Diagnostic assessment of dominant process controls on the functioning of ecohydrological systems

Patnaik, Swagat and Biswal, Basudev (2018) Diagnostic assessment of dominant process controls on the functioning of ecohydrological systems. PhD thesis, Indian institute of technology Hyderabad.

Text
Thesis_Phd_CE_5192.pdf - Submitted Version
Restricted to Repository staff only until June 2021.
Download (22MB) | Request a copy

Abstract

Freshwater is an important ingredient for the survival of any human civilization, and thus optimal usage of available water resources is one of the key objectives of engineers and policy makers. It is now widely acknowledged that many parts of the world are facing severe water crisis mainly due to steep increase in population and climate change. As a consequence, natural ecosystems in those parts are currently under tremendous stress and are showing signs of attending new equilibrium states. Although the primary solution for the above mentioned problems is to develop appropriate physical infrastructure (e.g., construction of dams, barrages etc.), a major hindrance is that our knowledge on natural systems is often inadequate because of complex interactions between soil-water, vegetation, and atmosphere. In particular, we know little about the variability of water in and space and time and its physio-climatological drivers. Several philosophical approaches have been proposed in the past to study the spatio-temporal variability of water availability, each having its relative merits and demerits. In this study, we follow the dominant process concept which recommends the identification of dominant factors and processes controlling an ecohydrological phenomenon. We focus on understanding the influence of physical and climatological factors on three distinct ecohydrological phenomena: i) large-scale change of gross primary productivity (GP P) ii) emergence of non-linear storage-discharge relationship at the basin scale, and iii) transformation of rainfall into runoff or streamflow. All the three phenomena are studied using fairly large datasets. We studied the first phenomenon considering GP P and several other datasets for India to understand how the state-wise GP P trends during the time period between 1982 to 2008 have been influenced by biotic and abiotic factors. Although it is well known that various natural and anthropogenic factors (e.g., climate change, agricultural activities, and atmospheric CO2 concentration change) can alter GP P, their relative influences are not clearly understood in India. We found the common suspects like CO2 concentration change and climate change to have insignificant influence on the GP P trends in India. Our results instead support the notion that GP P trends in India are mainly shaped by agricultural activities. Furthermore, we show that GP P trends in India are caused by irrigation as well as nutrient loading. Overall, our study reveals the potential of agricultural activities in altering the ecohydrology of a region. The second phenomenon was analyzed by studying power-law recession flow relationship between discharge (Q) and time rate in change of discharge (dQ/dt). Discharge during recession periods occur due to slow drainage of storage, due to which storage at any point of time is expected to influence future recession curves. Thus, one may expect some relationship between the dQ/dt − Q power-law coefficient (k) and storage state at any point in the past, characterized by past average discharge (QN ). The strength of this relationship, characterized by the coefficient of determination R2 N , provides crucial information on the catchments storage-discharge dynamics. Using discharge data from a set of 354 US basins, we studied the relationships between R2 N and several catchment characteristics in a multivariate analysis framework. In particular, we performed step-wise multiple linear regression followed by principal component analysis. Many catchment characteristics (e.g., standard deviation of elevation and average soil clay content) were found to be significantly influencing R2 N . Similar observations were made when we studied k − QN relationship based recession discharge predictions. Further, we observed that the dQ/dt − Q power-law exponent (α) is strongly influenced by several catchment characteristics. However, our analysis reveals the value of α being influenced by a unique set of catchment characteristics in every region, possibly suggesting that the vii storage-discharge relationship of a region is shaped by the way its landscapes have evolved across geological timescales. The third phenomenon was studied with the hypothesis that climate has a dominant control over hydrological partitioning and thus it is possible to obtain a universal calibration-free rainfall-runoff model. A recent calibration free model was chosen for this study. The model employs an empirically obtained instantaneous dryness index, which is a function of antecedent rainfall and solar energy inputs, to perform hydrological partitioning using only climatic information. The model was applied in 50 catchments in India without any calibration and the results were found to be similar to those from the US, where the model was developed. Overall, observations here seem to support the notion that climate mainly governs catchment hydrology. The insights from this study can thus help in performing hydrologic prediction in discharge data-scarce regions. In summary, this study highlights the usefulness of the dominant process concept based approaches in understanding how physio-climatic factors affect the functioning of an ecosystem. The main advantage of these approaches is that we can uncover many details about an ecohydrological system without employing purely mechanistic bottom-up approach, which is infeasible because of the enormous complexity that symbolizes the relationship between soil-water, vegetation, and atmosphere of a typical ecohydrological system.

[error in script]

IITH Creators: