CONCLUSIONS AND RECOMMENDATIONS - Integrating GIS, Remote Sensing, and Mathematical Modelling for Surface Water Quality Management - page 198

example of the lake Edko, call for the important need for the integration of different analytical and
management tools. This integration helps to better understand the current status of the system and
indicates the possible management scenarios. In particular, the developed computational framework
for lake water quality management is considered an essential component of a decision support system
for the management of a lake. The developed hydrodynamic, water quality and eutrophication models
were able to give better understanding of the catchment and lake systems and they also succeeded in
focussing on different water quality processes within the lake. The developed framework and the
applied modelling system and tools on Lake Edko are generic and can be applied to other coastal lakes
of the Nile Delta region or other regions with similar conditions.

10.3. RECOMMENDATIONS FOR FUTURE RESEARCH

This research study focused on two major issues: managing surface water on a watershed scale and the
integration of management tools for deriving modelling systems with improved confidence for
management purposes. For future continuation of research in this field the following recommendations
are highlighted:

This research was implemented originally to deal with data scarce environments, which is a
common situation with many regions where there is a current lack of planning and
management tools. But improvements in field monitoring in watersheds (or catchments) and
shallow coastal lakes is still recommended with the field monitoring having a clear objective
related to the modelling objectives and calibration needs. For water quality measurements and
sampling the temporal and spatial frequency of measurements is critical. A specific
monitoring plan should be developed to cope with the needs of both hydrodynamic and water
quality models, and this plan should be consistent and linked with the remote sensing data
collection.

Improving the water balance model of the lake system taking into consideration the specific
analysis and modelling of groundwater is recommended to be developed further in future
research. Also needed is an improvement in the evapotranspiration rates within the water
balance modelling as a key parameter in the water budget modelling. Evapotranspiration rates
can be further studied with more focus on the rates associated with different vegetation species
using remote sensing data and analysis techniques. Application of remote sensing in the
calculation of evapotranspiration rates would be another added value of the integration of
remote sensing with mathematical models; this in turn can contribute to better level of
confidence.

Remote sensing data sets should be refined for improved water quality and eutrophication
modelling calibration. Since remote sensing was the main source of calibration data for water
quality models in this research, it is recommended that more high resolution spatial and
spectral data sets that are suitable for water quality analysis and extraction of water quality
parameters are developed. An intensive frequency temporal data set of SPOT and MERIS
images would be of considerable importance for further validation of the eutrophication
modelling results.

Attention should be given to the development of a site specific bio-optical remote sensing
water quality model for the Delta region coastal lakes. In this present research the focus was
on the mathematical modelling tools and their calibration using alternative data sets from
remote sensing analysis. The remote sensing analysis methodologies used were based on
empirical and statistical approaches, but for further studies of coastal lakes and monitoring
their status, it is recommended to apply an advanced remote sensing analytical approach
through the development of a site- specific bio-optical model that can be applied to these
lakes. This model will require specific in situ measurements that can be done once and
verified and calibrated every year. This model would be the key model for the application of