CROSS Harmonization & HPC modelization of FOREST Datasets
The pilot project CAMBrIc – CAlidad de la Madera en Bosques mIxtos – focus on simulation of different management scenarios to generate a database on wood quality in pure and mixed forests in Spain. Following taxa have been selected and analyzed: Pinus sylvestris, Pinus pinaster, Pinus nigra, Fagus sylvatica, Quercus pyrenaica and, Quercus robur, pure and mixed stands. Input data were extracted from the Spanish National Forest Inventory (NFI) database, where these species were predominant. Growth and yield models have been selected from the relevant literature, and four new mixed stands have been developed, and all of them coded in SIMANFOR (https://www.simanfor.es – Bravo et al 2020), and setup in Caléndula HPC.
The following management alternatives, proposed following the schemes defined by Duncker et al (2012), have been run:
(1) Passive: Unmanaged forest nature reserve;
(2) Low: Close-to-nature forestry,
(3) Medium: Combined objective forestry,
(4) High: Intensive even-aged forestry and
(5) Intensive: Short-rotation forestry.
These management schemes, simulated across a gradient of forest mixtures (from pure to multi-species stands) for the target species, allow to classify timber products.
Simulations were run in a much shorter time than expected, thanks to Caléndula HPC array job, which allow to split independent task and execute them simultaneously. This means that we can split the input datasets in individual plots, and run all plot inside the whole dataset in the time you run only one plot in a normal PC. SIMANFOR output will be available after project finished, into the DSI of the project.