Exploratory project MIRRORS (2021 - 2023)

Predicting the response of plants exposed to chronic thermal stress

Climate change is characterised not only by variable and extreme intensities of the main climatic factors but also by an increased frequency of extreme events, such as heat waves, which are highly detrimental to field crop yields and harvest quality.

Background and challenges

In this context, improving predictions of plant performances under repeated heat stress scenarii is a major challenge. The MIRRORS project is based on the hypothesis that the effect of a succession of stressful events is not equivalent to the sum of the individual effects of each event.  Indeed, when plants have been exposed by an initial stress, their responses to subsequent stresses can be determined by this prior event as a consequence of  a "memory effect" (which can be either penalising or beneficial).

In order to improve predictions of plant performance in repeated stress situations that are expected to occur more frequenlty, the MIRRORS project proposes methods and tools for generic predictions of the response of plants subjected to repeated thermal stress in particular.

Goals

We propose the following approaches, based on existing data sets for rapeseed and sorghum (1):

PEERSIM replication model
  1. Analyse the non-additive nature of the effects of heat stress events using complete datasets (climatic variables and plant performance criteria).
  2. Identify agro-climatic indicators or specific thermal sequences related to the memory of heat stress.  We will identify particular thermal scenarii, with recurrent patterns, and associate them with the observed plant performances (grain yield  and seedquality criteria).
  3. Then, two complementary approaches to modelling the effects of repeated thermal stresses will be developed, on both rapeseed and sorghum: (i) "concept-driven" based on the implementation of predictive ecophysiological models parameterised on these two species, in order to take these memory effects into account, and (ii) "data-driven" guided by data mining methods with no mechanistic a priori.

 

Contacts :

Units involved and partnerships

INRAE participants

AgroEcosystem division

Expertise

UMR EVA

Ecophysiology, agronomy, modelling.

UMR AGAP

Ecophysiology, statistical analysis and data mining.

UMR ISPA

Ecophysiology, modelling, biogeochemistry.

Partners

Laboratoire Lorrain de Recherche en Informatique et ses Applications

Knowledge discovery, modelling.

Modification date : 18 September 2023 | Publication date : 21 March 2022 | Redactor : Com