A foundational tenet of the ecosystem services concept is that they arise from biophysical processes. Riverine landscapes are process-response systems where river flow and geomorphology generate a heterogeneous physical template that influences ecological processes, suggesting that the supply of ecosystem services in riverine landscapes should be congruent with the character and heterogeneity of the physical template. In this study, we examine the congruency between the physical template (river functional process zones; FPZs) and the supply of river flow dependent ecosystem services from riverine landscapes of the Koshi River Basin, Nepal. The supply of ecosystem services was congruent with FPZs. Social factors were shown to mediate the use and value of ecosystem services between FPZs. Heterogeneity of the physical template interacts with place, social activity, and demography to influence the use and potential value of ecosystem services across the riverine landscape. These spatial patterns of greater use of some types of riverine ecosystem services in certain areas of the riverine landscape are indicative of a highly coupled agricultural or "green loop" social-ecological system (SES) and show that maintaining riverine template heterogeneity is an important element of this green loop SES that supports 40 million people in the Koshi River Basin.

Study region: Koshi River basin, Eastern Nepal.

Study focus: Climate change is increasingly evident as the global surface temperature is warming with erratic rainfall patterns across the globe. In this regard, the Koshi Basin in the Himalayan region is also impacted, and it is important to understand the spatio-temporal details of the impact in the basin under future climate change. This study assessed the potential climate change and its impact on the hydrological regime using the Soil and Water Assessment Tool (SWAT) and Indicators of Hydrological Alteration (IHA) based on RCP4.5 and RCP8.5 of ensemble downscaled CMIP5 GCM runs.

New hydrological insights for this region: Results show the upper part of the basin warming faster than the lower part, the pre-monsoon season warming more than other seasons. There is no clear uniform trend in precipitation. However, the southeastern part of the basin will get more precipitation. Sub-basins will get more precipitation during the post-monsoon under RCP4.5, and during the monsoon under RCP8.5. The annual water availability will not decline but water availability within seasons and regions is projected to be highly variable. There is also a change in the spatial pattern of river discharge and the western part of the basin is likely to experience more impact. Therefore, these findings will be valuable in identifying how particular sub-basins within the Koshi Basin will be impacted by climate change and in stipulating effective planning and management of water resources for the future.

Riverine landscapes have been conceptualised as complex adaptive systems, characterised by many biophysical and social components, which interact at multiple scales. This complexity challenges traditional scientific methods because the multi-causal, multiple-scale character of riverine landscapes limits the usefulness of conventional reductionist falsification approaches, except at smaller scales and within limited domains. Identifying and understanding the various biophysical and social drivers of rivers, and their connections are challenging. Historically, most river science has been limited to relatively small-scale and locationspecific studies conducted over small-time scales and the application of these studies to entire riverine landscapes is questionable as it infringes on the basic principles of hierarchy theory.

The study of ecosystem services in riverine landscapes is biased in scale. The majority of studies of flow-dependent ecosystem services are at a single scale, i.e. at a site or reach scale, and do not consider the entire river network. In addition, studies only evaluate one or two types of ecosystem services and the effect of climate change on these flow-dependent ecosystem services is limited.

The Koshi River Basin is the main river system of the greater Himalayan region and one of the most complex Himalayan rivers. It drains a region sensitive to climate change. The upper part of the basin stores substantial fresh water in the form of snow and glaciers. The system plays a key role in irrigation, household water for downstream areas, has a large potential for hydropower development, and supports ecosystem functioning. The Basin is home to more than 40 million people, of which 80 percent are dependent on the ecosystem services the system provides. The Basin is also home to sensitive and crucial ecosystems, with protected areas that support high levels of biodiversity – it is a hot spot of ecosystem services and functions as a vital corridor for various fauna.

This thesis uses multiple lines of evidence to understand the relationships between the physical template of the Koshi’s riverine landscape, flow-dependent ecosystem services and the influence of climate change. The first study (Chapter 2) examines the congruence between the physical template and flow-dependent ecosystem services of the Koshi River Basin network. River characterization of the Basin shows a spatially heterogeneous river network and a high degree of congruency between the physical template of the river network and the abundance, use and value of flow-ecosystem services. The abundance and use of flow-dependent ecosystem services are heterogeneous among the various river zones identified within this river network. However, the potential value of ecosystem services is influenced not only by physical templates but also by the demography of the regional population.

The second study (Chapter 3) examines the potential effect of climate change on the flow regime of the Koshi River Basin. A hydrological model was developed to determine the potential changes to the flow regime of the Koshi Basin, under two climate scenarios over a 100-year period. Results show significant changes in the flow regime of all sub-basins in the Koshi due to climate change. Flow regime components were projected to increase in most sub-basins. However, flow regime changes vary among the six sub-basin studies and the rate of change varies over time. Changes to the frequency of high and extremely low flow events demonstrate the increase in hydrological extremes. Flow regime changes accelerate over 100 years with the largest projected increases for Representative Concentration Pathway 8.5 scenarios.

The third (Chapter 4) study examines the response of the flow-dependent ecosystem services to climate change and the influence of lateral position in the riverine landscape on this response. Flow regime data from Chapter 2, along with the location of ecosystem services at three different lateral positions in the riverine landscape were used. A matrix of flow dependent ecosystem service responses was constructed for the Sunkoshi River Basin. The distribution of the flow-dependent ecosystem services varied significantly between the river channel, riparian zone and floodplain sections of the riverine landscape. The potential change in the flow regime also differed by lateral position under the two climate scenarios. As a result, the response of ecosystem services to climate change in the Sunkoshi riverine landscape is predicted to vary under different climate change scenarios according to lateral position. Thus, the distribution and response of flow-dependent ecosystem services to climate change are not uniform in relation to the lateral position of the riverine landscape.

The fourth study (Chapter 5) examines the impact of climate change on flow-dependent ecosystem services based on the geomorphological organization of a river network. A river characterization schema, an inventory of ecosystem services among river zones, and flow regime data from Chapter 2 were used to construct a matrix of flow-dependent ecosystem services according to river zone. Results show the response of the flow-dependent ecosystem service to climate change will not be uniform within the Koshi River network. The responses varied most by river zones and lateral position but the nature of the response of ecosystem services to climate change was consistent among all sub-basins. The finding of this study highlights that the response of the flow-dependent ecosystem services to climate change is determined by the physical template of the river network. This study highlights the importance of scale and hierarchy to the response of ecosystem services in the riverine landscape.

This thesis highlights the importance of the physical template in the production of flowdependent ecosystem services across the riverine landscape and recognises that the heterogeneity of the physical template influences the response of ecosystem services in the riverine landscape to climate change within river networks. The Koshi River Basin does not support a simple clinal gradient river model, its river network is heterogeneous in terms of the arrangement of its physical template. Given the congruency between the physical template and ecosystem services, ecosystem services are also heterogeneous within the river network. Moreover, the response of ecosystem services to climate change within a river network is also not uniform and varies according to the lateral and longitudinal position of the river network. This thesis is the first to study the importance of the physical template for ecosystem services in a large Himalayan river basin. Furthermore, this study shows that the distribution and response of flow-dependent ecosystem services are complex and respond in a complex way to flow regime changes arising from climate change. In particular, this complex response depends upon lateral and longitudinal positions within the river system. Moreover, the relationship between flow and ecosystem services may change over time.

This research has enabled the construction of a social-ecological systems framework to understand the abundance, use and value of ecosystem services within entire river networks. The framework unpacks the complex interplay of biophysical and social components within riverine landscapes as well as understanding the role of controllers for the production of ecosystem services and their use and social value in the riverine landscape coupling the human dimension. Overall, this framework illustrates the riverine landscape as a complex adaptive system.

Overall, this thesis contributes to our understanding of riverine landscapes as a complex adaptive system. Complex adaptive systems are a mechanism to understand riverine landscapes as social-ecological systems. Ecosystem services are an ideal indicator to consider interactions between the environment and humans in a social-ecological system through a lens of complex adaptive systems. Finally, this thesis improves knowledge of the interaction among physical templates, ecosystem services and people at a large scale in a systematic way, and demonstrates how the riverine ecosystem services may respond to climate change.