What to Choose for Estimating Leaf Water Status-Spectral Reflectance or In vivo Chlorophyll Fluorescence?

Abstract

In the context of global climate change and the increasing need to study plant response to drought, there is a demand for easily, rapidly, and remotely measurable parameters that sensitively reflect leaf water status. Parameters with this potential include those derived from leaf spectral reflectance (R) and chlorophyll fluorescence. As each of these methods probes completely different leaf characteristics, their sensitivity to water loss may differ in different plant species and/or under different circumstances, making it difficult to choose the most appropriate method for estimating water status in a given situation. Here, we present a simple comparative analysis to facilitate this choice for leaf-level measurements. Using desiccation of tobacco (Nicotiana tabacum L. cv. Samsun) and barley (Hordeum vulgare L. cv. Bojos) leaves as a model case, we measured parameters of spectral R and chlorophyll fluorescence and then evaluated and compared their applicability by means of introduced coefficients (coefficient of reliability, sensitivity, and inaccuracy). This comparison showed that, in our case, chlorophyll fluorescence was more reliable and universal than spectral R. Nevertheless, it is most appropriate to use both methods simultaneously, as the specific ranking of their parameters according to the coefficient of reliability may indicate a specific scenario of changes in desiccating leaves.

Fig. 1.

Workflow and parameters used for comparative analysis.

Fig. 2.

Representative spectra of diffusive R measured on the adaxial side (R_D) of fresh (RWC = 98%), partially desiccated (RWC = 52% or 54%), and severely desiccated leaves (RWC = 5%) of tobacco (A) and barley (E). Comparison of R_D and R_B (R from the abaxial leaf side) in fresh and severely desiccated leaves of tobacco (B) and barley (F). WI (WI = R₉₀₀/R₉₇₀) and relative decrease of R in the 800- to 1,100-nm region (ΔR) estimated from R_D (indexed by “D”) and R_B (indexed by “B”) in desiccating leaves of tobacco (C and D) and barley (G and H). Each point in (C), (D), (G), and (H) represents a single leaf sample. RWC was estimated from the average of the sample weight before and after the measurement.

Fig. 3.

WI (WI_SWIR = R₁₀₀₀/R₁₄₅₀) estimated from measurement of directional R from adaxial side of desiccating leaf samples of tobacco (A) and barley (B).

Fig. 4.

Equivalent water thickness (EWT) during desiccation of tobacco and barley leaf samples within the RWC interval 100% to 50%.

Fig. 5.

Normalized difference vegetation index [NDVI = (R₇₈₀ − R₆₃₀)/(R₇₈₀ + R₆₃₀)] estimated from measurement of diffusive R from the abaxial (NDVI_B) and adaxial side (NDVI_D) of desiccating leaves of tobacco (A) and barley (D). SPAD values of desiccating leaves of tobacco (B) and barley (E). Relative SPAD and NDVI′_D (estimated from measurement of directional R from the adaxial side) of a representative leaf of tobacco (C) and barley (F) during its desiccation (in % of the value measured immediately after leaf detachment). For selected data points, the time after the leaf detachment is indicated.

Fig. 6.

Chl fluorescence parameters of desiccating tobacco and barley leaf samples. (A and D) Maximum quantum yield of PSII photochemistry in the dark-adapted state (F_V/F_M) and effective quantum yield of PSII photochemistry in the light-adapted state (ΦPSII_st). (B and E) Nonphotochemical quenching of Chl fluorescence after 1 min of exposure to actinic light (NPQ₁). (C and F) Nonphotochemical quenching of Chl fluorescence at steady state (NPQ_st). In case of ΦPSII_st and NPQ_st, RWC was estimated from the average of the sample weight before and after the measurement. Each point represents a single leaf sample. For tobacco, in addition to slow desiccating whole leaves, circular leaf segments (14 mm diameter) were used to achieve a lower RWC in a comparable time as with barley leaves.

Fig. 7.

Coefficient of reliability (CR), coefficient of sensitivity (CS), and coefficient of inaccuracy (CI) of parameters measured in desiccating leaf samples of tobacco and barley within the RWC interval 100% to 50%. The parameters have been divided into 5 groups according to the type of leaf characteristics they reflect. A horizontal line in CR plot indicates the reliability threshold (CR = 0.4).