EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Innovative physiological indicators for drought stress in banana

Mathilde Vantyghem1,2,4, Roel Merckx1, Rebecca Hood-Nowotny2, Bert Stevens3, Christian Resch4, Gruber Roman4, and Gerd Dercon4
Mathilde Vantyghem et al.
  • 1Division of Soil and Water Management, Faculty of Bioscience Engineering, KU Leuven, Belgium
  • 2Institute of Soil Research, Department of Forest and Soil Sciences, University of Natural Resources and Life Sciences Vienna, Austria
  • 3Laboratory of Tropical Crop Improvement, Faculty of Bioscience Engineering, KU Leuven, Belgium
  • 4Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture, IAEA, Austria

Banana is a very important crop in East-Africa, serving as a staple for millions of smallholder farmers. Aside from pests and diseases, lack of water is the major constraint to production. Climate change is expected to aggravate these problems, creating a need for improved resilience and better management practices. A major obstacle to the development and evaluation of such practices is the difficulty to measure drought stress in the field. In this research, we investigate physiological parameters that can provide information on drought stress in banana under field conditions. We evaluate the use of stable carbon isotope ratios (δ13C) and leaf temperature as indicators for stress, the former ones not well-established for banana. Leaf temperature is known to increase under drought stress due to stomatal closure. The existing methods to measure leaf temperature are however expensive and their use is limited to small greenhouse set-ups. In this research, we employ an infrared thermometer (±1°C) for temperature measurement under field conditions. The experimental set-up consists of a banana field trial with a blocked design (irrigated and rainfed treatments) in the Kilimanjaro region, Tanzania. Leaf samples for isotope analysis were taken from mature plants (mother plants) and the main on-growing sucker (daughter plants) in August 2019, during the dry season. Leaf temperature was monitored throughout the day. Results show significantly higher δ13C ratios in rainfed plants, compared to irrigated ones, indicating more drought stress. Within both groups, mother plants have higher δ13C ratios than daughter plants. At dawn, leaf temperature was similar for all treatments. During the day, rainfed banana plant leaf temperature increased 7°C more than in their irrigated counterparts. Daughter plants remained cooler than mother plants in both treatments. Leaf temperature and δ13C showed a strong correlation. While carbon isotope signatures are a known proxy, our results suggest that leaf temperature is a an easily measurable indicator of drought stress as well. The infrared thermometer is cheap, convenient to use in the field and provides in-situ information. Leaf temperature has an enormous potential as a drought stress sensor in banana, as well as in other plants. Our research will further optimize both methods for drought stress evaluation. This will facilitate management comparisons in the future as well as variety screening, eventually contributing to more resilient banana production systems.

How to cite: Vantyghem, M., Merckx, R., Hood-Nowotny, R., Stevens, B., Resch, C., Roman, G., and Dercon, G.: Innovative physiological indicators for drought stress in banana , EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-4844,, 2020

Comments on the presentation

AC: Author Comment | CC: Community Comment | Report abuse

Presentation version 1 – uploaded on 04 May 2020
  • CC1: Comment on EGU2020-4844, Sarah Garré, 05 May 2020

    Hi Mathilde, nice experiment. In the presentation, you report maximal differences between irrigation and rainfed and between mother/daughter. Could you elaborate a bit on the variability of the measurements too?

    • AC1: Reply to CC1, Mathilde Vantyghem, 05 May 2020

      Yes, of course. The variability in tempererature is not very large. Yet, I have to say we measured 20 plants per irrigation treatment and developmental stage. Importantly, we measured left and right side of the midrib. These can vary in temperature, since the leaves droop a bit, and therefore one side could be sun exposed and the other side shaded. The standard error per treatment-developmental stage is arround 0.4-0.7 °C, being higher at noon. Much smaller than the difference between groups.

      In terms of δ13C, the variability is limited as well. The standard error per treatment-dev stage is arround 0.3-0.4 ‰, based on 7 plants. We did take several samples per plant, and these values are based on the average per plant. Again, fairly smaller error than the differences between groups. 

      In general, I would say both temperature and δ13C differences are large enough to be used as a proxy for drought stress, but the question still is, which levels of stress can be measured. These measurements were done one month and a half after the last rains, so in fairly dry conditions. The next step will be to apply progressive water shortage and monitor both parameters at increasing levels of stress. 

      • CC2: Reply to AC1, Sarah Garré, 05 May 2020

        thanks mathilde!