Decreased root hydraulic traits in German winter wheat cultivars over 100 years of breeding

Abstract

Wheat (Triticum aestivum L.) plays a vital role in global food security, and understanding its root traits is essential for improving water uptake under varying environmental conditions. This study investigated how over a century of breeding has influenced root morphological and hydraulic properties in 6 German winter wheat cultivars released between 1895 and 2002. Field and hydroponic experiments were used to measure root diameter, root number, branching density, and whole root system hydraulic conductance (Krs). The results showed a significant decline in root axes number and Krs with release year, while root diameter remained stable across cultivars. Additionally, dynamic functional-structural modeling using the whole-plant model CPlantBox was employed to simulate Krs development with root system growth, revealing that older cultivars consistently had higher hydraulic conductance than modern ones. The combined approach of field phenotyping and modeling provided a comprehensive view of the changes in root traits arising from breeding. These findings suggest that breeding may have unintentionally favored cultivars with smaller root systems and more conservative water uptake strategies under the high-input, high-density conditions of modern agriculture. The results of this study may inform future breeding efforts aimed at optimizing wheat root systems, helping to develop cultivars with water uptake strategies better tailored to locally changing environmental conditions.

Figure 1.

Root diameter density distribution for 6 different cultivars of winter wheat (T. aestivum L.). Data corresponds to lateral A) and axile B) roots obtained from the field with the shovelomics technique, for 2 experimental years (n = 27 to 32 plants). Density plots of lateral and axile roots were separated for visualization purposes. The arrows are a reference of the median value, for the different root types. The color scale indicates the cultivar release year.

Figure 2.

The relationship between year of cultivar release and root morphological traits for 6 different cultivars of winter wheat (T. aestivum L.). Panels A to C) show root diameter, and panels D to F) show number of root axes for different root types. Data points and error bars represent the mean ± CI95% across 2 years of field experiment (n = 27 to 32 plants). The dashed lines and the shaded areas represent the regression line ± CI95% (only shown if significant, P < 0.05). Statistical significance was assessed using linear mixed models. Data were log-transformed prior to model fitting and back-transformed for visualization.

Figure 3.

The relationship between year of cultivar release and whole root system conductance (K_rs) of winter wheat (T. aestivum L.). A) shows pressure chamber measurements of K_rs in 10- to 12-d-old plants for 6 different cultivars, with data points and error bars representing the mean ± CI95% (n = 8 to 12 plants). The dashed line and shaded area represent the regression line ± 95%. Statistical significance was assessed using ordinary least squares regression. B) shows the development of K_rs over plant age for the winter wheat cultivars S. Dickkopf (release year 1895) and Tommi (release year 2002). The line and the shaded area correspond to the mean ± SE of K_rs simulations (n = 6 simulation runs) using the whole-plant model CPlantBox. Data in (A) were log-transformed prior to model fitting and back-transformed for visualization. Notice the log-scale in panel (B).

Figure 4.

Comparison between root hydraulic properties of wheat (T. aestivum L.) obtained from the literature and this study. A) Area-normalized conductance of individual roots (k_root) or whole root systems (K_{rs_area}); and B) whole root system conductance (K_rs) development with age. Black symbols represent literature values measured using a hydrostatic driving force under nonstress conditions (data extracted from a root hydraulic properties database, Baca Cabrera et al. 2024, n = 8 studies). The gray filled circles in (A) and (B) represent the mean and range of variation of pressure chamber measurements in this study (n = 6 cultivars). The gray shadowed area in (B) represents the mean ± SE of K_rs simulations using CPlantBox, as presented in Fig. 3 (n = 6 simulation runs). The dashed black line in (B) represents a fitted exponential model for the literature data (Baca Cabrera et al. 2024).

Figure 5.

Monthly average temperature and monthly total precipitation profiles at Campus Klein-Altendorf, Germany. Lines represent monthly average temperature and bars represent monthly total precipitation measured during the years 2022, 2023 and 2024. For comparison, the long-term means (based on data from 1956 to 2014) are also presented.