Increases in SST, upward energy flux, cloud cover, and precipitation over the STIO have been accompanied by diabatic heating in the mid-troposphere due to the con version of latent energy into DSE during condensation . Figure 9e shows that STIO surface warming corresponds with a significant increase in the amount of NCEP2 DSE exported from the region . As was the case for GPCP precipitation, divergence of DSE increased most rapidly in the northwestern portion of the STIO. In this region, DSE correlated well with GPCP precipitation. Agreement between these two independently derived data sets, and their cor relation with STIO SSTs, testifies to the strength of the inter-connection between STIO SSTs, precipitation in the STIO region, and the large-scale energetic response in the atmosphere.Increasing exports of DSE from the STIO region appear to be associated with increasing imports of DSE into the GHA . While Fig. 9e does not explicitly link divergence of DSE over the STIO to convergence of DSE over the GHA, the overturning trend shown in Fig. 8c does. This atmospheric response is in agreement with previous climate modeling work showing that an off-equatorial positive heating anomaly should lead to decreased convective activity over eastern Africa via a Gill-type response . This is further supported by Fig. 8c, d, which shows that upward vertical velocities over the Ethiopian Highlands have slowed by as much as 0.06 Pa s-1 . Finally, reduced convection over the GHA also reduces westerly moisture transport from the Congo Basin . This is particularly apparent in maps comparing mean states of atmospheric circulation during the 1948–1988 and 1989–2009 periods in Fig. 18b, d in ‘‘Appendix’’ .Climate change is increasing water stress in many areas as a result of increasing evaporative demand, altered precipitation and earlier snowmelt .
Moderate drought is a common cause of reduced growth and increased mortality ,ebb and flow trays whereas severe droughts have led to mass mortality . Differences in drought susceptibility between tree species and populations have both environmental and genetic components. There is evidence of local adaptation to climate in many trees despite high gene flow . Populations often do not respond equally to a given climate . To better manage tree populations, we need to understand the relationships between tree genotype and phenotype and between phenotype and ecological function . The relative importance of drought response plasticity vs genetic differences is still largely unknown, as are details of which traits are most important for drought tolerance and what genes underlie them. Here, we focus on the genetics of drought tolerance in conifers. Although some general principles apply to both angiosperms and gymnosperms, there are significant differences. Gymnospermsare generally more drought resistant as a result of lower stomatal sensitivity to vapor pressure deficit and more cavitation-resistant xylem . Conifer xylem is made up entirely of tracheids, whereas angiosperms may produce both tracheids and wide vessels, which have higher hydraulic conductivity, but a smaller margin of safety with regard to xylem pressures . In addition, angiosperms have more complex anatomical responses to drought, such as changes in vessel connectivity . Moreover, much research attention has been given to drought tolerance in conifers because they are a dominant component of many arid zone forests. We first address the definitions of ‘drought tolerance’ and the physiological mechanisms involved.We then describe three major methods that have been used to study drought tolerance genetics, and review the major findings to date. Finally, we discuss how genetic tools may aid forest management, and needs for future research.
In particular, we recommend the combination of complementary methods, and the broadening of the range of phenotypes, taxa, life stages and time spans examined.A basic definition of drought tolerance is the ability to survive, and sometimes grow, during periods of water shortage. Survival and growth are often correlated, with trees exhibiting a history of below average growth or abrupt decreases in growth having higher mortality . Because of this, and because tree genetics studies are often motivated by wood production, some studies define drought tolerance as growth maintenance . However, drought length and duration can affect the growth–mortality relationship. For instance, growth plasticity may be adaptive in variable environments . Populations adapted to extended drought often exhibit conservative resource use strategies, resulting in slower growth rates, even in favorable conditions . Drought tolerance can be broken down into several categories. ‘Drought avoidance’ strategies reduce exposure to drought stress . However, it is unclear to what degree trees truly avoid drought stress relative to plants that go dormant. ‘Drought resistance’ is the ability to withstand drought exposure, whereas ‘drought resilience’ is a measure of how quickly a tree can resume normal growth when conditions improve .Conifers manage tissue water potential in two main ways: isohydric trees close stomata to maintain water potential, whereas anisohydric species allow water potential to drop . Isohydric trees use increasing abscisic acid concentrations as a signal to keep stomata closed, whereas anisohydric trees use low leaf water potential itself as a signal to close stomata . Anisohydric conifers include many Cupressaceae and some Taxaceae . Xylem architecture affects how changes in stomatal conductivity influence cavitation risk, and anisohydric trees tend to have xylem that is more cavitation resistant . Wider tracheids increase conductivity and the risk of hydraulic failure , whereas those with smaller inter-tracheid pits or more lignified walls are less vulnerable . The reduction of leaf area with branch die-back, reduced needle number or smaller needles can also reduce water loss. Anisohydric species often exhibit branch die-back during drought, whereas isohydric trees typically retain a full canopy until death . Some conifer species can refill xylem following cavitation.
This is thought to be an energy-intensive process that depends on carbon reserves . This may explain why drought stressed trees can exhibit lower refilling capability . Picea abies refills freezing-cavitated xylem before soils have thawed by taking up water through its needles . This could explain why other conifers can refill xylem in the absence of positive root pressures, unlike co-occurring angiosperms . However, refilled xylem may be less resistant to future drought stress, a characteristic known as ‘cavitation fatigue’ .Loss of water potential in cells is associated with cell turgor loss,ebb flow tray denaturation of proteins and changes in membrane fluidity. To avoid cellular damage, plants synthesize molecules that act as osmotic balancing agents. These reduce cellular solute potential, and may increase turgor at lower water potentials. In addition, hydrophilic compounds can prevent the membranes from leaking . Other com pounds stabilize proteins or detoxify reactive oxygen species. These protective molecules include proteins such as chaperonins and dehydrins , the amino acid proline and various carbohydrates .We hypothesize that protective molecules may be produced earlier during a drought in anisohydric species because leaf water potential drops more quickly.Rooting depth affects access to deep soil water and is probably crucial for seedlings as well as adult trees in areas with seasonal drought . Deep roots may also redistribute water from deep to shallow soils . More small diameter roots, with high surface area : volume and a lower vulnerability to cavitation, may aid drought resistance . Structural changes can have long-lasting effects. Decreasing soil moisture can induce greater root production, but extended drought reduces root mass , which limits responsiveness to precipitation pulses . Lumen width and cell wall thickness of tracheids are plastic, with those produced in moist seasons and years generally being wider, more numerous and thinner walled than those produced in dry periods . Xylem is often functional for multiple years , and so current drought responses can affect water transport during future drought. The production of protective molecules typically drops soon after normal water potential is restored . However, transcriptional and physiological ‘memory’ in stomatal guard cells has been observed, with stressed plants maintaining smaller stomatal apertures when re-watered . There may also be ‘legacy effects’ on NSC production and traits such as growth and xylem anatomy . Plants that quickly return to normal could gain a growth advantage. In areas in which recurring drought is common, however, we hypothesize that this memory effect reduces mortality risk. There are multiple traits involved at different stages of the drought response . Stomatal control and patterns of root and shoot growth affect the degree to which a plant avoids drought stress. These traits plus xylem morphology, protective molecule production, changes in carbohydrate metabolism and pathogen defenses influence drought resistance. Finally, the recovery rate of photosynthesis and other processes, the degree of persistent changes in structure and the ability to refill xylem affect drought resilience. In the next two sections, we first review the methods used to date to examine genetic controls on ecologically important traits, and then explore how these methods have been and can be leveraged to test for genetic variation in, and identify the genetic basis of, the traits and processes addressed above.
Gene expression or transcriptome studies examine changes in the amount of RNA transcripts to identify genes that are upregulated or down regulated under different conditions. Changes in the amount of a gene product can result in different phenotypic responses, even if all individuals have the same gene sequence. Such changes are responsible for plasticity, and may involve temporary or heritable epigenetic modifications . Gene expression studies may involve a variety of techniques, but most recent studies have used microarray chips – DNA probes to which cDNA or RNA hybridize, resulting in flfluorescence – or cDNA sequenc ing . The latter avoids the need for probe and microarray design and can survey whole novel transcriptomes . Real-time quantitative polymerase chain reaction is highly sensitive, but is most often used to target specific candidate genes or to confirm a subset of expression changes . All techniques are sensitive to which tissues are sampled at what time . Moreover, unless expression responses in different genotypes or populations are explicitly compared, this approach does not address local adaptation.Provenance or common garden studies, where seedlings from many different sources are planted in a common environment, began to reveal heritable differences between tree populations long before the availability of genetic marker data . Provenance studies established in the mid-20th century to identify seed zones for replanting or highly productive genotypes have been re-purposed to investigate potential responses to climate change . Many recent studies have also used seedling common gardens . Studies conducted across multiple sites, or incorporating multiple treatments, can estimate the plasticity of traits, allowing the fitting of transfer functions that predict performance based on source and planting environments . However, such studies do not reveal which genes are responsible for observed differences unless paired with other techniques. It should be noted that there is usually substantial variation within tree populations . The third set of approaches can be used to investigate the causes of heritable variation between populations and individuals.These approaches aim to identify genes or genomic regions related to a trait or to adaptation along environmental gradients. QTL studies are a classic way to identify the loci involved in continuous trait variation. However, although QTLs for a number of traits have been identified in trees, this approach has had limited success for a variety of reasons, many of which are reviewed in Gonzalez Martınezet al.. For instance, a great deal of time and space is needed to cross parental tree lines and raise a sufficient sample size of progeny. Conifers also have very large genomes with low linkage disequilibrium and, without enough genetic markers avail able, most QTLs are undetectable . In addition, high-resolution genetic/physical maps or positional cloning is needed to identify causal genes/mutations . By contrast, genome scan and association studies make use of large numbers of newly available markers , and are carried out in highly diverse out crossing natural populations . Genome scans identify loci that differ more or less between populations than expected by chance . For instance, outlier Fst values can be used to infer the type of selection: balancing selection results in low Fst and shared alleles, and divergent selection in high Fst with segregated alleles.