Visualization of Water Services in Africa: Data Applications for Nexus Governance - Governing the Nexus: Water, Soil and Waste Resources Considering Global Change - page 198

support decision-making related to flood protection, where a finer scale is

required to account for soil or vegetation heterogeneity (e.g. needed for esti-
mation of water in

ltration capacity).

'
representations. An example of an abstract symbolic map showing relationships are
cartograms with a gridded surface. Each grid cell area is resized according to their
value (e.g. high number of water resources equal to large grid

A large number of further visualization options is available for ' realistic world

field area size), where
all grid cell areas are resized relative to each other. The method was developed and
publicized in the Worldmapper project (Sasi Group and Newman 2014 ).

3D - 4D Visualization

The increasing needs of spatial understanding have led to the development of 3D
data visualizations (e.g. volumetric models, discrete models and continuous mod-
els) (Lin et al. 2013 ). The number of software tools that create [e.g. Paraview
(paraview.org)] and display [Google Earth (Google), Layerscape (Microsoft
Research), World Wind (NASA), Skyline ( http://skylineglobe.com ) ] 3D data and
objects are increasing. These tools can illustrate three-dimensional geological fea-
tures (e.g. groundwater systems), topography, urban structures as well as heights of
natural land cover features (e.g. forest height). The illustration is often supported by
a remote sensing natural colour (RGB) image overlay. Some emerging web-based
spatial visualizations include data analysis results such as areas of global risk maps
or interactive statistical graphs (Fig. 8 ).
To get better insights into feedback systems (cause-effect relations) of varying
processes as well as between compartments, more dimensional time series (4D)
visualization tools are required. The understanding of feedback loops is important
for decision-makers, since each action seldom causes only desired responses, but
also undesired side effects. For instance, the extraction of river water for agriculture
purposes might increase yields and local income, but unsustainable water use can
decrease water availability downstream, which leads to an increase in drought risk,
decrease of yields and increase in hunger. Thus, the knowledge of feedback loops is
needed to choose appropriate decisions and to adjust them effectively. A visuali-
zation that aims at a full system interaction representation needs to be supported by
model simulations and prediction (Kwakkel et al. 2014 ). Models enable the sim-
ulation of complex relationships and processes based on mathematical equations as
well as on data of different sources. The integration of these data along with known
mathematical relationships allows us to give predictions about future system
behaviour under speci

ed conditions. For instance, Fig. 6 , shows the usage of
remote sensing data as input for a hydrological modeling framework (LIS

—

Land
Information System) that might simulate the actual water balance and future water
availability changes due to climate change (e.g. decrease of precipitation). Milewski
et al. ( 2009 ) described an approach to use remote sensing data as input for a
hydrological model to estimate run-off and recharge in an arid environment. With