Leaf-rolling in maize crops: from leaf scoring to canopy-level measurements for phenotyping

Abstract

Leaf rolling in maize crops is one of the main plant reactions to water stress that can be visually scored in the field. However, leaf-scoring techniques do not meet the high-throughput requirements needed by breeders for efficient phenotyping. Consequently, this study investigated the relationship between leaf-rolling scores and changes in canopy structure that can be determined by high-throughput remote-sensing techniques. Experiments were conducted in 2015 and 2016 on maize genotypes subjected to water stress. Leaf-rolling was scored visually over the whole day around the flowering stage. Concurrent digital hemispherical photographs were taken to evaluate the impact of leaf-rolling on canopy structure using the computed fraction of intercepted diffuse photosynthetically active radiation, FIPARdif. The results showed that leaves started to roll due to water stress around 09:00 h and leaf-rolling reached its maximum around 15:00 h (the photoperiod was about 05:00-20:00 h). In contrast, plants maintained under well-watered conditions did not show any significant rolling during the same day. A canopy-level index of rolling (CLIR) is proposed to quantify the diurnal changes in canopy structure induced by leaf-rolling. It normalizes for the differences in FIPARdif between genotypes observed in the early morning when leaves are unrolled, as well as for yearly effects linked to environmental conditions. Leaf-level rolling score was very strongly correlated with changes in canopy structure as described by the CLIR (r2=0.86, n=370). The daily time course of rolling was characterized using the amplitude of variation, and the rate and the timing of development computed at both the leaf and canopy levels. Results obtained from eight genotypes common between the two years of experiments showed that the amplitude of variation of the CLIR was the more repeatable trait (Spearman coefficient ρ=0.62) as compared to the rate (ρ=0.29) and the timing of development (ρ=0.33). The potential of these findings for the development of a high-throughput method for determining leaf-rolling based on aerial drone observations are considered.

Fig. 1.

Diurnal variation of global radiation (Rg), temperature (T), and vapor pressure deficit (VPD) recorded on the days on which measurements were taken in 2015 and 2016. UT, universal time. (This figure is available in color at JXB online.)

Fig. 2.

Location of digital hemispherical photography (DHP) measurements across the two rows of a microplot. A stick was placed in the positions shown and images were taken at the locations indicated by the circles. (This figure is available in color at JXB online.)

Fig. 3.

Example digital hemispherical photography (DHP) images taken at seven different time-points (from S1, morning, to S7, late afternoon) at the same location within a microplot. The changes of canopy structure from the morning to afternoon due to leaf rolling can clearly be seen. Some artifacts are apparent for S4 and S5 due to direct sunlight; however, these were fully accounted by the color classification that was used in image processing. (This figure is available in color at JXB online.)

Fig. 4.

Diurnal pattern of the leaf-rolling scores for the water-stress treatment for 30 genotypes in 2015 (left) and 16 genotypes in 2016 (right). In the box-plots the line within the box is the median, the tops and bottoms of the box are the 75th and 25th percentiles, respectively, the whiskers extend to the most extreme data points that the algorithm considers not to be outliers, and outliers are plotted individually as ‘+’. (This figure is available in color at JXB online.)

Fig. 5.

Diurnal pattern for nine segmental gap fractions, $\overline{P_{\text{o}}\left(\theta , {\phi }_q\right)}$ The directions (θ, φ_q) are indicated at the end of each curve. $\overline{P_{\text{o}}\left(\theta , {\phi }_q\right)}$ was computed as the average for 30 genotypes in 2015 (water-stress treatment only), 16 genotypes in 2016 under water-stress (WS) and four genotypes in 2016 under well-watered (WW) conditions. (This figure is available in color at JXB online.)

Fig. 6.

Distribution of the values for the fraction of intercepted diffuse radiation FIPAR_dif(t₀) observed in the early morning when leaves were in the unrolled state for all the 50 genotypes investigated. The values are sorted in ascending order. The years and treatments are indicated: WS, water stress; WW, well-watered. The identifiers of genotypes (Table 1) common between years and treatments are indicated above each bar. (This figure is available in color at JXB online.)

Fig. 7.

Left: the relationship between the fraction of intercepted diffuse radiation for unrolled leaves observed in the early morning, FIPAR_dif(t₀), and the values corresponding to the maximum leaf-rolling observed in the late afternoon, FIPAR_dif^max_roll. Right: the relationship between FIPAR_dif(t₀) and the difference in values observed between the early morning and the late afternoon, ΔFIPAR_dif^max_roll. The years and treatments are indicated: WS, water stress. The lines represent the linear best fit. (This figure is available in color at JXB online.)

Fig. 8.

Left: the relationship between ΔFIPAR_dif(t) and the leaf-rolling visual score [Score(t) – 1]. The solid lines correspond to the best-fit line verifying the constraint ΔFIPAR_dif(t) = 0 when Score(t) = 1 (eqn 3) for the 2015 and 2016 water-stress (WS) treatments. The dashed line corresponds to the best-fit over all the 370 available data points including the well-watered (WW) treatment in 2016. Right: the relationship between the canopy-level index of rolling, CLIR(t) (eqn 4) and the leaf-rolling visual score [Score(t) – 1]. The dashed line corresponds to Equation 5.

Fig. 9.

Left: distribution of the maximum value of the leaf-rolling visual score, max(Score – 1), observed for each microplot during the day. Right: distribution of the maximum value of the canopy-level index of rolling, max(CLIR), observed for each microplot during the day. The values are sorted in ascending order. The years and treatments are indicated: WS, water stress; WW, well-watered. The identifiers of genotypes (Table 1) common between years and treatments are indicated above each bar. (This figure is available in color at JXB online.)

Fig. 10.

Left: comparison between the maximum value of the normalized scores, (Score – 1), observed in 2015 and 2016. Right: comparison between the maximum value of the canopy-level index of rolling, CLIR, observed in 2015 and 2016. The numbers correspond to the genotype identifier (Table 1). The 1:1 line is indicated. The Pearson (r²) and Spearman (ρ) coefficients are provided.

Fig. 11.

The diurnal dynamics of the leaf-rolling visual scores, (Score – 1), evaluated at the leaf level (black line), and the score estimated from eqn (5) (red line) from canopy-level digital hemispherical photography (DHP) measurements. The years and treatments are indicated: WS, water stress; WW, well-watered. WW 2016, four genotypes; WS 2016 and WS 2015, the same eight genotypes. The genotype identifier (Table 1) is indicated in each graph. (This figure is available in color at JXB online.)

Fig. 12.

The dynamics of the normalized rolling score, (Score_norm) evaluated at the leaf level (circles) and the canopy level (squares) using eqn (6) for the time interval 09:00 to 13:00 h (UT, universal time). The lines correspond to the best linear robust fit. The years and treatments are indicated: WS, water stress. The same eight genotypes were used throughout. The genotype identifier (Table 1) is indicated in each graph.

Fig. 13.

Patterns of distribution of the time (t_max/2) when half the maximum rolling was observed (left), and the slope corresponding to the rate of change of rolling from minimum to maximum (α). The years and treatments are indicated: WS, water stress. The identifiers of genotypes (Table 1) common between years and treatments are indicated above each bar. (This figure is available in color at JXB online.)