Genevieve Tolhurst

In recent decades, Arctic near-surface temperatures
have risen at approximately twice the rate of the global
average. Arctic ice-loss has
accelerated polar warming more than lower latitudes in
what is called ‘Arctic amplification’.
Investigation into the effects of Arctic amplification has
become of great scientific and public interest catalysed
by recent extreme weather events in the Northern
Hemisphere’s mid-latitudes. Using past weather
records and satellite imagery to measure ice extent, ,
statistical analyses and mathematically modelling can
evaluate proposed mechanisms linking mid-latitude
weather to Arctic amplification.

Wildfire is a curious natural hazard; unlike other natural hazards such as hurricanes, and earthquakes, the physical size and intensity of fires may be influenced by human intervention. People can create fire and use it as a tool to manage their environment. In the past, people have successfully controlled and suppressed wildfires. This essay will discuss how human perceptions of nature influence the outcomes of disasters. Using Victoria’s 2009 Black Saturday fires as an example, the socio-ecological model of resilience will be evaluated in its application during Black Saturday. Human perceptions of fire change how society responds to the threat of fire in landscape management, property hazard reduction, and community preparation and response.

This essay will discuss how people exacerbate the impact of environmental disturbances to create disasters. We will consider governments’ attitudes and priorities; government disaster management policies; and societies’ entrenched values and inequalities. We will explore the human failures that led to Victoria’s 2009 Black Saturday bushfire disaster.  This essay will draw on recent global disasters such as Hurricane Katrina 2005, Hurricane Mitch 1998, the Bangkok floods of 2011, and the
2004 South Asia Boxing Day tsunami to illuminate the issues of Black Saturday.

Australia is a fire prone landscape which has been subjected to wild and planned fire for millennia.  The European settlement of Australia changed fire regimes in the landscape by imposing European landscape values to very different ecosystems. Prescribed burning can provide key aspects of the natural disturbance regime under human management (Boer et al 2009). In the last century, land managers have learnt the importance of prescribed burning to protect people, property, and to regenerate fire-prone ecosystems. This review will examine the historic fire regime in Australia, especially Mediterranean-climate regions, and the precedent for fire from Aboriginal management. It will broadly include contrasting management values of Aborigines, early Europeans, conservationists and foresters. Prescribed burning must be used effectively and efficiently to protect things we value; a summary of the literature which discusses these issues will be presented. Finally, limitations and suggested improvements for prescribed burning practices will be discussed and evaluated.

This is an exercise in SAS using a simulated database.

An Australian retailer “Homely Goods” has recently launched an incentive to encourage website users to register their details. Registered users receive a credit of 5% of their first purchase as a registered user which can be used for a future purchase. Registered users are expected to spend more than non-registered users. The aim of this investigation is to find if there is a difference in the behaviours of registered users compared to non-registered users.

This report uses MS Excel to
analyse crash data on Victorian roads to evaluate the extent to which alcohol contributes to the road toll. In
particular, it will look at the proportion of people involved in alcohol-related crashes as opposed to nonalcohol-
related cases; the incidence of fatalities in either case; the regional representation of alcohol-related
fatalities; and finally, the weekly periodicity of alcohol-related fatalities.

This report intends to classify three types of wine by thirteen of their physical and chemical properties.
The classification was done using K Nearest neighbour (KNN) and decision tree classifiers from sklearn in python. It was found that the wines could be classified with a
classification error rate of 0.057.
We found that the three wines were well classified by their physical and chemical properties, and decision tree classifier performed best with the data.