Beatriz González Pérez

METHODS FOR QUANTIFYING FIRE SEVERITY IN SHRUBLAND-FIRES

Methods are presented to relate temperature-residence-time at the soil surface, i.e., time above 150 ºC as an estimation of the severity of a fire, with measurements made during an experimental fire or on two post hoc measurements. The experiment was carried out in a shrubland dominated by the woody-legume Cytisus striatus subsp. eriocarpus, in Central Spain. Temperature-sensitive paints, and steam-releasing open-calorimeters were used as fire-meters during the burn. Post hoc measurements used were estimations of heat output per unit area, and measurements of the minimum diameter of branches of Cytisus remaining after the fire. Time above 150 ºC was obtained from measurements made with thermocouples placed at the soil surface in 20 contiguous 1×1 m squares of the burn plot. All other measurements were made at each 1×1 m in the 22×3 m rectangle surrounding, and including, the thermocouple squares. Various simple and multiple regression models were constructed to predict time above 150 ºC from each of the four measurements made during or after the fire. Maximum coefficients of determination obtained for regressions were 0.61 and 0.62 for water mass loss from open-calorimeters and branch diameter, respectively. Using all the variables in a multiple regression model, time above 150 ºC was related to water mass loss from open-calorimeters and heat output per unit area with a coefficient of determination of 0.77. It is concluded that estimations of time above 150 ºC at the soil surface during the passage of fire may be possible based on simple devices, such as open-calorimeters, or on biological indicators, such as minimum branch diameters. Additionally, combining two methods (open-calorimeters, estimations of heat output per unit area) may allow the reconstruction of the time above 150 ºC during the fire at a scale of 1 2, an important characteristic of a burn to understand ecosystem response to fire.

Fire-type and forestry management effects on the early postfire vegetation dynamics of a Pinus pinaster woodland

The objective of this research was to study the effects of type of fire, prefire-, and postfire-management on the postfire vegetation dynamics of a Pinus pinaster woodland in Central Spain, burned at 15 yr of age. The effects of type of fire (crown-, or surface-fire), prefire-management (thinning out of trees and clearing of brush or no such actions) and postfire-management (removal of burned trees one year after the fire or no such action) on the postfire vegetation were studied during the first three years after the fire. Herbaceous plant abundance, species richness, and diversity, as well as abundance, growth and density of the dominant shrub species (Cistus ladanifer) were measured during the first three years after the fire. Our results show that the effects of the type of fire on the vegetation were minimal. Prefire-management effects were significant on the abundance of herbaceous species, mainly during the second and third year after fire, in particular for the Leguminosae species. Prefire managed areas were more diverse in species, and produced higher plant biomass than unmanaged areas. Postfire-management effects on the shrubs and herbs were minimal, except for the Leguminosae, which increased their cover where the trees had been removed. Plant dynamics were marked by the interaction between prefire-management and fire-type through the dynamics of the shrub cover. On most occasions, plots that resulted in lower cover of C. ladanifer had greater abundance of herbaceous plants and, in particular, of the Leguminosae. In general, our results show that irrespective of fire-type, prefire-, or postfire-management all areas tended to be very similar in their vegetation three years after the fire.

Bayesian Analysis of Contingency Tables

ABSTRACT The display of the data by means of contingency tables is used in different approaches to statistical inference, for example, to broach the test of homogeneity of independent multinomial distributions. We develop a Bayesian procedure to test simple null hypotheses versus bilateral alternatives in contingency tables. Given independent samples of two binomial distributions and taking a mixed prior distribution, we calculate the posterior probability that the proportion of successes in the first population is the same as in the second. This posterior probability is compared with the p-value of the classical method, obtaining a reconciliation between both results, classical and Bayesian. The obtained results are generalized for r × s tables.

APROXIMACIÓN BAYESIANA APLICADA AL REPARTO MODAL EN MODELOS DE TRANSPORTE DE MERCANCIAS. CASO PRÁCTICO: CORREDOR FERROVIARIO BIOCEÁNICO CENTRAL)

La Planificacion de los Transportes de una determinada zona necesita herramientas que permitan representar o modelizar, de forma analitica, la situacion actual y futura de su red de infraestructuras. De esta manera, se pueden estimar, para diferentes escenarios, los flujos de pasajeros, vehiculos o mercancias que habra en dicha red. Una de estas herramientas son los denominados Modelos de Demanda, de entre los que destaca el modelo clasico de 4 etapas (Generacion-Atraccion, Distribucion, Reparto o eleccion modal y Asignacion). El presente articulo muestra una novedosa aproximacion a la tercera etapa, que es la correspondiente a la decision del modo de transporte a la que se enfrenta la unidad de decision dentro de un conjunto discreto de alternativas. Tradicionalmente, los modelos de eleccion discreta han sido los mas utilizados para esto, aunque existen otros metodos, como las redes neuronales, que tambien han sido utilizadas por otros autores. Las redes bayesianas se proponen como herramienta alternativa para la eleccion modal, tanto para pasajeros como para mercancias, y para este ultimo caso, se presenta un caso de estudio basado en el Corredor Ferroviario Bioceanico Central en Sudamerica. Los resultados obtenidos permiten comparar un modelo logit y una red bayesiana, y muestran como la aproximacion bayesiana surge como una herramienta prometedora en este tipo de aplicaciones. Palabras Clave: Modelos de Transporte, Modelo de Eleccion Discreta, Logit, Red Bayesiana.

An explanatory analysis of electricity prices within day-ahead Spanish energy market by using a graphical automatization with R

Since 1998, the Spanish and Portuguese Administrations began to share a common path in building the Iberian Electricity Market (MIBEL). This cooperation has been very successful, not only for its contribution to the existence of an electricity market on an Iberian level, but also on a European scale, as a significant step in building the Internal Energy Market. The price of electricity in the MIBEL is very changeable. This creates a lot of uncertainty and risk in market actors. Due to continuous changes in demand and marginal price adjustment, buyers and sellers can not know in advance the evolution of prices. Our interest is to study of this uncertainty since the perspective of the buyer and not the sellers perspective. The aim of this work is to develop a graphical analysis of the variables involved in the Spanish Energy Market in order to explain the electric price and provide to small traders, that could be interested in participating in that market, better knowledge of the schedule. On the other hand, large industrial consumers use this information to design strategies to optimize its production capacity and improve their production costs. In this article the variable of interest is the marginal price instead of demand. This paper provides a graphical analysis by means of an easily reproducible automatization with the R project for statistical computing that allows to explore, visualize and understand the key variables that define its final value. Mibel 2011 and 2012 data are used for ilustrations. The results show the importance of the calendar effect, seasonality and trend as principal factors to take into account for posterior fases: modeling and forecasting.