Plant-water relationships for several common bean genotypes (Phaseolus vulgaris L.) with and without drought stress conditions

Abstract

This research presents results related to plant-water relationships under drought and non-drought stress conditions for several common bean (Phaseolus vulgaris L.) genotypes, including the local and most-widely planted variety in Puerto Rico, and new genotypes known to be drought tolerant. The genotypes evaluated were: ‘Morales’ which is currently the most popular white-seed bean variety in Puerto Rico, and with unknown drought response, BAT 477 cream-seed, which was one of the first genotypes released with drought tolerant characteristics, SER 21 and SER 16 red-seed, and SEN 3 and SEN 21 black-seed, which are germplasm released by CIAT (Centro Intenacional de Agricultura Tropical, Colombia) with drought tolerant characteristics. The experiments were conducted in a greenhouse environment in Mayagüez, Puerto Rico in the Tropical Agricultural Research Station (TARS) facilities, and a field environment at the University of Puerto Rico, Experiment Station at Fortuna in Juana Diaz, Puerto Rico. During the three year study, a total of eight experiments were conducted, five in the greenhouse and three under field conditions, during 2005, 2006 and 2007. Under greenhouse conditions, no statistical differences in the stomatal resistance (rL) and leaf temperature (TL) were observed among genotypes without drought stress conditions. Statistical differences were observed in both, with moderate and strong drought stress. The genotypes with the lowest increases in rL and TL under high drought stress conditions were, in the following order: SER 21 and SEN 3. The genotypes BAT 182 477 and Morales both exhibited the highest rL and TL. Under field conditions, without drought stress, no statistical differences were observed in rL and TL among genotypes, however, with drought stress the genotypes BAT 477 and Morales were statistical different from the others and the genotypes SER 16 and SER 21 showed the lowest values of rL and TL. With respect to the water status measured with the relative water content (RWC) in the greenhouse experiments, statistical differences were observed among genotypes. The genotype SER 21 exhibited the highest values and BAT 477 the lowest values, indicating the high capability of SER 21 to conserve water under strong stress conditions. The poor response of BAT 477 to drought stress in these experiments could be associated with the high leaflet size and high total leaf area. The genotypes with the greatest reduction in the leaf area under drought stress was SER 21, and also this genotype showed the highest water use efficiency (WUE) and harvest index (HI) values. The genotype Morales can be considered to have some degree of drought tolerance, based on its response under moderate drought stress conditions, in field and greenhouse environments. Crop coefficient were derived following the methodology proposed by the Irrigation and Drainage Papers (FAO-24 and 56) for two genotypes (SER 16 and Morales) during two years of experiments. The crop coefficient (Kc) derived in this study were lower than those reported by the Irrigation and Drainage Paper No. 56 (Allen et al. 1998) due to several factors, such as: different atmospheric demand, low plant density especially for SER 16, and the irrigation system used (drip) that reduced significantly the soil evaporation. Also the Kc was estimated indirectly measuring the fraction of soil covered by vegetation (fc) or with the cumulative growing degree days (CGDD). In addition to drought stress, high wind speed contributed to stress. The genotype most susceptible at high wind conditions was Morales compared with SER 16. This 183 susceptibility under windy conditions generated an inverse relation between stomatal resistance and the aerodynamic resistance (i.e., with increased wind speed, the aerodynamic resistance decreased while the stomatal resistance increased). The genotype SER 16 under field conditions with 6 plants.m-2 do not show statistical differences in the seed yield with respect to Morales with 13 plants.m-2, but statistical differences were observed in biomass and pods, indicating a yield compensation phenomenon, that is highly desirable under limited water conditions. SER 16 exhibited lower cumulative evapotranspiration rates, and higher WUE and transpiration efficiency values than Morales. The critical variable in the generalized Penman-Monteith (GPM) methodology is the surfaces resistance (rs) which is a function of the stomatal resistance (rL) and the leaf area index (LAI). A disadvantage in applying the GPM method is the necessity to directly measure rL and LAI, which are difficult and time consuming. In this study, the GPM method with the measured rs was referred to as “Measured”. We also considered other methods for estimating rs based on the latent heat flux (λE), such as as the “inverse of the GPM model”, the vertical gradient “Szeicz and Long method” (Szeicz and Long,1969), the “ET station” (Harmsen et al. 2006), and the micrometeorologicallybased method of Ortega-Faria et al. (2004), which depends on net radiation (Rn), vapor pressure deficit (VPD), soil heat flux (G) and the change in soil moisture. These results indicated that the ET can be estimated directly using the GPM method if the rs is appropriately parameterized. We found that rs could be reliably estimated based on the method recommended in the Drainage and Irrigation Paper (FAO56) when: a) the LAI is greater than 1.0. Conversely, if the LAI is less than 1.0, this indicates all of the leaf area is contributing to transpiration and not just the effective area (i.e., LAI x 0.5). b) In this study, the inverse of the GPM method performed poorly for large values of the aerodynamic resistance (ra), causing rs to increase, which is contrary to the physiological response, and subsequently under estimating the ET. c) the Szeicz and Long method and the ET station predicted correctly the rs with LAI values over 1.0, but 184 did not work well under high drought stress conditions (Figure 5.8), and d) the OrtegaFarias et al.(2004) method estimated appropriately the rs with and without drought stress if soil moisture is correctly estimated in the root zone. Under windy and drought stress conditions, the rs estimation was not appropriate for any of the methods for the genotype Morales, which is stomatally susceptible under windy conditions. The upper and lower baselines for the crop water stress index application (CWSIIdso et al. 1981) were developed for these genotypes, indicating genotypic variations in the baselines. The drought tolerant genotypes showed higher upper baselines, and the rate of change (slope) in the lower baseline was also higher in the most drought susceptible genotypes. The upper and lower baselines, in this study were different than those previously reported for common beans, indicating also the environmental and genotypic variability. The CWSI was well related with the water content in the root zone. When the soil reached the field capacity, the CWSI for both genotypes was between 0.1 and 0.2, which has been previously reported for other common bean genotypes, and also the maximum relative yield under greenhouse conditions corresponded for this range of CWSI. The CWSI also detected the “physiological stress” induced by windy conditions in the genotype Morales under field conditions.