Identify Crop Characteristics Conferring Stress Adaptation
Many crop characteristics have been reported as conferring improved stress adaptation (e.g., Rebetzke et al. 2009; Cossani and Reynolds 2012) and well controlled phenotypic studies can indicate whether these traits may be complementary in different combinations (Reynolds et al. 2007a). However, the only way to confirm their value definitively is to find sources with good expression of the traits and introgress into elite backgrounds. The approach has been successful within the IWIN for producing a new generation of drought adapted lines as will be discussed in the last section. Some of the traits that have proven to be most useful for both heat and drought adaption include (see Richards 2006; Rebetzke et al. 2009; Reynolds et al. 2010):
• Cooler canopy temperature that indicates access to subsoil water under drought and an adequate vascular system –including rootsto match evaporative demand at high vapour pressure deficit typical of heat stress in dry environments.
• Rapid early ground cover to avoid wasteful evaporation of water at the soil surface back to the atmosphere.
• Intrinsic transpiration efficiency that ensures conservative use of water when it is relatively abundant as a water budgeting strategy.
• Epicuticular leaf wax that reflects excess radiation and reduces evaporation from the leaf surface thereby reducing the risk of photo-inhibition and dehydration, respectively.
• Accumulation and remobilization of water soluble carbohydrates in storage organs like stems permitting grain-filling to continue even when post-anthesis stress is too severe to permit adequate carbon assimilation.
• Membrane thermo-stability that can be screened using electrolyte leakage, and chlorophyll fluorescence that, although it has not been used systematically in selecting parents, has become a recent focus for screening.
Exploration of Genetic Resources for Adaptive Traits: Landraces
CIMMYT and other wheat breeding programs have been using landraces to broaden the wheat gene pool for decades (Smale et al. 2002), mostly for disease resistance traits as these are relatively easy to detect in un-adapted material and select for in segregating progeny (Reynolds and Borlaug 2006). More recently Mexican landraces were collected -after approximately 500 years of essentially natural selection in some of Mexico's harshest rain-fed environments-. These were screened for heat and drought adaptive traits and a few of the lines show exceptional characteristics such as deep roots and ability to store water soluble carbohydrates in the stem (Reynolds et al. 2007b). Very recently, over 70,000 accessions of the World Wheat Collection have been screened for adaptation to heat and drought stress -mostly for the first time-, and diverse panels have been assembled for genetic analysis. These include elite durum and bread wheat lines, landraces from hot and dry regions worldwide, and lines derived from interspecific hybridization with wheat's relatives, including 're-synthesized' wheat (see next section).
Other more targeted approaches include characterization of panels identified using agro-ecological data also known as the Focused Identification of Germplasm Strategy or FIGS (figs.icarda.net/). The principle is that landraces from heat and drought stressed regions are more likely to contain stress adaptive-traits. In the case of drought, we obtained 204 landraces originating from drought-affected areas. These were grown in Cd Obregon, Sonora, under drought conditions (only one irrigation at sowing providing approximately 180 mm of available water to 1.2 m depth, and no significant rainfall during the growing season). CIMMYT varieties were included as checks. Total above ground biomass was estimated and grain yield after threshing (Fig. 41.1).
As shown in Fig. 41.1, lines capable of producing high biomass were identified. Over 45 % of the entries had greater biomass under drought than the CIMMYT check varieties, some with 40 % greater biomass. The harvest index of the lines was lower than that of CIMMYT checks and as a result only 5 % of the entries had
Fig. 41.1 Frequency distribution of 208 FIGS lines grown under drought for (a) total biomass (b) grain yield. The mean of three CIMMYT checks (Sokoll, Roelfs, Weebil) is indicated by the green arrow
greater yield than the CIMMYT varieties. This is not unexpected for landraces: most of the FIGS lines were tall, and some were late-flowering resulting in greater stress during grain filling. These traits would have contributed to the low harvest index of the landraces in addition to not necessarily being adapted to the photoperiod conditions of the experimental environment. The lines that demonstrated the capacity to produce high biomass are being crossed to CIMMYT elite lines as part of the Physiology Group's pre-breeding program. The progeny lines will be selected for semi-dwarf stature and appropriate maturity.