Romain Bertrand

Centre for Biodiversity Theory and Modelling
Station d’Ecologie Théorique et Expérimentale du CNRS
09200 Moulis, France

Phone: +33 5 61 04 05 83

Research interests

I am particularly interested by the impact of global changes on species distribution and productivity, and communities. Until now my work is mainly focused on the forest ecosystem.
I investigated the velocity of plant community reshuffling in response to the recent climate warming [3], the importance of non-climatic factors (such as soil acidity), ontogeny and seed dispersal to predict actual and future distribution of tree species [1,5], the monitoring of environmental conditions by using floristic assemblage [2,3] and the ontogenetic niche shift that occur along the edaphic dimension among tree species [4]. I also modeled the current potential distribution of French forests habitats defined in the European Habitats Directive for the 2013 assessment of the Natura 2000 network by a direct and indirect modeling approach (focus on habitat and species, respectively), and modeled French climate conditions over the period 1960-2010 [5].

Currently, I'm studying :
  • the mitigating or amplifying effects of abiotic (e.g. precipitation, light availability, pH and N-availability changes) and biotic (e.g. habitat fragmentation, niche evolution and thermic tolerance) factors on the response of forest plant communities to the recent climate warming.
  • the effect of the recent climate change on the productivity-biodiversity relationship in the forest ecosystem, and also the potential stabilizing effect of biodiversity on the forest productivity.
  • the relationship between ecological niche and phylogeny among a large set of forest plant species.

Key results

Changes in forest plant communities lag behind climate warming
Climate change is driving latitudinal and altitudinal shifts in species distribution worldwide, leading to novel species assemblages. I reported that forest plant communities had responded to 0.54°C of the ef-fective increase of 1.07°C in highland areas (500–2600 m above sea level) over the period 1965-2008, while they had responded to only 0.02°C of the 1.11°C warming trend in lowland areas. These partial compensations demonstrate the current occurrence of a climatic debt in forest vegetation caused by cli-mate warming. The high climatic debt observed in lowland forest compared to the highland (3.1 times higher in lowland areas) is likely due to extinction and migration debts caused by a wider temperature tolerance of plants in lowland communities and a plant migration capacity (i) limited by the high spatial fragmentation of the lowland forest habitat and (ii) shorter than the shift of thermal conditions in lowland areas, respectively. The ecological and geographical specificity of lowland and highland forests led me to consider two different threats induced by climate change: (i) biotic attrition in lowland areas, and (ii) loss of a specific and an historic biodiversity in highland forest (mountaintop extinction of alpine species due the surface decrease of their potential habitat) likely replaced by more common species assemblage (in-crease of generalist plants).

Figure: Comparison of floristically (green) and climatically (red) reconstructed temperature trends between 1965 and 2008 in lowland (a) and highland (b) forest areas [3]. One explanation of the higher lag observed in lowland forest plant communities is a likely marked migration debt in this area due to a plant migration capacity shorter than the shift of thermal conditions which is also limited by the high fragmentation of lowland forests (c, d). Panels c and d show the minimal distance between analog temperature conditions in 1965-1986 and 1987-2008 across the French forest territo-ry and along the elevation gradient, respectively [5]. To conserve the same temperature conditions a species needs to migrate poleward and upward over 34.4 and 5.7 km in the French lowland and highland forests, respectively.

Edaphic dimension, seed dispersal and ontogeny are crucial factors to predict plant distribution

Most of the studies have used only climate variables to predict future potential species range shifts and have omitted some important abiotic and biotic factors. Based on the simRShift model, I assessed the importance of the edaphic dimension, ontogeny and seed dispersal to predict the species distribution of Quercus pubescens over the 21st century. Adding the edaphic dimension to the climate-only species distribu-tion model (SDM) substantially improved the niche-space definition of Q. pubescens, highlighting an in-crease in species tolerance in confronting climate constraints as the soil pH increased. Spatially, future predictions over the 21st century showed that disregarding the edaphic dimension in SDM led to an over-estimation of the potential distribution area, an underestimation of the spatial fragmentation of this area, and prevented the identification of local refugia, leading to an underestimation of the northward shift capacity of Q. pubescens and its persistence in its current distribution area. Spatial discrepancies between climate-only and climate-plus-edaphic models are strengthened when seed dispersal, age of first repro-duction and forest fragmentation are accounted for in predicting a future species distribution area [1,5].

Figure: State of the potential distribution areas of Quercus pubescens in 2100 simulated from climate-only (a) and cli-mate-plus-edaphic (b) SDMs [1]. Panels c and d show the shifts of the Southern and Northern edge of the Q.pubescens distribution simulated in 2100 from different scenarios of seed dispersal distance (c) and age of first reproduction (d) (negative and positive values of edge shift reveal a southward and northward shifts, respectively) . An increase in the seed dispersal distance and in the age of first reproduction led to a northward expansion and a contraction (mainly from the northern edge) of the future potential distribution of Q. pubescens, respectively.

simRShift is a semi-mechanistic and semi-stochastic spatially explicit model developed in the environment to simulate species range shifts. simRshift is mainly based on habitat suitability for species, but also takes into account ontogenetic niche shifts (juvenile and reproductive niche), mortality, seed dispersal, age of maturation,…[5] I still develop the model and currently I’m working to extend simRShift at the communi-ty scale and to add a module to simulate species productivity. A stable version with a graphic interface will be soon available.

Ecological niches converge towards richer nutritional conditions over tree ontogeny

Niche changes during a species’ lifespan are known as ontogenetic niche shifts. These shifts reflect changes in resource availability, requirements, organisms’ foraging ability, and/or size-dependent biotic interactions. In the plant kingdom, however, this issue remains poorly covered. I showed that some gen-eral changes in the ecological niche optimum and amplitude occurred over ontogeny among the main temperate tree species [4]. The tree stage was found to occur mainly at higher nutrient availability than the seedling (+16.3% on the nutritional gradient) or sapling (+11.1%) stages. In addition, nutritional niches of tree species exhibited, successively, a niche enlargement in eutrophic conditions and a niche restriction in oligotrophic conditions during growth. These global nutritional niche shifts observed over the species’ lifespan contributed moderately but significantly to the niche separation in temperate tree communities (up to 4.5%). We interpreted niche shifts as a response to an increase in nutritional re-quirements over ontogeny, leading to an intra-specific selection where individuals established in eu-trophic soils have the maximal fitness. Biotic interactions and temporal changes in the environment may secondarily enhance or counteract the process. Ontogenetic niche shifts require consideration in the study of species autecology and plant community organization, especially in a context of global changes where they can restrict their ability to track the shift of their potential habitat caused by environmental changes [4,5].

Figure: Main nutritional niche shifts over the ontogeny of temperate tree species [4]. Arrows represent shifts leading to the nutritional niche in the next life-history stage. Arrow thickness varies with the intensity of the niche shift. P = probability of presence.

Spatio-temporal database of yearly and monthly climate conditions over the period 1960-2010

During my doctorate, I produced (in collaboration with C. Piedallu, LERFoB) a set of French spatial grids of yearly and monthly temperature and precipitation conditions over the period 1960-2010 from an efficient modeling approach relying meteorological observations to topographic and geographic variables [3,5]. The model performance is really appropriated for ecological studies (R² and RMSE average over the period = 0.889 and 0.04°C, and 0.724 and 1.1mm for precipitation). Please contact me for more information and request.

Figure: Spatial projections of the temperature (a, R²=0.856 and RMSE=0.04°C) and precipitation (b, R²=0.856 and RMSE=0.9mm) models for August 2003 (one of hottest and driest month over the last 100 years in France).


- 2013 - 2016: Post-doctorate fellowship at the Centre for Biodiversity Theory and Modelling (CBTM, USR2936 CNRS, Moulis, France). Outline: Biodiversity and ecosystem functioning in a changing environment. Supervisor: Dr. Michel Loreau (CNRS).

- 2011 - 2012: Post-doctorate fellowship (TEECH project) at the Laboratoire d’Etude des Ressources Forêts-Bois (LERFoB, UMR1092 AgroParisTech-INRA, Nancy, France). Outline: Modelling of the current potential distribution of French forest habitats defined in the European Habitats Directive for the 2013 assessment of the Natura 2000 network. Supervisor: Dr. Damien Marage (AgroParisTech).

- 2011: Engineer at LERFoB (UMR1092 AgroParisTech-INRA). Outline: Spatiotemporal modelling of climatic conditions in France over the period 1960-2010. Supervisor: Dr. Christian Piedallu (AgroParisTech).

- 2008 - 2012: Ph. D. in forest ecology at LERFoB (UMR1092 AgroParisTech-INRA). Outline: Spatio-temporal response of the forest vegetation to climate warming: assessment of the vegetation reshuffling and characterisation of the effect of ecological and geographical factors modulating this process at the species and community scales. Supervisor: Prof. Jean-Claude Gégout (AgroParisTech).

Awards and distinctions

- 2014: Young scientist prize (SFE). Read more.


  • Riofrío-Dillon G., Gegout J.-C. and Bertrand R. (2017) — Decreasing before increasing: Evolution of nitrogen availability conditions in French forest ecosystems over the last century according to forest herbs. bioRxiv. Download
  • Legrand D., Larranaga N., Bertrand R., Ducatez S., Calvez O., Stevens V. and Baguette M. (2016) — Evolution of a butterfly dispersal syndrome. Proceedings of the Royal Society B - 283: 20161533. Download

  • Bertrand R., Riofrio-Dillon G., Lenoir J., Drapier J., de Ruffray P., Gegout J.C. and Loreau M. (2016) — Ecological constraints increase the climatic debt in forests. Nature Communications. 7: 12643. Download - Article recommended by F1000

  • Bertrand R., Perez V. and Gégout J.-C. (2012) — Disregarding the edaphic dimension in species distribution models leads to the omission of crucial spatial information under climate change: the case of Quercus pubescens in France. Global Change Biology 2648-2660. Link

  • Riofrío-Dillon G., Bertrand R. and Gégout J.-C. (2012) — Toward a recovery time: forest herbs insight related to anthropogenic acidification. Global Change Biology 18, 3383-3394. Link

  • Bertrand R., Lenoir J., Piedallu C., Riofrío-Dillon G., de Ruffray P., Vidal C., Pierrat J.-C. and Gégout J.-C. (2011) — Changes in plant community composition lag behind climate warming in lowland forests. Nature 479, 517-520. Link

  • Bertrand R., Gégout J.-C. and Bontemps J.-D. (2011) — Niches of temperate tree species converge towards nutrient-richer conditions over ontogeny. Oikos 120, 1479-1488. Link


Other papers are available here.

Thesis:Bertrand R. (2012) — Spatio-temporal response of the forest vegetation to climate warming: assessment of the vegetation reshuffling and characterisation of the effect of ecological and geographical factors modulating this process at the species and community scales.Ph.D. thesis, AgroParisTech, Nancy, 305 pp. [pdf]

Recently passed Ph.D. students: (2010 - 2013) Co-supervisor of Gabriela Riofrío-Dillon (LERFoB, UMR1092 AgroParisTech-INRA). Outline: Evolution of the acidity and nitrogen availability in the French forest soils over the 20th century - A spatiotemporal and multiscale approach based on the bioindicator character of plants. Director: Prof. Jean-Claude Gégout (AgroParisTech).