The current drought affecting southern and eastern Australia is profoundly affecting Sydney’s water supply. A desalination plant was developed in response to the Millennium drought to reduce pressure on water supply during future droughts; however it has provided only limited relief. Increasing population, a warming climate, land surface changes, and extensive agricultural use all increase water demand and potentially reduce catchment runoff. It is necessary to understand how Sydney water supply might be changing in order to better manage water resources and lower the associated risks of severe water shortages. This study aims to better understand Sydney water supply by analysing precipitation and temperature trends across the Sydney catchment. A decreasing trend in annual precipitation is apparent across the catchment. Annual precipitation is also significantly less variable, due to a reduction in years recorded above the 80^th percentile. These changes are the result of significant reductions in precipitation during both spring and autumn, especially over the last 20 years. Wavelet analysis is applied to assess how the influence of climate drivers has changed over time. Preliminary attribute selection was carried out using various linear regression and machine learning techniques including artificial neural networks, random forests and support vector regression.