Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Master → Australie → Identifying drivers of streamflow salinity in data-limited catchments

University of Melbourne (2016)

Identifying drivers of streamflow salinity in data-limited catchments

Malana, Mohammad Naeem

Titre : Identifying drivers of streamflow salinity in data-limited catchments

Auteur : Malana, Mohammad Naeem

Université de soutenance : University of Melbourne

Grade : Masters Research thesis 2016

Résumé
Salinisation of water and land is a serious problem affecting global environment, economy and food security. The strategies to mitigate salinity problem require an improved understanding of the factors causing it. This requires exploring the variability of streamflow salinity, which is an indicator of catchment salinity and highly variable in space and time. Previous research has focused on estimating streamflow salinity through a variety of models that are usually data-intensive, site specific and do not have wider applicability. The arid and semi-arid regions of the world, which are worst affected by the salinity problem, have data-limitations and commonly cannot support calibration of complex models. Therefore, a simple approach in analysis is required. This research project developed a simple approach for assessing the variability of streamflow salinity in data-limited catchments. This aim was achieved by developing a data-based approach to identify drivers of spatial and temporal variability of streamflow salinity. The spatial variability of streamflow salinity was analysed with the help of two methods. The first method used the Budyko curve to classify study catchments between the less-saline catchments and the catchments at risk of salinity. The Budyko curve analysis identified the long-term annual catchment water balance and climate as the main drivers of the spatial variability of streamflow salinity. The second method applied a statistical analysis and identified five drivers of streamflow salinity. These drivers are climate, topography, land-use, streamflow characteristics and mean annual evapotranspiration. The statistical model used long-term climate data and catchment static metrics. The catchment static metrics were derived from data on topography, climate, land-use and land-cover, geology, streamflow and salinity for the characterisation of study catchments. The 78 study catchments were projected into a physiographic space using the principal component analysis. This analysis was based on the catchment static metrics significantly associated with streamflow salinity. The statistical models involved ordinary least squares regression and inverse distance weighting interpolation method as the tools for prediction in the study region. The temporal variability in streamflow salinity was analysed at an annual and monthly time steps using a catchment water balance. It was demonstrated that catchment water balance is the driver of temporal variability in streamflow salinity based on the observed data. The Budyko curve and the Zhang et al. (2001) model were used to assess the inter-annual variability in the average streamflow salinity. A simple Thornthwaite-type monthly catchment water balance was used to explore the seasonal variations in streamflow salinity on time-series plots. However, no significant relationship between monthly water balance and streamflow salinity was detected on an ‘x-y’ scatter plot for all the study catchments, This is mainly due to uncertainty in data inputs, model error, the time-lag effect and bias in the observed streamflow salinity data. A simple data-driven approach was developed to analyse drivers of the spatial and temporal variability in streamflow salinity. This approach is suitable for data-limited catchments. The Budyko curve has the potential to explain both the spatial (geographical) as well as temporal (inter-annual) variability in the average annual streamflow salinity. The variability in streamflow salinity is derived by climate, catchment water balance and physical attributes of catchments. Most of these drivers are natural processes except for changes in land use. Therefore, the management of catchment water balance through vegetation management (e.g. reafforestation) is a practical measure to mitigate the salinity problem.

Présentation

Page publiée le 2 août 2017