A data-driven prediction model for Fennoscandian wildfires

Wildfires are recurrent natural hazards that affect terrestrial ecosystems, the carbon cycle, climate and society. They are typically hard to predict, as their exact location and occurrence are driven by a variety of factors. Identifying a selection of dominant controls can ultimately improve predictions and projections of wildfires in both the current and a future climate. In this study, we applied a data-driven machine learning approach to identify dominant hydrometeorological factors determining fire occurrence over Fennoscandia, and produced spatiotemporally resolved fire danger probability maps. A random forest learner was applied to predict fire danger probabilities over space and time, using a monthly 2001–2019 satellite-based fire occurrence dataset at a 0.25° spatial grid as the target variable. The final data-driven model slightly outperformed the established Canadian fire weather index (FWI) used for comparison. Half of the 30 potential predictors included in the study were automatically selected for the model. Shallow volumetric soil water anomaly stood out as the dominant predictor, followed by predictors related to temperature and deep volumetric soil water. Using a local fire occurrence record for Norway as target data in a separate analysis, the test set performance increased considerably. This improvement shows the potential of developing reliable data-driven prediction models for regions with a high quality fire occurrence record, and the limitation of using satellite-based fire occurrence data in regions subject to small fires not picked up by satellites. We conclude that data-driven fire prediction models are promising, both as a tool to identify the dominant predictors and for fire danger probability mapping. The derived relationships between wildfires and its compound predictors can further be used to assess potential changes in fire danger probability under future climate scenarios.

Development of Google Earth Engine Fire Weather Index Calculator for Indonesian Fire Danger Rating System

Tropical peatlands make up almost ten percent of the land surface in Indonesia, making peat fires detrimental not only for global atmospheric carbon levels, but also to public health and socioeconomic activities in the region. Indonesian Fire Danger Rating System (FDRS) was developed based on the Canadian Forest Fire Weather Index System (CFFWIS), using three different fuel codes and three indices representing fire behaviour. Daily Fire Weather Index (FWI) calculation is done by the Meteorological Climatological and Geophysical Agency (BMKG) with data from its synoptic weather stations network. Distribution of such weather stations are sparse, therefore this paper reports on the development of Fire Weather Index calculator on Google Earth Engine, using high resolution weather data, provided by weather model and remote-sensing open datasets. The resulting application is capable of generating daily maps of FWI components to be used by the Indonesian Fire Danger Rating System.

Fire prediction with logistic regression on territory of Bosnia and Herzegovina

The Canadian Fire Weather Index system [1] has been used worldwide by many countries as classic approach in fire prediction. It represents system that account for the effects of fuel moisture and weather conditions on fire behaviour. It numerical outputs are based on calculation of four meteorological elements: air temperature, relative humidity, wind speed and precipitation in last 24h. In this paper meteorological data in combination with Canadian Fire Weather Index system (CFWI) components is used as input to predict fire occurrence using logistic regression model. As logistic regression is a supervised machine learning method it’s based on user input in the form of dataset. Dataset is collected using NASA GES DISC Giovanni web-based application in the form of daily area-averaged time series in period of 31.7.2010 to 31.7.2020, it’s analysed and pre-processed before it is used as input for logit model. CFWI components values are not imported but calculated in run-time based on pre-processed meteorological data. As a result of this research windows application was developed to assist fire managers and all those involved in studying the fire behaviour.

Fire Danger Harmonization Based on the Fire Weather Index for Transboundary Events between Portugal and Spain

Portugal and Spain have a cross-border cooperation protocol on wildfires response for a buffer strip of 25 km for each side of the border. In spite of the success of this collaboration, there are issues to be improved, since Portuguese and Spanish authorities use different methodologies to assess the daily fire danger. A methodology to harmonize fire danger and its interpretation by the Portuguese and Spanish Civil protection authorities in the transboundary buffer strip area is hereby presented. The fire danger index used is the Canadian Fire Weather Index (FWI), which requires input from meteorological data and gives an indication of fire intensity. The fire danger class is an important decision support tool for preventing and fighting wildfires. Since the meaning of FWI values change from region-to-region according to its specific characteristics, a calibration process was performed based on statistical data of the daily FWI values, the number of fires and burned area between 2005 and 2013. The results of the FWI calibration and harmonization of the data for the five danger classes minimizes the fire danger discrepancies across the border. This methodology has the potential to be reproduced in other areas.

Weather Factors Associated with Extremely Large Fires and Fire Growth Days

Abstract“Megafires” are of scientific interest and concern for fire management, public safety planning, and smoke-related public health management. There is a need to predict them on time scales from days to decades. Understanding is limited, however, of the role of daily weather in determining their extreme size. This study examines differences in the daily weather during these and other, smaller fires, and in the two sets of fires’ responses to daily weather and antecedent atmospheric dryness. Twenty fires of unusual size (over 36 400 ha), were each paired with a nearby large fire (10 100 to 30 300 ha). Antecedent dryness and daily near-surface weather were compared for each set of fires. Growth response to daily weather was also examined for differences between the two sets of fires. Antecedent dryness measured as the Evaporative Demand Drought Index was greater for most of the fires of unusual size than it was for smaller fires. There were small differences in daily weather, with those differences indicating weather less conducive to fire growth for the unusually large fires than the smaller fires. Growth response was similar for the two sets of fires when weather properties were between 40th and 60th percentiles for each fire pair, but the unusually large fires’ growth was observably greater than the smaller fires’ growth for weather properties between the 80th to 100th percentiles. Response differences were greatest for wind speed, and for the Fosberg Fire Weather Index and variants of the Hot-Dry-Windy Index, which combine wind speed with atmospheric moisture.

ESTIMATING FINE DEAD FUEL MOISTURE CONTENT UNDER EQUATORIAL CLIMATE CONDITIONS

The measurement of the fine dead fuel moisture content (FDFMC) is extremely important for forest fire prevention and suppression activities, as it has a great influence on the ignition probability and fire behavior. The Fine Fuel Moisture Code (FFMC) from the Fire Weather Index (FWI), is one of the most used models to estimate the FDFMC. Nevertheless, studies that assess the efficiency of this model in Brazil or in low latitude regions are rare. The present study aimed to evaluate the efficiency of the FFMC in an equatorial climate area and to develop a new model capable of estimating the FDFMC with greater precision. For this purpose, 861 random samples of fine dead fuel had their moisture content determined through oven drying. The obtained values were compared with those estimated by the FFMC and correlated with meteorological parameters to build a regression model. The results obtained show that the FDFMC was overestimated by the FFMC. The independent variables with the greatest influence on the FDFMC were, in decreasing order of significance: air relative humidity, air temperature, amount of rainfall in the last 24 hours and number of days without rainfall. The developed model presented good statistical parameters (r2 = 0.86; p <0.0001; RMSE = 0.22) and can be used, in areas with similar characteristics of the study area, to estimate the daily fire risk and to determine ideal conditions for prescribed burns.

Updated Operational Implementation of the Canadian Forest Fire Weather Index System in Ireland

&lt;p&gt;Since 2006 the Canadian Forest Fire Weather Index System (FWI) has been used operationally at Met &amp;#201;ireann to predict the risk of forest fires in Ireland (Walsh, S, 2006). Although only around 11% or ca 770,000 ha of the total land area of Ireland is afforested, there are also large areas of open mountain and peatlands that are covered in grasses, dwarfshrub and larger woody shrub type vegetation which can provide ready fuel for spring wildfires, when suitable conditions arise. Following winter, much of this vegetation is either dead or has a very low live moisture content, and the flammability of this vegetation can be readily influenced by prevailing weather, most especially following prolonged dry periods. The Department of Agriculture, Food and Marine is the Forest Protection authority in Ireland and issues Fire Danger Notices as part of this work. These notices permit improved preparedness for fire responses and are based on information provided by Met &amp;#201;ireann on the current status of FWI and FWI components using observation data at synoptic stations and the predicted FWI for the next five days ahead based on numerical weather prediction input data.&lt;/p&gt;&lt;p&gt;The FWI is based on&lt;/p&gt;&lt;ul&gt;&lt;li&gt;three different types of forest fuel, ie how quickly these dry out/get rewetted. These are the Fine Fuels Moisture Code (FFMC), the Duff Moisture Code (DMC) and the Drought Code (DC).&lt;/li&gt; &lt;li&gt;components based on fire behaviour: the Initial Spread Index (ISI), the Build-up Index (BUI), and the Fire Weather Index (FWI) which represents fire intensity as energy output rate per unit length of fire front. It is then used to determine the Daily Severity Rating (DSR) of the fire danger.&amp;#160;&lt;/li&gt; &lt;/ul&gt;&lt;p&gt;Of these components, the FFMC and ISI components have been found to provide the most accurate indication of risk under Irish conditions, based on the fuels involved and ignition patterns observed to date.&lt;/p&gt;&lt;p&gt;The DSR was based on a climatology of 1971 to 2005 at the time of operational implantation of the FWI at Met &amp;#201;ireann. An updated climatology based on the new reference period of 1990 to 2020 will be shown as well as the change of the 98 percentiles of extreme rating using this new reference period. &amp;#160;&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Walsh, S.&lt;/strong&gt;&amp;#160;&amp;#8220;Implementation in Ireland of the Canadian Forest Fire Weather Index System.&amp;#8221; In&amp;#160;&lt;em&gt;Making Science Work on the Farm. A Workshop on Decision Support Systems for Irish Agriculture&lt;/em&gt;, 120&amp;#8211;126. Dublin: AGMET, 2007.&amp;#160;&lt;/p&gt;

Climate change impacts on the Greek tourism sector

&lt;p&gt;In this study we examine the impacts of climate change on the tourism sector using a number of tailored climate indicators assessing whether climate conditions are suitable for touristic activities such as the Tourism Climate Index -focusing on outdoor activities- and the Beach Climate Index -focusing on beach activities- as well as fire danger indicators such as the Canadian Fire Weather Index, focusing on forest fire risk. To this aim daily or sub-daily data for a number of meteorological variables from a large ensemble member of Regional Climate Models from the EURO-CORDEX data base are used. The data cover the period 1971-2100 under three RCP emissions scenarios, namely RCP2.6, RCP4.5 and RCP8.5. The analysis is performed for three periods, the 1971-2000 which is used as a reference period and two future periods, the 2021-2050 and 2071-2100. The results indicate that the robust warming projected on a seasonal basis, under all three climate scenarios, drives the changes on all indicators examined. Regarding the climate suitability indicators for tourism the results indicate a lengthening of the tourist season suitable climate conditions while for the fire danger indicators, an increase in the number of days with high and very high fire danger conditions is projected. The most pronounced changes are evident towards the end of the century and under the RCP8.5 future emissions scenario. This study is performed in the framework of CLIMPACT, a&lt;strong&gt; &lt;/strong&gt;Greek national funded project which aims&lt;strong&gt; &lt;/strong&gt;to immediate integration, harmonization and optimization of existing climate services and early warning systems for climate change-related natural disasters in Greece, including supportive observations from relevant national infrastructure.&lt;/p&gt;&lt;p&gt;&amp;#160;&lt;/p&gt;