Desert Rivers: Considerations for Restoration Practices

by Colleen McConvill

1. Introduction

Each river system is unique in its geomorphic, hydrologic and ecologic characteristics. This is what makes classifying rivers into broad categories for restoration and management purposes so difficult. Applying a model developed on one river system too broadly to another system could result in restoration projects that ignore key elements of a specific system. Keeping this in mind though, classification can help us make better decisions about restoration by allowing us to learn from other river systems. Because desert rivers share some key characteristics, they can be a useful category in which we place some river systems so that comparisons can be made and restoration information can be shared. The purpose of this report is to explore some of the unifying concepts of desert rivers, explore the effect of disturbances on desert river systems and start a discussion of how these factors will change our restoration goals.

1.1 What is a Desert River?

This brings us to what exactly is a desert river. A desert is generally defined as an area that receives less than 500 mm per year of precipitation or alternatively as an area in which evapotranspiration exceeds precipitation. Deserts can be subdivided into arid (less than 250 mm/year) or semi-arid (250-500 mm/year) (USGS). While deserts are sometimes thought of as hot areas, precipitation, and not temperature, is the key defining characteristic. Deserts are also referred to as drylands,and these terms are used interchangeably here in. A desert river then would be a river that flows through a desert region, either entirely or partially (Kingsford and Thompson, 2006). The discussion here will be limited to non-ephemeral systems for the most part, as ephemeral systems are discussed as a separate topic. Desert rivers can be evaluated in a similar manner to any other river system by looking at geomorphic, hydrologic and ecologic components. Desert rivers, however have an additional component that should be addressed, climate. Periodic drought and climate change are two factors that should be important considerations when approaching restoration issues in desert regions.

Desert rivers can be divided into allogenic and endogenic (Figure 1), depending on the source of their waters. Allogenic rivers, such as the Colorado River and the Rio Grande in the American Southwest, source most of their flow from outside of desert regions and tend to be through flowing systems. Endogenic systems, such as the Cooper River in Australia, are fed by systems within the desert region and tend to be more ephemeral than allogenic systems. This is an important distinction to make because the two systems will have very different hydrologic and ecologic characteristics as well as different responses to disturbances.

Figure 1 - Cooper Creek, Austrualia on the left is an endogenic system, while the Rio Grande on the right is an allogenic system.

1.2 Historical Background

Studying desert rivers is important because approximately one-third of the earth’s land surface falls under this category (IALC). Due to global air circulation patterns, deserts tend to occur at or near 30 degrees latitude; however they can also occur in the interiors of large continents, such as Asia and Australia. These regions are predicted to expand with a warming climate (IPCC, 2007). The more knowledge we have about riverine systems in deserts the better we will be able to deal with changes to these systems in the future. Even though most river research has been conducted on humid systems, a large base of knowledge exists on dryland fluvial systems. This knowledge will need to be built upon in order to alleviate future stressors on dryland rivers.

Early research on dryland rivers was conducted on many continents, but because economic evaluation was the driving factor behind this research, it is mostly geomorphic and geologic in nature. During the late 19th century G.K. Gilbert and J.W. Powell completed surveys of the American West and particularly of the Colorado River system. The work completed by these individuals, particularly Gilbert, forms the basis of what we now call fluvial geomorphology. In the early 20th century research was also conducted by Europeans in the drylands of Africa and Australia, though much of this research was purely descriptive (Graf, 1988, p.20). Research conducted by Carter and Blanford in the Thar Desert of Pakistan added to the work done in the American West. L.B. Leopold and J.P. Miller built upon the research from the 19th and early 20th centuries by conducting studies on small fluvial systems in the Southwest (Nanson et al., 2002).

More recent research has focused on particular river systems and testing of specific models of various types, including hydrologic, geomorphic and ecologic. Because much of the historical research was conducted on river systems with steep gradients, a gap existed in the research for streams with low gradients dominated by suspended sediment loads. Australian researchers such as L. J. Bull and M.J. Kirby have sought to fill this void from the hydrologic and geomorphic perspective, while research done by R. T. Kingsford and others focused on ecological aspects of low gradient allogenic dryland rivers. Water resources are limited in desert regions, and with population growth and climate change acting as additional stressors on already stressed systems, there has been a push to understand more about how river systems evolve and interact with the environment.

2. Unifying Concepts of Dryland Rivers

2.1 Geomorphic Landforms

According to Kingsford and Thompson (2006), there are only two features that are unique to dryland rivers, billabongs (waterholes) and flood outs. While this is true, these features are more specific to low gradient rivers or rivers with very large seasonal variation in precipitation, such as those in Africa and Australia. A billabong is a small lake that is formed when water floods and then retreats from an area. These areas can act as refuges for aquatic organisms when flood water recedes, and are often areas of high bio productivity (Hillman, 1995). Flood outs happen when a river becomes unconfined and separates into many small channels. Flood outs are common on rivers in Australia and Africa, but not so much in other dryland regions (Nanson et al., 2002). While these landforms should be considered, as they can provide critical habitat, they are not a good characteristic for providing a unifying frame work for dryland rivers.

River forms vary greatly in dryland regions and can be braided, anatomizing, meandering, or relatively straight. Because river morphology is determined by a balance of resisting forces and driving forces, desert rivers can take many shapes. Please visit the fluvial geomorphology section of this wiki for more a more detailed explanation.

2.2 Sediment Supply

Sediment supply in dryland rivers varies just as geomorphic features do, though with sediment there are a few more generalizations we can make. According to Landbein and Schumm, (1958) sediment yield-precipitation relationships for small to medium catchments tend to be at their highest in arid to semiarid regions (Figure 2). Regions with more rainfall have enough vegetation (resisting force) to attenuate movement of sediment, whereas arid to hyper arid regions do not have enough rainfall (driving force) to produce enough overland flow to move substantial amounts of sediment. This relationship has been criticized as an ‘oversimplification’ (Nanson et al., 2002) and limited in its application across systems, though its use on some systems is probably appropriate. Topography must also be considered in this relationship, as the gradient of the system will affect how much sediment can be transported. Sediment is a key consideration in restoration because disruption of sediment supply by impoundments alters stream morphology and ultimately physical habitat. Sediment fluxes can also be of concern from an economic standpoint, as high sediment can harm recreational fishing waters.

Figure 2 - Sediment yield vs. precipitation relationships based on various models

Vegetation in drylands is sparse, at least when compared to more humid areas; this affects interception of rainfall and can reduce infiltration leading to an increase in overland flow (Ludwig et al., 2005; Graf, 1988, p. 121). This overland flow can then transport more material from surrounding slopes and into waterways. Timing and intensity of rain storms can enhance sediment flux in arid regions. Many dryland regions (American Southwest, Australia, India) are subject to high intensity rain storms from monsoonal systems or otherwise high intensity but low frequency events. A study conducted in the arid zones of the eastern Himalayas found that during an abnormal monsoon year (decadal scale events) twice the amount of normal sediment was transported in the river systems (Wulf et al., 2005). We can reasonably expect dryland rivers with low relief, such as those dominant in Australia, to have very different sediment supply and transport regimes. Based on gradient alone one might expect these systems to be dominated by suspended load (clay and silt), but according to work done on the Burdekin river in Australia, highly variable rainfall has a strong influence on what type of sediment can be transported (Amos et al., 2004).

2.3 Hydrology

Hydrology of any given watershed is unique. As with sediment supply, it is hard to place a group of rivers into one category, but there are a few properties of dryland rivers that will be useful in addressing restoration concerns. Drylands generally share three major hydrologic characteristics: 1) high evapotranspiration to precipitation ratios, 2) large in-stream transmission losses (seepage), and 3) “flashy” stream hydrographs due to physical characteristics of land forms and variability of precipitation. In addition to these three factors, allogenic systems tend to also have a flood pulse in the spring as snow melts, whereas some endogenic systems have a seasonal flood pulse related to monsoonal moisture.

Water budgets, using the concepts of the water cycle, are one way to quantify the amount of water entering and leaving a watershed. The main components of a water budget are: inflows (precipitation), outflows (evapotranspiration, surface and groundwater flow from the basin) and storage (soil moisture, surface water, ground water). One way of defining a desert is by the ratio of potential evapotranspiration (PET) to precipitation, where deserts have high PET to precipitation ratios. Seepage (loss of water into the ground) coupled with high PET leads to high transmission losses in desert watersheds. In some basins this could be considered the natural state of the system; where this becomes a restoration concern is when waters are diverted for uses such as irrigation and factors that affect transmission losses, such as increasing actual evapotranspiration (AET), are multiplied. Alluvial systems tend to have naturally high seepage rates. This is due to permeable, and often very deep, sediments resulting in loosing reaches (more water lost than gained). Much of this water enters the shallow alluvial aquifers. One example of this is the Salt River in Arizona where it is estimated that 29% of its flow is transferred to ground water storage (Graf, 1988, p.94).

Hydrology is affected by physical basin characteristics and the amount of vegetative or imperious ground cover (resistant bedrock or anthropogenic). Lithology, soil properties and vegetation in a watershed can affect the amount and depth of water infiltration. Many dryland systems have little vegetative cover, thin soils, and exposed bedrock. These are all conditions that lead to more overland flow and less infiltration. Infiltration and subsequent movement of water through the subsurface tend to attenuate flow from slope to valley floors and into river systems. In dryland systems a higher proportion of water is transported by overland flow and flows virtually unimpeded to the valley. This coupled with high intensity storms creates the ‘flashy’ hydrograph that is typical in many desert regions (Figure 3). These flows tend to be important hydrologic and geomorphic factors because less water can be transferred and stored in the ground and higher amounts of sediment can be transported in short periods of time.

Figure 3 - Typical flashy hydrograph (from the USGS)

Snowmelt from upland regions can be a large part of the water budget for allogenic dryland systems, for example the Rio Grande and Colorado rivers. Snowmelt in the spring provides pulses of water to the river. These pulses are important to the ecology of the stream especially for those organisms specifically adapted to this type of environment (Poff et al., 2007). Flooding and overbanking of the river help to sustain riparian vegetation, and scouring and deposition of sediments alter stream morphology habitat. The absences of this flood pulse in systems that have been altered by dams and impoundments is a major restoration concern in dryland rivers and will be discussed more fully in the section on disturbances in dryland rivers.

2.4 Ecology

Repair of ecological structure is one of the main aims of many restoration projects. Several factors contribute to why this is the case. Restoration projects are often undertaken in order to save an endangered species, such as the Rio Grande Silvery Minnow in New Mexico, United States. Other times ecosystem services need to be repaired to improve water quality. In those respects, dryland river ecology is much like any other river system. Some characteristics, while not exclusive to drylands, stand out as being important: high primary productivity, riparian vegetation, and boom and bust cycles.

Water and energy are the two main ingredients for life. Primary producers have the advantage in many dyrland rivers because of the abundance of sunny days and the lack of dense riparian vegetation that would otherwise shade the water. Primary production in dryland rivers can be double that of forest streams (Bunn et al., 2006). Because light is not a limiting factor in most dryland rivers other nutrients, such as nitrogen, become limiting nutrients (Grimm et al., 1981), but this is dependent on what else is happening in the rest of the watershed (e.g. agriculture, ground and surface water chemistry).

Riparian vegetation lining desert rivers is diverse in density, type, and ecological function. Deep canyon systems, like the Grand Canyon reach of the Colorado River (Figure 4), will have only thin riparian zones, though relatively cool moist environments in deep narrow side canyons can support a high diversity and density of organisms. This can be contrasted with rivers such as Cooper Creek in Australia where seasonal floodplain can be 100's of km2 (Bunn et al., 2006) though on a more seasonal basis than rivers with a more reliable perennial flow. Riparian vegetation has been at the forefront of restoration efforts in the American Southwest. Species introduced in the 19th and 20th centuries, such as the tamarisk, have altered riparian ecology and in some cases stream morphology (Graf, 1978) (see report on invasive species). The function of the riparian zone is broad and affects: 1) hydrology (Tabacchi,et al., 2000), 2) water quality (Dosskey et al., 2010) and 3) ecology (Naiman & Decamps, 1997). Bunn et al. (2006) state that “Desert rivers …represent the ecological arteries of dryland landscapes,” and riparian zones are a major part of this system.

Figure 4 - Floodplain riparian zones of Channel Country, Australia and thin riparian zones of the Grand Canyon, United States

The variable nature of dryland environments has an impact on the type of organisms that can colonize such a volatile system. Groups of organism that inhabit these highly variable systems tend to go through ‘boom and bust’ cycles. In ecology this is referred to as an ‘r’ selective species. R selective species produce a large amount of offspring, sometimes waiting until conditions are favorable, with the hopes that at least a small fraction of offspring will survive. In years that are favorable to the offspring, populations explode (the boom). In years that conditions are not as favorable, the population will fall or move from the area (the bust), but even if a small percentage of boom population survives, the species will be successful. How different organisms respond to boom and bust cycles depends. During bust periods many fish and aquatic invertebrates will die, though some invertebrates can encyst (Williams, 1987). Waterfowl populations can reach large numbers during boom cycles and then fly to other regions when drier conditions return. Boom and bust cycles are well documented on dryland rivers in Australian for micro-invertebrates (Jenkins & Boulton, 2007) and water birds (Kingsford et al., 1999), and to a lesser degree for species in the American Southwest such as the Rio Grande Silvery Minnow (Magana, 2012) and Riparian cottonwoods (Lytle & Merritt, 2004). Natural and anthropogenic disturbances to flow regimes in rivers that exhibit boom and bust character can be harmful to organism that are adapted to this sort of environment (Poff et al., 1997).

2.5 Climate

Climate varies vastly among dryland river basins, though amount of precipitation (sometimes coupled with PET) is what we use to define a desert. Precipitation in drylands is spatially and temporally variable, and regions can experience droughts that last months, years or even decades. Paleoclimate records show that some desert regions have experienced large scale climate fluctuations, including multiyear and decadal scale droughts (NCDC). Couple this with climate change, and some deserts could experience semi-permanent drought conditions (Gutzler & Robbins, 2011). Drought impacts all parts of the basin, not just the amount of flow in the river. Vegetative densities, sediment fluxes, and basin hydrology will all influence how the river responds geomorphically and ecologically. Biogeochemical cycles have also been shown to be affected by drought cycles (Dahm et al, 2003). Increases in wildfires due to droughts can alter sediment fluxes (University of Arizona) and water quality (Earl & Blinn, 2003), in turn affecting the ecology of the system. Understanding desert regions and their fluvial systems will become increasingly important if climate predictions come true.

3. Restoration Considerations and Goals in Dryland Rivers

3.1 Climate Change and Drought

Thus far, this report has discussed some of the factors that help unify desert rivers and their associated watersheds. Now that this background information has been provided, we can discuss how this can influence restoration efforts and future goals. Restoration efforts are many times focused on returning a system to some approximation of the past state of the system. Missing from the restoration conversation is the impact of future climate change and long term (decadal and millennial scale) climate fluctuations on restoration efforts. This section will briefly look at some of these factors and how they might influence our restoration.

3.1.1 Precipitation Variability

Restoration efforts that do not account for future change in the system could ultimately fail, costing the taxpayer money and wasting the effort of organizations responsible for restoration work. Variability and timing of precipitation is predicted to be a large factor in global climate change (IPCC, 2007; Barnett et al. 2005). Some of the impacts of climate change on desert regions include:
  • Decrease in the amount of snowpack and changes in timing of precipitation and snowmelt (Barnett, et al.,2005), affecting flow regimes.
  • Changes in sediment flux due to loss of vegetation (Ludwig et al, 2005), affecting river morphology and physical habitat.
  • Increase in forest fires causing changes in sediment flux (Ludwig et al, 2005), water quality (Earl & Blinn, 2003), and basin hydrology (Pierson et al., 2003).
  • Drought-induced changes in surface-ground water interactions affecting the hyporheic zone and biogeochemical cycling (Dahm et al, 2003).

Knowing these are possible future outcomes, how will we adjust our restoration goals? Part of the answer might be that we accept that certain goals can only be met on short term scales (years to decades), such as sustaining populations of endangered species. Looking at restoration on longer time scales, decisions could be made to accept that we cannot get back to historical conditions, and must abandon some goals so that a more sustainable dynamic can be achieved. Adaptive management is an option in regions with a lot of uncertainty; a good discussion of this strategy can be found in the adaptive management section of the wiki. Many of the issues listed above can be helped by evaluating the buffer zones along rivers. Riparian zones can act to attenuate sediment and nutrient fluxes into the river (Naiman & Decamps, 1997) and, as will be discussed in the next section, can act as buffers to climate changes.

3.1.2 Natural Resistance to Climate Change

Natural river systems have built in mechanisms for dealing with change in the environment. These mechanisms of change are viewed by humans as either negative (such as changes in sediment flux or movement of the river across the floodplain) or positive (riparian zones function as buffers). Restoration efforts tend to focus on making the environment look like we want it or protecting areas that we have deemed important, instead of restoring the environment to a state in which it can dynamically respond to change. While restoring the river system to a natural or historical state is impractical in many areas, incorporating many natural buffers into restoration plans could have huge benefits in the future. Of particular importance in this area are riparian zones and wetlands.

Naiman and Decamps (1997) document important roles that riparian zones can play in the riverine environment. Many of these roles can help to explain why riparian zones act as buffers to change. Seavy et al. (2009) identify four reasons why riparian restoration is important for a river's resistance to climate change:

  • Promote aquatic and terrestrial linkages
  • Provide thermal refugia for organisms
  • Enhance landscape connectivity
  • Attenuate flood due to more frequent extreme events

Up to a certain point positive feedback mechanisms help sustain these dynamics as the environment changes. This can give organisms in the system time to adapt to the change, or it can give the biological community time to develop into a community that is in equilibrium with the new environment (e.g. change species composition). Resilience to change does not necessarily need to mean maintaining the status quo; it can also mean slow change to a new normal. Seavy et al. (2009) recommend collecting seeds from plants across the watershed so that as the climate changes these plants can be worked into the successional environment; Incorporating the idea that healthy riparian corridors and river system might not reflect the historical river system could help many restoration efforts. This type of adaptive management that is not based on what the river has looked like in the past could help restoration efforts be successful on larger time scales.

3.2 Disturbances on Dryland Rivers

3.2.1 Dams, Impoundments and Channelization

By definition water is scarce in drylands. Because of this, development tends to cluster around river systems. The Rio Grande in the Southwestern United States provides a good case example for looking at the consequences of development of the flood plain and the river system. The Rio Grande basin near Albuquerque, New Mexico had been inhabited and developed by the Spanish since the 1500’s, and centuries before that by Native Americans. While the Spanish and Native Americans presumably changed the river system in many ways, it is modern development that has had the greatest impact on the river system. Post World War II the Albuquerque region experienced a population boom. This sudden increase in population stressed a region in which water resources were already fully allocated (and probably over allocated). Development of the flood plain at this point, while substantial, was mostly limited to agricultural practices. From 1930-1970 many dams were built on the Rio Grande and its tributaries for purposes of water storage, sediment control and flood control. During this time and to the present day, the floodplain has been developed for residential and commercial use following the overall trend in the southwestern United States (O'Donell, 1997) .

These developments have had a large impact on all parts of the Rio Grande River system (please see the Rio Grande and Chama sections of this wiki for a more detailed report). Channelization caused by development of the floodplain and construction of levees has restricted the Rio Grande to its current location in the basin, whereas historically (geologic time scales) this river has moved laterally across the valley. This, combined with sediment trapping in dams, has caused parts of the river to incise. Lateral movement is now seen as something that needs to be fixed in order to protect infrastructure and development along the river corridor. The biological community has also been affected by channelization. Plant and animal species are adapted to an environment where flooding and variable flows were part of the natural flow regime [add link and citation]. Cottonwood trees in the riparian zone need flooding in order to germinate [add citation]. The cottonwoods along the middle Rio Grande, for the most part, have been unable to germinate which has reduced the number of sapling that are replacing the now mature trees. Flow regime disturbance and physical habitat alterations have altered in-stream biota and disturbed the life cycle of species such as the endangered Rio Grande Silvery Minnow [need citation].

While the details of the disturbance on the Rio Grande might be unique, the implications of development could be seen on any dryland river. Restoration work is currently being conducted on the Rio Grande to repair some of the past physical damage to the system so that biological systems can return to a healthy state [citation, MRG Plan?]. Environmental flows (discussed here) are begin developed on many rivers to help mitigate some of the damage from dams. Many large scale concepts have also been developed to help us understand the natural river system, such as ‘The Natural Flow Regime’ (Poff et al., 1997) and ‘The River Continuum Concept’ (Vannote et al., 1980). These concepts are many times developed on systems that have very different characteristics than those of dryland regions. While these concepts can help us better understand rivers as a whole, restoration groups must be careful not to apply them too broadly.

3.2.2 Forest Fires

During much of the 20th century wildfire in the United States was seen, from a human perspective, as a natural event with negative consequences. Suppression of wildfire led to an increase in forest litter and fuel loads in many forests that were once adapted to episodic burning. Now instead of low intensity burns that clear out the understory and help some species seeds to germinate (Keeley, 1987), we are experiencing catastrophic fires that eliminate all vegetation, sterilize soils, and create soil crusts. This type of fire has a tremendous effect on the hydrology of a watershed. As discussed in previous sections vegetative cover acts as a resisting force to erosion, and loss of this cover increase surface runoff and erosion. Drought in many desert regions predicted to increase in both severity and length, increasing the probability of wild fires. Fires during the summer of 2011 in Northern New Mexico followed by monsoonal storms caused Frijoles Creek, a stream that averages 10 cfs, to experience 3000-4000 cfs flows (figure 5). The fire had not affected the canyon to a great degree, though major areas of the upstream watershed had been completely burned and left free of vegetation. Ash, fire debris, and dirt was carried through Frijoles Canyon, altering water chemistry and sediment dynamics throughout the system. Effects of the fire and subsequent floods were seen downstream in the Rio Grande beyond Albuquerque. This example shows the importance of whole watershed health when restoring or reclaiming portions of a river. Fuel loads have been recognized as a problem for several decades and currently prescribed burns and fuel load removal is part of many forest management plans.

Figure 5 - Fire damage from 2006 fire near Los Alamos, New Mexico and 2011 flooding in Frijoles Canyon, Bandalier National Monument, New Mexico

4. Conclusion

Each river system is unique, and placing rivers into a larger hierarchy that is meaningful for restoration purposes can be difficult. By identifying key features that unify most desert river systems, might not make it easier to come to larger decisions about restoration practices, but it can help water resources and land managers learn more about the processes shaping the river systems in desert rivers. Many desert regions are predicted to become warmer and drier over the next century. Knowing this and the consequences, such as water shortages and increase in wild fire incidence, can help prioritize decisions about how we can improve watershed health.

5. References

  • Amos, K.A.J., Alexander, J.A.N., Horn, A., Pocockà, G.E.D., and Fielding, C.R., 2004, Supply limited sediment transport in a high-discharge event of the tropical Burdekin River , North Queensland , Australia: Sedimentology, no. 51, p. 145-162, doi: 10.1046/j.1365-3091.2003.00616.x.
  • Barnett, T.P., Adam, J.C., and Lettenmaier, D.P., 2005, Potential impacts of a warming climate on water availability in snow-dominated regions.: Nature, v. 438, no. 17, p. 303-9, doi: 10.1038/nature04141.
  • Bunn, S. E., Balcombe, S. R., Davies, P. M., Fellows, C. S., Mckenzie-Smith, F. J., Aquatic Productivity and Food Webs of Desert River Ecosystmes, in Ecology of Desert Rivers, ed. Kingsford, R. T., Cambridge Univer. Press: Cambridge, p. 76-99.
  • Dahm, C.N., Baker, M. a., Moore, D.I., and Thibault, J.R., 2003, Coupled biogeochemical and hydrological responses of streams and rivers to drought: Freshwater Biology, v. 48, p. 1219-1231, doi: 10.1046/j.1365-2427.2003.01082.x.
  • Dosskey, P. V., Vidon, P., Gurwick, N. P., Allan, C. J., Duval, T. P., and Lowrance, R., 2010, The Role of Riparian Vegetation in Protecting and Improving Chemical Water Quality in Streams: Journal of the American Water Resources Association, v. 46, no. 2, pp. 261-277.
  • Earl, S.R., and Blinn, D.W., 2003, Effects of wildfire ash on water chemistry and biota: Freshwater Biology, v. 48, p. 1015-1030.
  • Graf, W. L., 1978, Fluvial Adjustment to the Spread of Tamsrisk in the Colorado Plateau, Geological Society of America Bulletin, v. 89, p. 1491-1501.
  • Graf, W. L., 1988, Fluvial Processes in Dryland Regions, Springer-Verlag, New York.
  • Grimm, N. B., Fisher, S. G. and Minckley, W. L., 1981, Nitrogen and Phosphorus Dynamics in Hot Desert Streams of Southwestern USA: Hydrobiologia, 83, 303-312.
  • Gutzler, D.S., and Robbins, T.O., 2010, Climate variability and projected change in the western United States: regional downscaling and drought statistics: Climate Dynamics, v. 37, no. 5-6, p. 835-849, doi: 10.1007/s00382-010-0838-7.
  • Hillman, T, 1995, Billabongs, floodplains and the health of rivers : Water (Australia) Vol. 22, no. 2, pp. 16-19.
  • IALC, Desert Facts,, accessed March 31,2012.
  • IPCC, 2007, Intergovernmental Pannel On Climate Change Fourth Assessment Report: Climate Change.
  • Jenkins, K.M., and Boulton, a. J., 2007, Detecting impacts and setting restoration targets in arid-zone rivers: aquatic micro-invertebrate responses to reduced floodplain inundation: Journal of Applied Ecology, v. 44, p. 823-832, doi: 10.1111/j.1365-2664.2007.01298.x.
  • Keeley, J.E., 1987, The Role of Fire in Seed Germination of Woody Taxa in California Chaparral: Ecology, v. 68, no. 2, p. 434-443.
  • Kingsford, R.T., Curtin, A.L., and Porter, J., 1999, Water Flows on Cooper Creek in arid Australia determine ` boom ’ and ` bust ' periods for waterbirds: October, v. 88, no. April 1998, p. 231-248.
  • Kingsford, R. T. and Thompson, J. R., 2006, Desert or Dryland rivers of the World: an Introduction, in Ecology of Desert Rivers, ed. Kingsford, R. T., Cambridge Univer. Press: Cambridge.
  • Landbein, W. B. and Schumm, S.A., 1958, Yield of Sediment in relation to Mean Annual Precipitation: Transactions of the American Geophysical Union, 39, p. 1076-1084.
  • Lytle, D., and Merritt, D.M., 2004, Hydrologic Regimes and Riparian Forests?: A Structured Population Model for Cottonwood: America, v. 85, no. 9, p. 2493-2503.
  • Ludwig, J.A., Wilcox, B.P., Breshears, D.D., Tongway, D.J., and Imeson, C., 2005, Vegetation Patches and Runoff-Erosion as Interacting Ecohydrological Processes in Semiarid Landscapes: Ecological Society of America, v. 86, no. 2, p. 288-297.
  • Magaña, H. a., 2012, Habitat use of the Rio Grande silvery minnow (Hybognathus amarus) during a long-term flood pulse in the Middle Rio Grande, New Mexico: Environmental Biology of Fishes,, doi: 10.1007/s10641-012-9977-5.
  • NCDC, National Oceanic and Atmospheric Administration: National Climate and Data Center,, Accessed on 2-20-12.
  • Naiman, R. J. and Decamps, H., 1997, Ecology of Interfaces: Riparian Zones: Annual Review of Ecology and Systematics, v. 28, p.621-658.
  • Nanson, G. C., Tooth, S. and Knighton, A. D., 2002, A Global Perspective on Dryland Rivers: Perceptions, Misconceptions and Distinctions, in Hydrology and Geomorphology of Semi-Arid Channels, ed. Bull, L. J. and Kirby, M. J., Chichester: John Wiley and Sons ltd.
  • O’Donell, M., 1997, Land-Use Trends in the Southwestern United States, from the USGS, Accesed on March 30, 2012
  • Pierson, F.B., Robichaud, P.R., Kenneth, E., and Moffet, C.A., 2003, Impacts of Fire on Hydrology and Erosion in Steep Mountain Big Sagebrush Communities: The First Interagency Conference on Research in the Watersheds,
  • Poff, N.L., Allan, J.D., Bain, M.B., Karr, J.R., Prestegaard, K.L., Brian, D., Sparks, R.E., Stromberg, J.C., and Richter, B.D., 1997, A paradigm for river conservation and restoration: BioScience, v. 47, no. 11, p. 769-784.
  • Seavy, N.E., Gardali, T., Golet, G.H., Griggs, F.T., Howell, C.A., Kelsey, R., Small, S.L., Viers, J.H., and Weigand, J.F., 2009, Why Climate Change Makes Riparian Restoration More Important than Ever?: Recommendations for Practice and Research: Ecological Restoration, v. 27, no. 3, p. 330-338.
  • Tabacchi, Eric and Lambs, Luc and Guilloy, Hélène and PlantyTabacchi, Anne-Marie and Muller, Etienne and Decamps, Henri ( 2000) Impacts of riparian vegetation on hydrological processes. Hydrological Processes, Vol. 14 (n° 16-17). pp. 29 59-2976. ISSN 0885-6087
  • University of Arizona,
  • USGS, What is a Desert, accessed 2/24/2012
  • USBR, United State Bureau of Reclamation, accessed on 2-20-2012
  • Vannote, R.L., Minshall, W.G., Cummins, K.W., Sedell, J.R., and Cushing, C.E., 1980, The River Continuum Concept: Canadian Journal of Fish and Aquatic Science, v. 37, p. 130-137.
  • Williams, D. D., 1987,The ecology of Temporary Waters, Croom Helm:London
  • Wulf, H., Bookhagen, B., and Scherler, D., 2010, Seasonal precipitation gradients and their impact on fluvial sediment flux in the Northwest Himalaya: Geomorphology, v. 118, no. 1-2, p. 13-21, doi: 10.1016/j.geomorph.2009.12.003.