Statistical Downscaling
Statistical downscaling techniques are reliable to correlate the cardinal and sensitive
climate variable maximum temperature and precipitation obtained from GCMs at
coarse resolution and fine-scale spatial resolution for impact assessment at the
regional scale (Ahmed et al. 2013). Statistical downscaling, despite having limitations (Wangsoh et al. 2017), is a practical technique requiring relatively less
computational effort than dynamic downscaling (Tripathi et al. 2006); it finds its
major applications for the assessment of climate change on the themes of water
resources in response to hydrological modeling. Weather classification, weather
generators, and transfer functions are among the prominent subcategories (Anandhi
et al. 2008). Wilby et al. (1998) compared and contrasted various statistical downscaling methods.
Among the sophisticated methods under statistical downscaling techniques, the
change factor (CF) method is a simple and linear spatial downscaling method for
impact analysis studies widely accepted for rapid assessment of climate change. The
CF method simulated daily times series data of temperature and precipitation and
further compared the performance of simulation in three mountainous basins of the
United States (Hay et al. 2000). The CF method was used to compare the bias
corrections-based methods for downscaling projections of extreme flow indices to
quantify the impacts from floods (Hundecha et al. 2016). Owing to its ease of
implementation to assess the impact of climate change at hydrological scale (Boé
et al. 2007) and enabling the representation of multiple GCMs and emission scenarios with minimal computing effort, CF estimates values of climate variables at future
time scales and at spatial scales that are appropriate for regional climate change
impact assessment (Anandhi et al. 2011). The difference in these projections from
future and historical simulations are considered the means of simple addition/
multiplication or scaling the mean climatic CF to each baseline observation datum
(Fowler et al. 2007). This method is also termed constant scaling as the changes in
the different percentile of climate variables (such as precipitation) in magnitude are
assumed equal. Scaling approaches are simple and consider multiple GCMs for the
downscaling (Wang et al. 2017).
Change Factor
The change factor (CF) method is a perturbation method based on the assumptions of
relative changes in future and historical climate projections by GCMs even though
the GCM is biased. Additive change factor (ACF) and multiplicative change factor
(MCF) are the classifications for assessing temperature and precipitation (Akhtar
et al. 2008). ACF calculates the arithmetical difference between a GCM variable
derived from a current climate simulation and from a future climate scenario
considered at the same GCM grid location, as shown in Eq. 2.1:
26
N. S. Patil and R. S. Laddimath
Statistical downscaling techniques are reliable to correlate the cardinal and sensitive
climate variable maximum temperature and precipitation obtained from GCMs at
coarse resolution and fine-scale spatial resolution for impact assessment at the
regional scale (Ahmed et al. 2013). Statistical downscaling, despite having limitations (Wangsoh et al. 2017), is a practical technique requiring relatively less
computational effort than dynamic downscaling (Tripathi et al. 2006); it finds its
major applications for the assessment of climate change on the themes of water
resources in response to hydrological modeling. Weather classification, weather
generators, and transfer functions are among the prominent subcategories (Anandhi
et al. 2008). Wilby et al. (1998) compared and contrasted various statistical downscaling methods.
Among the sophisticated methods under statistical downscaling techniques, the
change factor (CF) method is a simple and linear spatial downscaling method for
impact analysis studies widely accepted for rapid assessment of climate change. The
CF method simulated daily times series data of temperature and precipitation and
further compared the performance of simulation in three mountainous basins of the
United States (Hay et al. 2000). The CF method was used to compare the bias
corrections-based methods for downscaling projections of extreme flow indices to
quantify the impacts from floods (Hundecha et al. 2016). Owing to its ease of
implementation to assess the impact of climate change at hydrological scale (Boé
et al. 2007) and enabling the representation of multiple GCMs and emission scenarios with minimal computing effort, CF estimates values of climate variables at future
time scales and at spatial scales that are appropriate for regional climate change
impact assessment (Anandhi et al. 2011). The difference in these projections from
future and historical simulations are considered the means of simple addition/
multiplication or scaling the mean climatic CF to each baseline observation datum
(Fowler et al. 2007). This method is also termed constant scaling as the changes in
the different percentile of climate variables (such as precipitation) in magnitude are
assumed equal. Scaling approaches are simple and consider multiple GCMs for the
downscaling (Wang et al. 2017).
Change Factor
The change factor (CF) method is a perturbation method based on the assumptions of
relative changes in future and historical climate projections by GCMs even though
the GCM is biased. Additive change factor (ACF) and multiplicative change factor
(MCF) are the classifications for assessing temperature and precipitation (Akhtar
et al. 2008). ACF calculates the arithmetical difference between a GCM variable
derived from a current climate simulation and from a future climate scenario
considered at the same GCM grid location, as shown in Eq. 2.1:
26
N. S. Patil and R. S. Laddimath