought effects on vegetation water content and fluxes in chaparral with the 970 nm water band index. Remote Sensing of Environment 103(3): 304–311.

Fensholt, R., and I. Sandholt. 2003. Derivation of a shortwave infrared water stress index from MODIS near- and shortwave infrared data in a semiarid environment. Remote Sensing of Environment 87(1): 111–121.

Ferrio, J.P., D. Villegas, J. Zarco, N. Aparicio, J.L. Araus, and C. Royo. 2005. Assessment of durum wheat yield using visible and near-infrared reflectance spectra of canopies. Field Crops Research 94(2-3): 126–148.

Filella, I., T. Amaro, J.L. Araus, and J. Penuelas. 1996. Relationship between photosynthetic radiation-use efficiency of barley caeopies and the photochemical reflectance index ( PRI ). Physiol. Plant 96: 211–216.

Filella, I., and J. Penuelas. 1994. The red edge position and shape as indicators of plant chlorophyll content, biomass and hydric status. International Journal of Remote Sensing 15(7): 1459–1470.

Gamon, A., L. Serrano, and 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.

Gao, B. 1996. NDWI A Normalized Difference Water Index for Remote Sensing of Vegetation Liquid Water From Space. 266(April 1995): 257–266.

Huete, A. 1988. A Soil-Adjusted Vegetation Index ( SAVI ). Remote Sensing of Environment 309: 295–309.

Inman, D., R. Khosla, R. Reich, and D.G. Westfall. 2008. Normalized Difference Vegetation Index and Soil Color-Based Management Zones in Irrigated Maize. : 60–66.

Inman, D., R. Koshla, R. Reich, and D. Westfall. 2007. Active remote sensing and grain yield in irrigated maize. precision agric 8: 241–252.

Jiang, Y., and R.N. Carrow. 2007. Broadband Spectral Reflectance Models of Turfgrass Species and Cultivars to Drought Stress. Crop Science 47(4): 1611.

Linke, R., W. Schneider, and P. Weihs. 2008. Occurrence of repeated drought events : can repetitive stress situations and recovery from drought be traced with leaf reflectance ? periodicum biologorum 110(3): 219–229.

Mänd, P., L. Hallik, J. Peñuelas, T. Nilson, P. Duce, B. a. Emmett, C. Beier, M. Estiarte, J. Garadnai, and T. Kalapos. 2010. Responses of the reflectance indices PRI and NDVI to experimental warming and drought in European shrublands along a north–south climatic gradient. Remote Sensing of Environment 114(3): 626–636.

Marti, J., J. Bort, G.A. Slafer, and J.L. Araus. 2007. Can wheat yield be assessed by early measurements of Normalized Difference Vegetation Index? Annals of Applied Biology 150(2): 253–257.

Maxwell, K., and G.N. Johnson. 2000. Chlorophyll fluorescence – A practical guide. Journal of experimental botany 51(345): 659–668.

Peñuelas. 1997. visible near-infrared reflectance assessment of salinity effects on barley. Crop Sci. 37: 198–202.

Penuelas, J., I. Filella, and J.A.. Gamon. 1995. Assessment of Photosynthetic Radiation-Use Efficiency with Spectral Reflectance. new physiologist 131(3): 291–296.

Penuelas, J., J. Pinol, R. Ogaya, and I. Filella. 1997. Estimation of plant water concentration by the reflectance Water Index WI (R900/R970). International Journal of Remote Sensing 18(13): 2869–2875.

Price, J.C., and W.C. Bausch. 1995. Leaf Area Index Estimation from Visible and Near-Infrared Reflectance Data. 65(November 1994): 55–65.

Reynolds, M. 2002. Application of Physiology in Wheat Breeding. CIMMYT, Mexico, D.F.

Serrano, L., S.L. Ustin, D.A. Roberts, J.A. Gamon, and J. Pen. 2000. Deriving Water Content of Chaparral Vegetation from AVIRIS Data. Remote Sensing of Environment 581: 570–581.

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(2-3): 337–354.

Sims, D. a, and J. a Gamon. 2003. Estimation of vegetation water content and photosynthetic tissue area from spectral reflectance: a comparison of indices based on liquid water and chlorophyll absorption features. Remote Sensing of Environment 84(4): 526–537.

Teal, R.K., B. Tubana, K. Girma, K.W. Freeman, D.B. Arnall, O. Walsh, and W.R. Raun. 2006. In-Season Prediction of Corn Grain Yield Potential Using Normalized Difference Vegetation Index. Agronomy Journal 98: 1488–1494.

Vogelmann, J.E., B.N. Rock, and D.M. Moss. 1993. Red edge spectral measurements from sugar maple leaves. International Journal of Remote Sensing 14(8): 1563–1575.

Yu, G., T. Miwa, K. Nakayama, N. Matsuoka, and H. Kon. 2000. A proposal for universal formulas for estimating leaf water status of herbaceous and woody plants based on spectral reflectance properties. Plant and Soil 227: 47–58.

 

June 5th, 2012
Topic: Crop Science, Plant breeding Tags: None

2 Responses to “Spectral reflectance indices”

  1. Secondary traits for maize grain yield under drought stress conditions | cropscience.ch Says:

    [...] Spectral reflectance indices [...]

  2. Statistical methods to analyze large spectral data sets | cropscience.ch Says:

    [...] Spectral reflectance indices [...]

≡ Leave a Reply

You must be logged in to post a comment.