Derived Triangulated Irregular Network (TIN) for each study basin with the finalized basin boundaries and channel networks

•    A Digital Elevation Model (DEM) with a resolution of about 29 meters obtained from the National Elevation Dataset (NED) was used to generate the basin boundaries, channel networks and deriving the TINs for the Santa Cruz and San Pedro River basins.

•    TINs were derived using the TIN Index Analysis Program (TIAP). The preprocessing of the DEM included the extraction of a floodplain from the DEM to be used in the Terrain method (slope based) that was used to generate the TINs.

•    The terrain or slope criteria method is based on the topographic relevance of DEM points in describing the terrain. Regions with drastic elevation changes (rugged) are characterized with a much higher resolution, whereas flatter areas have a lower resolution, except in the floodplain areas where high resolution is retained.

•    We generated TINs at various resolutions which resulted in the decision to use a coarser resolution for the larger domains (Santa Cruz and San Pedro watersheds) and a higher resolution for the smaller domains (Box Canyon and Walnut Gulch subbasins).  The stream network used for the large domains has a threshold area of about 10 km², whereas the smaller domains have a channel upslope area of 0.03km².

•    The TINs of the entire NSF-WSC domain including the higher resolution smaller domains are shown in Figure 1. Notice the high resolution around the regions where the smaller domains are located.

 

 Figure 1. Final derived Triangular Irregular Network’s (TINs) for the NSF-WSC study area.

Model Attributes and Outcomes for Use in Economic Work

I met with David Brookshire, Craig Broadbent, and Don Coursey (by phone) in Albuquerque on June 14. We had a discussion about the scenario’s approach we will be taking to look at alternative futures in our two systems. There is a strong interest on their part to tackle the spatial and temporal components that they were not previously able to address in their work on the San Pedro. As this works in well with the overall goals of the project it seems like a good way to go. Ultimately we will have 3 or 4 scenarios that will be used by the economics group to pursue but this will only be the case once all of the population projections, physical and biological modeling can be completed. In the meantime we need to develop qualitative and then quantitative scenarios that can be used by the economists to assess individuals preferences for differing outcomes along the San Pedro and Santa Cruz. As this will be an involved process first we should step through the whole process for a single instance. These scenarios will also inform the perturbations we will have to impose upon the coupled models of the system we have proposed to conduct. For simplicity and existing state of knowledge among the team we should first develop a scenario for the San Pedro.

The EPA ICLUS project (http://cfpub.epa.gov/ncea/global/recordisplay.cfm?deid=203458) used a set of storylines that developed from a detailed application of the last IPCC’s emission storyline’s (http://www.ipcc.ch/ipccreports/sres/emission/093.htm). I think using a similar approach here would be useful. Since Francina et al. have been using the A2 emission scenario.

From http://www.ipcc.ch/ipccreports/sres/emission/index.php?idp=94

4.3.2. A2 Storyline and Scenario Family

The A2 scenario family represents a differentiated world. Compared to the A1 storyline it is characterized by lower trade flows, relatively slow capital stock turnover, and slower technological change. The A2 world "consolidates" into a series of economic regions. Self-reliance in terms of resources and less emphasis on economic, social, and cultural interactions between regions are characteristic for this future. Economic growth is uneven and the income gap between now-industrialized and developing parts of the world does not narrow, unlike in the A1 and B1 scenario families.

The A2 world has less international cooperation than the A1 or B1 worlds. People, ideas, and capital are less mobile so that technology diffuses more slowly than in the other scenario families. International disparities in productivity, and hence income per capita, are largely maintained or increased in absolute terms. With the emphasis on family and community life, fertility rates decline relatively slowly, which makes the A2 population the largest among the storylines (15 billion by 2100). Global average per capita income in A2 is low relative to other storylines (especially A1 and B1), reaching about US$7200 per capita by 2050 and US$16,000 in 2100. By 2100 the global GDP reaches about US$250 trillion. Technological change in the A2 scenario world is also more heterogeneous than that in A1. It is more rapid than average in some regions and slower in others, as industry adjusts to local resource endowments, culture, and education levels. Regions with abundant energy and mineral resources evolve more resource-intensive economies, while those poor in resources place a very high priority on minimizing import dependence through technological innovation to improve resource efficiency and make use of substitute inputs. The fuel mix in different regions is determined primarily by resource availability. High-income but resource-poor regions shift toward advanced post-fossil technologies (renewables or nuclear), while low-income resource-rich regions generally rely on older fossil technologies. Final energy intensities in A2 decline with a pace of 0.5 to 0.7% per year.

In the A2 world, social and political structures diversify; some regions move toward stronger welfare systems and reduced income inequality, while others move toward "leaner" government and more heterogeneous income distributions. With substantial food requirements, agricultural productivity in the A2 world is one of the main focus areas for innovation and research, development, and deployment (RD&D) efforts, and environmental concerns. Initial high levels of soil erosion and water pollution are eventually eased through the local development of more sustainable high-yield agriculture. Although attention is given to potential local and regional environmental damage, it is not uniform across regions. Global environmental concerns are relatively weak, although attempts are made to bring regional and local pollution under control and to maintain environmental amenities.

As in other SRES storylines, the intention in this storyline is not to imply that the underlying dynamics of A2 are either good or bad. The literature suggests that such a world could have many positive aspects from the current perspective, such as the increasing tendency toward cultural pluralism with mutual acceptance of diversity and fundamental differences. Various scenarios from the literature may be grouped under this scenario family. For example, "New Empires" by Schwartz (1991) is an example of a society in which most nations protect their threatened cultural identities. Some regions might achieve relative stability while others suffer under civil disorders (Schwartz, 1996). In "European Renaissance" (de Jong and Zalm, 1991; CPB, 1992), economic growth slows down because of a strengthening of protectionist trade blocks. In "Imperial Harmonization" (Lawrence et al., 1997), major economic blocs impose standards and regulations on smaller countries. The Shell scenario "Global Mercantilism" (1989, see Schwartz, 1991) explores the possibility of regional spheres of influence, whereas "Barricades" (Shell, 1993) reflects resistance to globalization and liberalization of markets. Noting the tensions that arise as societies adopt western technology without western culture, Huntington (1996) suggests that conflicts between civilizations rather than globalizing economies may determine the geo-political future of the world.

Since it will be hard for us to produce and for participants to digest a year by year description of changing conditions on the San Pedro the discussion I had with the economics team focused on descriptions of the system as specific time points in the future - year 0, year 5, year 10, year 25 and year 50. At each of these times there would be decisions that could be made that would affect the future ecosystem services of the river system.

Here are the three tasks we have in front of us-

1. First we need to define the specific outcomes that our modelling effort can provide. I have started on this step below. Please correct and add information as needed.
2. Describe decision points that are at year 0, year 5, year 10, year 25 and year 50
3. Describe conditions at these same years.

Outcomes

Outcomes from the models are the things that we can predict that people can understand and potentially care about. I have taken a first swing here that needs better detail from each modeling group and the economists. Also each outcome must be explicit in space, time and the value of what can be described (i.e. what can be predicted at what temporal and spatial resolution?). More precise and more detailed descriptions are better we can always simplify down later.

Climate Modelling-

Outcomes - Changes in annual precipitation (at 35km resolution) and occurrence of “extreme”/intense rainfall events, changes in mean annual temperature

Rainfall-Runoff Modelling-

Outcomes- changes in surface runoff and groundwater recharge at 10 km resolution and on an hourly basis, changes in magnitude frequency and duration of floods through the San Pedro channel network.

Scour Modelling-

Outcomes- stream power at 250 m² resolution, seed bed creation at 250 m² resolution

Groundwater Modelling-

Outcomes - Seasonal depth to groundwater, variability of depth to groundwater and streamflow presence/absence at 250 m² resolution. Seasonal ET flux at 250 m2 resolution.

Biological Modelling-

Outcomes - at river reach scale (~2-3km) vegetation condition class, species richness of herbaceous cover

Minutes/results from March 4th Meeting

Notes and Action Items from March 4^th WSC Project Meeting

In Attendance: Tom Meixner, Tom Maddock, David Brookshire, Craig Broadbent, Don Corsey, Julie Stromberg, Amanda Zuchi, Keith Nelson, Katherine Bao, Jennifer Duan, Bai Yang, Kirsten Neff, Imam Mallakpour, Enrique Vivoni, Augustin Robles, David Plane, Francina Dominguez

LINKING MODELS

Vivoni ASU - Enrique’s group needs the shapefile from the groundwater model.

Meixner/Maddock UA - Groundwater Model- The southern portion of the basin has an improved hydrologic characterization – this is why we used the Pool and Dickinson model.. In the north, the confined aquifer’s characterization is still not ready – we will have to wait about a year. Should we use the Anderson equation? The tRIBS model will replace the Anderson equation.

Needs: Historical pumping data, agricultural recharge, population growth and projected increased municipal water use in the basin, downscaled regional climate model.

Vivoni ASU- Will use the 35km downscaled climate model results to do a continuous run and use as an initialization for the 10km downscaled climate model results (which will only be available for 10 year time slices).

David Plane - Conclusion: The wells follow the people!

Dave Brookshire: We would like maps of wet/dry of the river through time.

Groundwater Model - Modflow has a 3D grid. The aquifer sees 500-year climate while the river sees the short-scale climate. This is why we need the tRIBS to characterize the river and get the seasonal variation in recharge and discharge correct. The simplified model of the interactions between the river and the groundwater captures the general pattern of summer floodwater in riparian groundwater – this model agrees with isotope data. We can adjust for pumping through the model.

Beavers? We don’t really have the scientific understanding to include beavers.

Is the importance of flood recharge a result of a natural property of the system or is it due to anthropogenic influence.

Brookshire: What is the baseline? We are stuck with today!

Plane: Does the location of the wells matter? In the short term yes, in the long term no.

Groundwater Models: Wet/Dry scenarios can be used to construct thresholds. Sources of water shift between wet and dry scenarios.

Model Scott C. Simpson (Modeling stream-aquifer interactions during floods and baseflow).

Using tRIBS will give them the lateral contributions to the main stem.

We need to have a consistent treatment of channel losses and coordinate between models – know the physics of infiltration losses. Rewati has some more recent results on this,

Duan: tRIBS – rectangular cross-section of river. Sediment

Jennifer needs center of channel network.

Sediment model needs:

-upland soil erosion from each cell

- Jennifer thinks that one-dimensional unsteady flow model is the best approach to do this. Begin with one-dimensional and then go to two-dimensional.

- Model predicts how much flow will recharge and this has already been done. Scour or deposition occurs because of the non-equilibrium conditions in the flow. Stochastic channel bank erosion. Channel planform evolution and riparian corridor prediction.

We need to focus on first order effects on sediment transport (probably temperature effects on viscosity is not first order.)

Use Laursen’s equation for sediment transport – USDA recommends. Non-equilibrium bed load transport model.

Selective transport model by size.

Julie: How do we deal with shear stress and the relationship with vegetation. How do we deal with this?

Jennifer will be able to simulate either on an event by event basis or continuous. There is more certainty on the shear stress.

Jennifer will feed information to Julie. Julie needs seed beds.

We will look at the migration of the river corridor and the location of riparian vegetation.

Julie Stromberg

Is trying to see if she should use a temporal or spatial approach. The spatial approach would be bulked, Enrique thinks that model space will be a good place to explore the link between spatial and temporal approach.

We will need to go from reach 1 to reach 11, and this will define the entire zone for the San Pedro.

Three links will give us an idea of restoration- at 3links we have an increase in the Marshland. Is there hydrologic knowledge at 3links?

The input data needed is the spatial cover (%) of perennial flow within each reach.

The output is woody condition by reach.

How long does it take for the cottonwoods to die?

Group discussion on meaning

are there actual thresholds and discontinuities how do we define these?



Develop some synthetic pathways of groundwater and biological change. In both spatial and temporal variability?

250 m resolution of groundwater MODFLOW model

Economists – stay over meeting

What make a place special? - San Pedro – birds, fly ways water

What do birds want? How do these things vary in time?

How do people think about what birds want?

Movie:

A) Time lapsed - current conditions, 5 years, 50 years

B) Show water trees birds across time space

C) From perspective zoom into reaches walk about

D) Iterate through different scenarios

Do we get to thresholds?

Paths - Current conditions to future how does drying or wetting play out?

1) Synthetic Pathways - Discontinuous - versus smooth

2) There may be difference between biophysical and resulting behavioral pathways.

3) End points

4) Use condition class reach classification.

Next Steps- Timeline for Valuation Group-

March: CE lit on bundles, literature on comparing trajectories

April- Pathways from Tom, schematic of the structure of valuation and resulting questions, outline of story board.

May - Who will puttogether the movie?

Juen Structure of valuation framework

September - Test story voards - focus group

Some MODFLOW River Modelling Issues

Posted for Tom Maddock


1) A Reach is that portion of the river or stream that that is associated with a particular finite-difference cell used by the groundwater flow model.

· 2)A Segment is a group of reaches that have 1) uniform or linearly changing hydraulic properties such as streambed elevation, streambed thickness, streambed hydraulic conductivity, and stream depth and width; 2) tributary flows or specified inflows or outflows (at boundaries); 3) diversions.

· 3) The numbering of segments and reaches is important. Segments are numbered sequentially from the furthest upstream to the last downstream segment.

· 4) Reaches within a segment must be numbered sequential from the furthest upstream reach in a segment to the last downstream reach of the segment.

· 5) The length of river within a cell is a function only of the cell size. Cell size is a precision issue.

· 6) Stream depth is used to compute the head in each reach by adding stream depth to the top of the streambed elevation for each reach. Stream depth is computed at the midpoint of each reach.

7) There are five options for computing stream depth: 1) Specify the stream depth for the first and last reach of a segment, and the program then linearly interpolates over the other reaches within the segment; 2) Uses Manning’s equation to determine depth as a function of flow,

Q = (C/N)A R^(2/3) So^(1/2)

where Q is stream discharge, C is a constant depending on length units, n is Manning’s roughness coefficient, A is the cross-sectional area, R is the hydraulic radius, and S₀ is the slope of the stream channel; 3) also uses Manning’s equation but the channel geometry is divided into an eight point cross-sectional; 4) uses depth-discharge and width-discharge power function relations for estimating stream depth and width as function of flow,

y = cQ^f and w = aQ^b

where y is the depth, Q is streamflow, w is width and variables a, b, c, f are numerical coefficients determined by regression; 5) uses values of depth and width for given streamflows on the basis of observations at a streamflow gaging point (the data is entered in tabular form).

· A different option can be used for each segment.