Task 1 - Inventory and modeling
Duration: 12 months
Start date: March 19th 2019
Person * months: 42
Researchers involved: Adrian Pascual Arranz; Américo Mendes; Brigite Botequim;
Carlos Caldas; Marco Marto; Marlene Marques; Miguel Sottomayor; Susete Marques
Consulting: Margarida Tomé and José Pereira
This task aims at characterizing the case study area and projecting its inventory over time under scenarios of climate change. It aims at producing the information needed to define the decision space to be used by Task 3 and to develop wildfire behavior models in Task 2. For that purpose, it will require computing power and is structured in four subtasks.
1 .1 Data acquisition, treatment and validation
It encompasses the collection of:
geographical, environmental and fuel cover type data to characterize the study area and to assess the impact of wildfires (wildfire perimeters in the period 2012-2017);
inventory data from 200 plots measured in 2012;
inventory data to be measured in unburnt plots;
Daily meteorological data from local weather station; d
data from country-level permanent plots measured at least twice.
It will proceed as follows: firstly, by overlay of wildfire perimeters with the inventory plots in the Vale Sousa Joint Management Area (ZIF_VS) to identify burned plots. Secondly, by carrying out an inventory to update the stand descriptors and fuel types in the 2012 plots and by acquiring additional data to support wildfire simulation (e.g. canopy base height (CBH), canopy bulk density (CBD) and canopy cover (CC)). Thirdly, by identifying the current land use in burned plots with remote sensing technologies (e.g SENTINEL-2A and Pléiades imagery), this will be further validated by field visits to these plots. Fourthly, by characterizing local fire weather (temperature, humidity and wind speed/direction). This subtask will develop a database with a logical structure adequate to store all relevant data from ZIF_VS that is needed to develop all Tasks.
1.2 Process based modeling
It encompasses the calibration of maritime pine process-based models to project the ZIF_VS forest inventory over time under climate change scenarios. Bayesian calibration will generate an estimate of posterior parametric uncertainty for the process-based forest growth model 3-PG. This subtask will expand of the current knowledge base of process- based models (it includes already a 3PG calibration for the eucalypt). The use of process-based models will be instrumental to simulate the decision space under climate change scenarios in 1.3.
1.3 Simulation of decision space
It encompasses the projection of growth of the ZIF_VS forest stands, using silviculture models and the process-based model knowledge base. We will consider three IIASA climate change scenarios (Forsell et al 2016): Reference, where climate change mitigation is minimal, resulting in 3.7 degrees of temperature increase and EU bioenergy and Global bioenergy associated to 2.5 º and 1.75º temperature increases at the end of the planning horizon, respectively. This sub-task will generate the stand-level prescriptions and the forest ecosystem management decision space, to be considered in other Tasks.
1.4 Fire simulators inputs
It will encompass: a) assignment of fuel cover types to the stand-level forest ecosystem prescriptions, from a set of customized fuel models for Portugal conditions and b) deﬁnition of an extreme “wildfire conditions” scenario, based on meteorological data collected in 1.1, to run the wildfire simulation in Task 2. Further, fuel moisture content for high fire risk season will be computed. This subtask will derive topographic and fuel map layers (SH, CBH, CHB and CC), with information on Portuguese custom fuel models distribution under climate scenarios.