Identifying systemic change in catchment hydrology

Temporal change in landscape systems is mostly studied with a focus on temporal variation in system states (e.g. groundwater level, discharge, denudation rate). These changes are driven by the landscape system, which includes all driving forces active in the landscape. In most cases, this system is considered constant, which implies that it is assumed that the processes and their interconnections remain the same. For instance, model calibration against observational data (aiming at parameter identification) mostly assumes that the set of modelled processes and their associated parameters remain the same: a single set of equations and parameters is assumed to represent the past and future behaviour of the system. In many cases, however, the system itself may change over time, due to external forces or due to internal mechanisms in the landscape that completely alter the system processes and system behaviour. An example is the land use system. Land use change is driven by many factors, including land prices, transport costs, housing costs, environmental properties. In many cases, these factors are considered constant and land use change is modelled with the same set of rules for all time steps. In reality however, many of these factors may change due to implementation of new technology or environmental laws, which implies systemic change of the land use system.

In this study you will address systemic change in the hydrological system. Hydrological models are nowadays important tools for forecasting drought and flooding. To reduce uncertainty in forecasts, these models are calibrated against observational data, in most cases river discharge time series. As noted above, it is mostly assumed that one unique set of model parameters can be used to represent hydrologic behaviour for all time periods (both past and future simulations, for all years). In reality, however, systemic change will occur, which will be associated with changes in parameter values. Systemic change in catchment hydrology may be due to changes in land use (causing changes in interception, infiltration), changes in geomorphology (causing changes in soil depth and subsurface hydrology), or other changes such as implementation of new reservoirs. In this study you will identify these changes by an inverse method, by calibrating a catchment model separately for each time period (typically one year) in a series of time periods. This will result in a time series of parameter values (i.e., a value for each year), that represent the temporal change in the hydrologic system. Following this approach you can address the questions of 1) What is the temporal change in parameter values of a catchment model? 2) Is it possible to relate these temporal changes to changes in the modelled catchment (e.g. landuse, geomorphology, reservoirs) that caused this systemic change in the hydrology? 3) What are the possible implications of this systemic change for forecasts of catchment discharge?

For this study you will use existing calibration techniques on an existing data set (large timeseries data are available for multiple decennia) and model (one of the data sets available, most likely the Danube catchment). This is an interesting topic if you would like to apply your knowledge in hydrology in a challenging rather innovative study. You will get technical support from staff and PhD students at our institute to get the calibrations running. The content of the study can be adjusted to your interests (e.g. you could also study systemic change in other landscape systems).

Supervision: Dr Derek Karssenberg (Utrecht University)

Location: Utrecht University

Period: to be determined

Number of students: 1-2

Program/track: Earth Surface Hydrology or Natural Hazards and Earth Observation

Prerequisites: courses in (stochastic) hydrology, spatio-temporal modelling (content of project can be adjusted to your background)

Contact/info: Derek Karssenberg (d.karssenberg@uu.nl)