What is a Virtual River Basin?

A basis for integration and application of multi-scale information

Multi-scale Resource Challenges in River Basins

River basins and their downstream coastal zones are facing a series of challenges critical to their future, centered on the availability and distribution of water. Floods and droughts impact freshwater resources, agriculture, biodiversity, and livelihoods. Development of hydropower provides much-needed energy, but alters the flow regime and sediment transport of the river. Climate change is then superimposed on all aspects of the system, bringing alterations in temperature and rainfall regimes, and reduction of snow cover. Global economic impacts and food shortages are a growing concern. Management options cover a range of issues, from bringing safe water to local villages for the rural poor, to adaptation strategies for large infrastructure, as climate evolves. International efforts must be made to predict and to develop sound adaptation strategies.

But such strategies require information. Answers are needed to such questions as,

  • What would be the impacts of changes in agriculture (including irrigation) and forestry practices on local and regional water balances? Will forest harvests lead to increases in stream flow and potential erosion?
  • How would changes in land use practices, with varying climate, affect water supply and water quality?
  • If some indication of climate over a growing season was provided, could crop selection (and fire management) be improved?
  • Can floods or droughts be predicated, or at least anticipated, one or two months into the future, as an early-warning system?
  • How would developments of infrastructure affect downstream flow, water quality, hydropower and fisheries resources?
  • How would changes in stream flow affect fisheries?

The answers have very practical outcomes, for management:

  • At field scales (for individual farm or plots), how can best sustainable land management (SLM) or natural resource management (NRM) practices be implemented?
  • At watershed scales (for governance and regional policy) - How can SLM/NRM practices be expanded and scaled up? How can agriculture, biodiversity, infrastructure, and cultural values be optimized?
  • At whole basins (for food and water security) - How far downstream will local demands or actions be propagated ? How would climate change impact water, temperature, and nutrients, for agriculture and biodiversity?

These issues represent a very complex set of intersecting scales, cross-sector science and technology, education, politics, and economics. One of the most significant challenges for evaluation of past performances, and especially for laying the basis for future decisions, is how to analyze, in a quantitative manner, the multiple pathways  that have produced current, or baseline, conditions. A template where decision-makers can consider rigorous scenarios of alternative futures could play an important role in making complex environmental and economic decisions. This requires an accurate understanding of linkages between water and multiple allocations, with the ability to quantitatively forecast individual and combined impacts of demand. With such information, the need is to evaluate the trade-offs between sectors is a basis for future policy interventions and financial investments.

The Virtual River Basin: A Foundation for Multi-Sector Integration of Information

To address these questions, information from multiple sources must converge, be organized and evaluated (preferably according to guiding principles), and disseminated. In this spirit, it is useful to think of a Virtual River Basin (VRB), as both a metaphor and a practical engine, for organizing and processing information and decision needs (Figure 1). Everyone not only needs, but understands and can relate to, water. A river basin provides a tangible, physical boundary on the movement of water, and in the process, on all of the activities within that basin, from agriculture to fisheries to forestry to urban development. "Virtual" refers to a construct that can be thought of in the abstract and used to explore "information space," but can be efficiently applied to (multiple) specific basins.

Figure 1
Mekong Basin Annual Precipitation

The theoretical structure for a VRB is to track the overall pathways and processes of water, as water moves from the atmosphere to and through the landscape and down river channels, through reservoirs and lakes, to the sea, on a geospatially-explicit, multi-temporal basis (as described, below). The knowledge necessary to track water includes an understanding of, and a mobilization of information for, all aspects of the landscape, including agriculture practices, landcover, topography, soils, fisheries, infrastructure, and human interactions. A Baseline assessment of the current and past environmental conditions of a basin (to establish both the extent and processes of change) provides the foundation to build from. With such a baseline established, then future scenarios can be analyzed, and the evolution of key system variables can be monitored.

Figure 2
Dynamic Information Framework

As a means to start the discussion on developing a VRB, consider the following conceptual framework (Figure 2).  The construct is that each module of the framework represents internally consistent data and information with sequential exchanges between each module. The information is derived from direct measurements and observations, and from models that help interpret and refine that information.

The first set of modules establishes the basic structure and dynamics of the Basin. The Drainage Basin (Module 1) establishes basic attributes of the landscape, including topography, soils, landuse and landcover. The Climate Forcing (Module 2) “drives” the landscape with precipitation, temperature, and winds. Climate can be derived from surface observations (including telemetry to a home base), satellites, and climate models. The Water Movement (hydrology, Module 3) then proceeds as the product of the climate acting across the templates of the landscape. Such models can then be used, or “coupled” to other models; e.g., for climate or hydropower or carbon exchange with the atmosphere, and used to evaluate the impacts of landuse change, irrigation, dams, and climate change on the hydrologic cycle. Chemical loading (Module 4) is the product of water inputs (from Module 3).

The second set of modules addresses the production basis, building on the “physics” of the basin. The Landscape Production (Module 5) represents primary production by landcover (including natural vegetation and agriculture), and secondary production (including livestock), responding to the structure of the drainage basin, and climate forcing (including changes in climate). Coupled to the hydrology models, net ecosystem (carbon) production can be calculated. At progressively finer resolution (“downscaling”), specific agriculture crops can be represented, with data from multiple sources and models. This would allow better insight into primary production along gradients.

Finally, the third set of modules addresses how economics and policy interact with the “biosphere.” Asset value (Module 6) represents both the economic consequences and feedback of the utilization of ecosystem goods and services, and aesthetic/cultural values placed on the environment. Policy (Module 7) represents the legislative intersection with the management of the basin, including policies from land tenure decisions to specific, nominally informed, legislation.

The concepts are equally relevant at progressively fined scales, down to individual projects. The construct allows upscaling, as well, relating how an individual project or locale is “nested” in larger regions. The execution of an architecture such as the one sketched out here provides a framework for identifying specific field sampling requirements, from climate stations to suspended sediments to economics of resources. The framework can then serve as the organizing structure for the activities of a region, including providing a basis for development of management scenarios. A Basin Baseline can be developed, as organizing and analyzing the information required to bring each module “to life.”

Whereas the information and decision issues confronting a basin are challenging, they are not unique. The recognition of the need for more holistic views is broadening, especially in the last few years. The robust framework for enabling a VRB is the emergence of a new generation of “Earth System Science,” based on the rapidly evolving capabilities for addressing global change issues. This involves use of satellites, new generations of dynamic computer “ models,” field measurements focused by model requirements covering wide areas, and, especially, a thinking and practice of “integrated systems.” Fundamental to these is a new class of hydrology models, which can be regarded not only as hydrology models, but as overall landscape models, because of the processes (and data layers) they represent. A key aspect to these models is that they are geospatially-explicit, fully-distributed, recognize the spatial heterogeneity of the watershed, and are process-based. Because these models can, and must, “meld” information from multiple sources, they can be functional in specific regions where local data are relatively sparse.

Functionally, the core of the VRB is built by progressive information layers, identified as the required inputs for the geospatial hydrology and landscape models, but can serve multiple purposes (Figure 3).  For example, information on land cover classifications identify the biophysical attributes of vegetation needed for modeling can provide the basis fo r carbon inventories, regional zoning, and so on. The first layer of information is provided by topography, which defines the boundaries of a river basin. These data can be derived in many ways, from local maps to the Shuttle Radar Topography Mission (SRTM). The topography data is used to derive river networks, and grids about how flow is accumulated.  Political boundaries can be superimposed on the basin, recognizing that such boundaries most frequently do not correspond to the basin itself, leading to transboundary issues. Information on soils is needed, including soil type, depth, texture, and fertility. Such data are typically derived from local knowledge, or from global datasets. Landcover information, from regional surveys and different satellites, is critical for multiple purposes. An all-important “driver” of the land surface is climate, expressed as the minimum and maximum temperate, precipitation, and winds. These data can be derived from local weather station networks, and from regional and global data assimilation schemes and climate models.  Finally, dynamic simulation models of the movement of water and energy can be built, addressing water distributions, agriculture productivity, and carbon cycles.

Figure 3
Models-Movement of water, energy

Thus a VRB can be thought of as the common environment for the overall information sources describing a Basin, organized in a highly systematic fashion, to facilitate analyses, and to “visualize” outcomes. By being organized according to landscape and hydrologic principles, information can serve multiple purposes, with specific targets for information identified and prioritized. The intersection of biophysical processes and environmental stressors can be seen in a geospatially-explicit fashion. Careful attention must be paid to how the mechanics of information is organized, displayed, and distributed.


A Dynamic Information Framework

Establishing such a process is not a trivial task, for several reasons.

  • The information required comes from multiple sources, from individual rain gauges to statistics on rice yield and fisheries. The information required comes from multiple disciplines, which presents problems with even communication between specialists. Existing data holdings are not always readily obtainable, sometimes for institutional reasons. New field measurements, especially holistic and cross-boundaries, are challenging.
  • Handling such diverse data and executing models is not straight-forward. There are very real problems in converting data streams into useful information that go beyond a database.
  • Perhaps most challenging is how to not only create such information, but how to get it into the hands of users of different levels, from the specialist to the local and regional decision makers to the local farmer or fisherman.
  • Few, if any, institutions in the world have sufficient in-house expertise to execute all parts of such a process.


Some of what is discussed here may seem obvious, but in practice it is not. Information frequently doesn’t pass from one office to the next, and remains in cylinders, or is sequestered. Models can be so complex that it is difficult to see the outcome. Models that are actually used frequently don’t represent the most recent advances, especially amongst practitioners. Even the most modern models are difficult to use, and represent more generations of graduate students than a coherent evolution, and lack a community. At the core of a Virtual River Basin is a Dynamic Information Framework (DIF), with the objective of providing a consistent theoretical basis, and the overall capability of integrating across sectors.

  • “Dynamic” refers to the fact that the landscape is a evolving – that we must look at not only the present, but at the past and especially the future needs. Neither is information static.
  • “Information” means that more than just which data need to be considered – e.g., which products must be developed from data?
  • “Framework” means that an overall set of information must be logically arranged and communicated, within a flexible environment. The structure for readily interacting with the DIF must be clear.

Underlying the technical details are the issues of dealing with

  • Ownership of and access to primary data,
  • Where systems reside (national, ministry, agency),
  • Accessing and using core information from multiple locations for inclusion in analysis, synthesis, and outputs, and
  • The communication of scenarios and likely outcomes.
Figure 4
DIF-Dynamic Information Framework

The computational and data organizational issues represented in executing the DIF require explicit attention. This schematic expresses the sequence of issues to be resolved, from the details of metadata and data storage, to facilitated access (Fig. 4). It is useful to think in terms of mobilizing the data from archives (and its attendant issues) to “data str eams,” which focus on specific outcomes, as represented by the modules. The actual operation of transferring data from archives to an accessible, quantifiable data layer can be expedited by including data services, for processing the data into usable forms. Given the complexity of outcomes, experience has shown that attention to providing visually compelling data products is very important for effective communication not only with decision and policy makers, but also with the public at large.

To enable a functional DIF, then specific components of the DIF need to include:

  • Base data layers;
  • Directed data layers, focused on synthetic objectives;
  • Geospatially-explicit, process-based, cross-sector simulation models (requiring data from the directed data layers). A modular structure allows ready swapping of models (while focusing on getting work done);
  • Facilitated input/output (including visualizations);
  • Decision support system and scenario testing capabilities.

Essentially, a DIF is a numeric and quantitative “Commons,” which builds on the legacy of knowledge from experience, with the goal of “harmonizing” watershed function for multiple users. The goal is to provide an instrument for a (quantitative) analysis of complex interdependent problems. The process of creating the model provides an integration of data from multiple sources (of interest to many). The framework provides a means for interpolation for sparse data, provides the basis for cross-scale/upscaling analyses, and provides the foundation for “scenarios.” The framework should be cross scale, allowing accurate representation of large regions and far-field effects, while being able to “zoom in” to a specific site of project. While flexibility is highly desirable, hence the term “framework,” emphasis must be achieving a stated goal.

Current and Planned Applications of the DIF

The DIF construct is not an esoteric, theoretical exercise. Rather it is a construct that is not only realistic at this point, but practical. The basic principles have been developed through work on the Amazon and Mekong river systems, and the Puget Sound basin of Washington state (and the broader global change community). The ideas have been presented at World Bank for a, including Water Week, February 2007 (Quantitative Approaches to Optimizing Water, Land and Biodiversity Management), the Sustainable Development Network (SDN) February 2008 (Watershed and Basin Management – Integrated Approaches across the SDN Practice), and Rural Week (Agriculture and Climate Change: Modeling and Managing Water from Basin to Nozzle).

It is currently being applied to, and is being developed from, emerging World Bank/GEF projects. The Ministry for the Coordination of Environment Affairs (MICOA) of Mozambique is developing a GEF project, Zambezi Valley Market Led Smallholder Development Project: Baseline Data on Landuse, Biodiversity (Zambezi) for the lower Zambezi River basin, including developing baseline and scenario datasets on landcover, biodiversity and hydrology, in support of improving smallholder agriculture. VIC is the core model for the ongoing World Bank/GEF project on the China 3H Basin project Mainstreaming Adaptation to Climate Change into Water Resources Management and Rural Development. The Ministry of Agriculture (MoA) of the Royal Government of Bhutan is developing the capability DrukDIF to use landuse and climate change scenarios to proactively address land and natural resource management to minimize and reverse land degradation, and impact on floods and hydropower. The Mekong River Commission is building on the outcomes of several projects, to produce a "Virtual Mekong Basin," to synthesize and communicate potential outcomes of climate and landuse change and hydropower development on water, agriculture, and fisheries. Through consultation with the Lake Victoria Basin Commission (LVBC), and the national teams for Kenya, Uganda, and Tanzania, the framework elements for a Lake Victoria basin Dynamic Information Framework were elaborated for the proposed IAD Lake Victoria Environmental Management Plan 2.