Water stress physiology and crop modeling of spring safflower under different irrigation management strategies

Increased pressure on irrigation water resources coupled with periodic drought in the Southern High Plains has increased the need for drought tolerant crops and irrigation management strategies in the region. Overall aim was to understand how a deep rooting desert originated crop like safflower fits in the Southern High Plains where irrigation is applied by spraying water frequently. We were also interested whether any management strategy like preseason irrigation improves its adaptability. The objectives of this study were to: (1) understand drought physiology and yield formation of spring safflower cultivars under different irrigation management practices, (2) study water extraction patterns of spring safflower under different irrigation management practices, (3) adapt the Decision Support System for Agrotechnology Transfer (DSSAT) Cropping System Model (CSM-CROPGRO) to simulate growth and seed yield of spring safflower. To achieve these objectives, two studies were conducted at Agricultural Science Center (ASC) of New Mexico State University, Clovis, NM.;In the first study, drought physiology and yield formation of two diverse spring safflower cultivars were assessed under different irrigation levels with or without preseason irrigation. One half of the experimental blocks received preseason irrigation of 164 in 2012 and 153 mm in 2013 to refill the soil profile utilized by the previous crops, while the other half remained depleted. Five in-season irrigation levels (I1 to I5) ranging from 88 to 392 mm in 2012 and from 83 to 373 mm in 2013 were imposed on both preseason irrigation and no-preseason irrigation blocks. Higher leaf water potential (Psi1) was observed under increased water availability either by preseason irrigation or by higher in-season irrigation level in safflower during two observation dates in both years. Osmotic potential at full turgor (Psipi100), photosynthesis rate (P n) and transpiration rate (Tr) decreased with a reduction in Psi1 under water stress conditions. The relative water content (RWC) was affected only by the in-season irrigation levels in both years. The preseason irrigation increased seed yield of safflower by 39 and 118% over no-preseason irrigation in 2012 and 2013, respectively. A gradual increase in seed yield was observed with an increase in irrigation levels; and the highest irrigation level, Is increased seed yield by 85 and 171% over the lowest irrigation level, I1 in 2012 and 2013, respectively. Seed yield increased with increase in Pn, plant biomass, number of heads per plant, and number of seeds per head but not with 1000-seed weight under increased water availability. Overall, increased availability of water through preseason irrigation or through in-season irrigation levels improved safflower physiology and yield formation.;The soil moisture data from the first study was used to assess the water extraction patterns and water use efficiency (WUE) of spring safflower (Appendix A). A significant water extraction was observed in preseason and no-preseason irrigation treatments at each depth in both years. However, preseason irrigation treatment increased water extraction over no-preseason irrigation treatment at all depths. The total water extraction of preseason irrigation treatment was 35 to 49 mm greater compared to no-preseason irrigation treatment, which significantly increased evapotranspiration (ET), WUE and oil yield of preseason irrigation over no-preseason irrigation treatment. Irrigation levels and cultivars did not affect the soil water extraction depth in both years. Safflower in different irrigation levels relied on moisture from the upper depths during early growing season and explored moisture from the deeper depths during late growing season. The total water extraction was increased in I1 treatment compared to Is treatment in both years. The WUE did not differ among irrigation levels in 2012; but in 2013, irrigation levels higher than I 3 had greater WUE. The highest and lowest ET and oil yield were recorded in I5 and I1 treatments respectively; however, oil yield of 15 treatment did not differ from I4 treatment. Water use (ET) of cultivar PI8311 was significantly lower compared to cultivar S333, while WUE and oil yield of PI8311 were higher or equal to higher in both years. Preseason irrigation and cultivar PI8311 are promising to improve WUE and oil yield in safflower; however, excessive irrigation through in-season irrigation was not always efficient in increasing WUE.;In the second study, drought physiology, yield formation and water extraction patterns of spring safflower cultivars were assessed under growth stage based irrigation management. A process based crop model (DSSAT) was also adapted to simulate growth and seed yield of spring safflower. Three cultivars (99OL, P18311 and Nutrisaff) were grown under four irrigations treatments [fully irrigated (FI), stress at vegetative stage (VS), stress at reproductive stage (RS) and dryland (DL)]. Measurements of water potential (Psi1) indicated that irrigation treatments imposed water stress on safflower. Safflower responded to stress by regulating stomata and osmotic adjustment. Water stress significantly reduced photosynthesis (Pa), leaf area index (LAI) and light interception in all stress treatments, resulting in decreased biomass accumulation. Seed yield of safflower was strongly related to biomass production (R 2 = 0.61). Among yield forming traits, heads per plant were more influential in yield formation than seeds per head and 1000-seed weight. Compared to the FI treatment, RS treatment reduced seed yield by 19% and 20% in 2013 and 2014, respectively; while the same reduction in VS treatment was 25% and 22%, respectively. From the results, it seems like withholding irrigation during reproductive stage (RS treatment) is less detrimental to safflower yield compared to withholding irrigation during vegetative stage (VS treatment). However, more research is needed to support our findings.;The soil moisture data supported our results. All irrigation treatments or cultivars extracted water to a depth of 1.6 m in both years. The maximum soil water extraction in each irrigation treatment was observed from 1.0 to 1.6 m soil layer. The average total water extraction from planting to harvest was highest in RS treatment compared to FI, VS and DL treatments. The highest evapotranspiration (ET) and oil yield were recorded in FI treatment; while the lowest values of these traits were found in DL treatment. The RS treatment showed the second highest ET and oil yield after FI treatment. The numerically higher WUE of RS treatment compared to FI and VS treatments suggest that RS treatment utilized the available water more efficiently. The cultivar 99OL showed higher WUE and oil yield due to higher water extraction compared to other cultivars. Results suggest practicing deficit irrigation at the reproductive stage of safflower along with adoption of 99OL cultivar to receive maximum WUE and oil yield benefits under limited water conditions; and the resulting decrease in oil yield might be 19% with saving of 28% irrigation water. Future research under different environments is also needed to refine these findings.;In the crop modeling component, the CROPGRO template approach was used, and parameters in species and cultivar files were developed based on safflower literature and calibration to field data. The entered base temperatures for photosynthetic, vegetative and reproductive processes of safflower ranged from 0 to 5 °C while corresponding optimum temperatures varied from 19 to 40 °C. Simulated results were compared with observed data collected from field experiments conducted at Clovis, NM, USA, during summers of 2013 and 2014. The model predicted the crop life cycle (anthesis and harvest maturity date) with relative root mean square error (RRMSE) of 0.07. Average plant biomass, head mass, head number and seed number were satisfactorily simulated when compared to observed values. Seed yield, averaged over irrigation treatments and years, was predicted as 1963 kg ha-1 compared to measured value of 1902 kg ha-1 with RRMSE of 0.12. Reasonable prediction of phenology, growth and yield by the model adapted for safflower suggested that the CROPGRO-safflower model is promising to simulate safflower production in semi-arid climates. However, further testing of the CROPGROsafflower model under different environments is needed.