Artificial intelligence techniques for the detection of animals in forest


The article can be downloaded here:


An artificial vision software that contain image processing and deep learning functions has been developed to detect animals on videos recorded by camera traps in forests. Monitoring by camera traps is used on the OPTMix experimental facility in the Orléans forest to estimate wild ungulate pressure (roe deer, red deer and wild boar) on study plots. Scientists and conservation biologists use camera traps to monitor wildlife populations and biodiversity. Automatic image processing that count animals and identify species could facilitate and improve the use of camera traps in biodiversity monitoring programs and make the methodology accessible to a greater number of end users.

OPTMix annual meeting, you can download the presentations

The presentations made during the OPTMix annual meeting on January 9 and 10, 2020 are available by clicking on the download link below for a period of 30 days:

OPMTix annual meeting, 9, 10 and 13 january 2020

The annual meeting of the project will take place on 9, 10 and 13 January, 2020 at Irstea (INRAE) Nogent-sur-Vernisson.

French version of the meeting annoucement and full program for the 9 and 10 of january


Article published in “Notre Forêt”, the journal of Regional Forest Ownership Center Ile-de-France Centre-Val de Loire

The journal “Notre Forêt” of the Regional Forest Ownership Center Ile-de-France Centre-Val de Loire has just published an article presenting the OPTMix device and the first results obtained about the effects of tree mixing on tree growth, water resource, mineral nutrition and biodiversity.

Korboulewsky, N., P. Balandier, Y. Dumas, M. Gosselin, A. Marell and T. Perot (2019). « OPTMix scrute la mixité en forêt. Qu’apporte le mélange d’essence face aux changements globaux ? » Notre Forêt 88(septembre): 4.

The full article is available here

Article published: Stand growth and structure of mixed-species and monospecific stands of Scots pine (Pinus sylvestris L.) and oak (Q. robur L., Quercus petraea (Matt.) Liebl.) analysed along a productivity gradient through Europe

An article has just been published in the journal “European Journal of Forest Research” about stand growth and structure of mixed and monospecific stands of Scots pine and Sessile oak. This work was based on plots set up along a productivity gradient in Europe. OPTMix is one of the plots used in this study.

Pretzsch, H., M. Steckel, M. Heym, P. Biber, C. Ammer, M. Ehbrecht, K. Bielak, F. Bravo, C. Ordóñez, C. Collet, F. Vast, L. Drössler, G. Brazaitis, K. Godvod, A. Jansons, J. de-Dios-García, M. Löf, J. Aldea, N. Korboulewsky, D. O. J. Reventlow, A. Nothdurft, M. Engel, M. Pach, J. Skrzyszewski, M. Pardos, Q. Ponette, R. Sitko, M. Fabrika, M. Svoboda, J. Černý, B. Wolff, R. Ruíz-Peinado and M. del Río (2019). « Stand growth and structure of mixed-species and monospecific stands of Scots pine (Pinus sylvestris L.) and oak (Q. robur L., Quercus petraea (Matt.) Liebl.) analysed along a productivity gradient through Europe. » European Journal of Forest Research. doi: 10.1007/s10342-019-01233-y


Past failures of monocultures, caused by wind-throw or insect damages, and ongoing climate change currently strongly stimulate research into mixed-species stands. So far, the focus has mainly been on combinations of species with obvious complementary functional traits. However, for any generalization, a broad overview of the mixing reactions of functionally different tree species in different mixing proportions, patterns and under different site conditions is needed, including assemblages of species with rather similar demands on resources such as light. Here, we studied the growth of Scots pine and oak in mixed versus monospecific stands on 36 triplets located along a productivity gradient across Europe, reaching from Sweden to Spain and from France to Georgia. The set-up represents a wide variation in precipitation (456–1250 mm year−1), mean annual temperature (6.7–11.5 °C) and drought index by de Martonne (21–63 mm °C−1). Stand inventories and increment cores of trees stemming from 40- to 132-year-old, fully stocked stands on 0.04–0.94-ha-sized plots provided insight into how species mixing modifies stand growth and structure compared with neighbouring monospecific stands. On average, the standing stem volume was 436 and 360 m3 ha−1 in the monocultures of Scots pine and oak, respectively, and 418 m3 ha−1 in the mixed stands. The corresponding periodical annual volume increment amounted to 10.5 and 9.1 m3 ha−1 year−1 in the monocultures and 10.5 m3 ha−1 year−1 in the mixed stands. Scots pine showed a 10% larger quadratic mean diameter (p < 0.05), a 7% larger dominant diameter (p < 0.01) and a 9% higher growth of basal area and volume in mixed stands compared with neighbouring monocultures. For Scots pine, the productivity advantages of growing in mixture increased with site index (p < 0.01) and water supply (p < 0.01), while for oak they decreased with site index (p < 0.01). In total, the superior productivity of mixed stands compared to monocultures increased with water supply (p < 0.10). Based on 7843 measured crowns, we found that in mixture both species, but especially oak, had significantly wider crowns (p < 0.001) than in monocultures. On average, we found relatively small effects of species mixing on stand growth and structure. Scots pine benefiting on rich, and oak on poor sites, allows for a mixture that is productive and most likely climate resistant all along a wide ecological gradient. We discuss the potential of this mixture in view of climate change.


New measurements of soil water content with a neutron probe

On 1 and 2 July 2019, new measurements of soil water content were carried out by Cyril Dejean (Irstea G-EAU unit in Montpellier) on the OPTMix plots using a neutron probe. This work was carried out under intermediate soil moisture conditions in order to complete the range of measurements required to calibrate the soil water content sensors (CS616 – Campbell Scientific) installed in the OPTMix plots (see also here).


Set up of 144 automatic band dendrometers

144 automatic band dendrometers (DRS26 SDI12 sensor, Environmental Measuring Systems) are being installed in low and medium density plots. These sensors will allow us to follow trees response to drought and other stressful events (pathogen attack for example) and to compare the species response (sessile oak and Scots pine) according to stand density and stand composition.

On each plot, the sensors were connected to the data logger with a single cable and the data are recorded every 30 minutes (see below for an example on a Scots pine plot).





UR SOLS research unit came to visit OPTMix on June 24, 2019

30 people from INRA’s UR SOLS ( research unit came to visit the OPTMix experimental plots on June 24, 2019. After a general presentation by Philippe Balandier, the scientific studies were presented in little workshops:

  • Water resource and sensors (Jordan Bello and Camille Couteau);
  • Tree growth and biodiversity (Thomas Pérot and Marion Gosselin);
  • Understorey dynamics, ungulates and soil (Anders Marell and Yann Dumas).



Article published: Effects of wild ungulates on birds’ nesting failure

An article has just been published in the journal “Forest Ecology Management” about the direct and indirect effects of wild ungulates (roe deer, red deer and wild boar) on forest birds’ nesting failure. The study was conducted as part of a Master 2 internship by Amanda Cocquelet in connection with the COSTAUD research project (Contribution des Ongulés Sauvages au foncTionnement des écosystèmes et AUx services rendus à ChamborD). In comparison with the first study site (Chambord), the predation rate was particularly low on the OPTMix site such as wildlife attendance.

Cocquelet, A., A. Mårell, S. Bonthoux, C. Baltzinger and F. Archaux (2019). « Direct and indirect effects of ungulates on forest birds’ nesting failure? An experimental test with artificial nests. » Forest Ecology and Management 437: 148-155. doi: 10.1016/j.foreco.2019.01.025


Increasing deer and wild boar populations in North America and Western Europe are suspected of threatening low-nesting forest birds. Ungulates may directly affect these birds by predating eggs or chicks, or indirectly attract other mammalian or avian predators or reduce nesting site availability and habitat quality. To test some of these various mechanisms in two French forests, 528 artificial nests, 77 of which (14.6%) were monitored with camera traps, were set up in 44 plots which were either unfenced or fenced to respectively include or exclude ungulates. We used generalized linear mixed models to relate the nest failure rate to indices of plot frequentation by ungulates and other nest predators, to vegetation structure and to local bird community richness indices. Nest failure rate was significantly higher in the unfenced plots (39%) than in the fenced plots (32%) but this was related neither to red deer nor to wild boar frequentation of the plots. Furthermore, levels of frequentation for other mammalian and avian predators tended to be positively correlated to ungulate presence but not to nest failure rate. Nest failure rate depended on both nest height and shrub cover: nests on the ground (31% failure rate) were less predated or disturbed than nests in low shrubs (41% failure rate). Nests surrounded by low shrub cover experienced a slightly higher predation risk. We conclude that ungulates have a limited influence on nest failure for low-nesting forest birds; in particular, they seem to contribute very little to egg predation compared to other predators.

Article published: Transmitted light as a tool to monitor tree leaf phenology and development applied to Quercus petraea

An article has just been published in the “Agricultural and Forest Meteorology” journal on how transmittance could be used to monitor tree leaf phenology and development for sessile oak (Quercus petraea).

Perot, T., P. Balandier, C. Couteau, S. Perret, V. Seigner and N. Korboulewsky (2019). “Transmitted light as a tool to monitor tree leaf phenology and development applied to Quercus petraea.” Agricultural and Forest Meteorology 275: 37-46.


Better knowledge and a more complete long-term monitoring of tree species phenology and tree foliage development are crucial to accurately estimating the impact of climate change on forest ecosystems functioning and on forest species distribution. We set up global solar radiation sensors under the forest canopy and continuously recorded solar radiation in order to follow the development of sessile oak (Quercus petraea L.) foliage in pure stands over four consecutive years and for two levels of stand density. At the same time, we made phenological observations to determine bud burst date. One of our goals was to link observed bud burst dates and transmittance measurements. Our results show that solar radiation transmittance during the leaf unfolding period followed a sigmoid-shaped curve; and that it was possible to fit a generalized logistic model to determine a set of parameters characterizing tree foliage and its development during the unfolding period. Among these parameters, we suggest that the date at which solar radiation interception by foliage reaches 50% could be used to monitor the beginning of the growing season over the long term. The relationship between observed bud burst date and the transmittance model parameters was more complex than we expected. We tested several indices to deduce the date of bud burst from transmittance. We showed that when radiation interception due to foliage reached 10%, this indicated the observed bud burst date. Finally, a linear model including parameters from the generalized logistic models of transmittance as predictors explained 57% of the variability in bud burst date. Complementary research combining light measurements and phenological observations over a longer period of time is needed to know whether these relationships could help estimate bud burst date regardless of the year or the stand density.