معیارهای اقتصادی برای تخمین استفاده از منابع فعلی و آینده با تمرکز بر برداشت آب

Demand for, and use of, water, food and energy resources is already straining mostly non-renewable supplies (e.g. freshwater resources). Population growth, climate change and shifting socio-economic standards mean that these resources will be stretched to critical levels. This paper uses historical data of GDP and GDP-per-capita to identify correlations with 19 water, food and energy metrics (e.g. water withdrawals, total food production and electricity generation and consumption), and subsequently attempts to estimate plausible resource use and demand for a suite of seven GDP growth scenarios, focusing on water resources. It is shown that GDP-per-capita is weakly correlated with all metrics, however total GDP shows stronger correlation. Best-fit regressions and the statistical distributions of historical data were used to replicate the historical data and validate derived correlations. Following this, the GDP scenarios were used to estimate plausible global total water withdrawals, food production and electricity generation and consumption to 2100. If recent GDP growth is maintained, then the ‘safe planetary boundary’ for total freshwater withdrawal of 4000 km3  yr−1 is surpassed in c. 29% of simulations (out of 8700 simulations). However, in scenarios of global GDP shrinkage, the ‘safe’ limit was exceeded in only 14% of simulations. Estimates reported here for total water withdrawals agree closely with similar estimates from the literature, which tend to use more complex methods and models. Similarly, the food production and electricity generation totals can be related to available viable agricultural land or potential CO2 emissions respectively (e.g. through assumptions of future energy generation methods). It is hoped that this work will contribute to the growing global policy and development debate on resource use and sustainable development, particularly through the provision of probabilistic estimates for resource use and by using open-data sources and a relatively simple methodology.