This laboratory exercise is designed for students who have had some exposure to rivers and geomorphic features associated with rivers, such as cutbanks, oxbows, point bars, bars, braided channels, meandering streams, and channel migration. They should have at least an introductory geology course, with some exposure to landforms and rivers. It certainly will help students if they have some exposure to plate tectonic features, such as the Altiplano (tectonic plateau), uplift, and foreland basins. They should also have some exposure to climate variation around large mountain ranges. I implement this exercise in an undergraduate junior-senior level course which serves geology, hydrogeology, and environmental science majors.

Students should be able to recognize cutbanks, oxbows, point bars, bars, braided channels, meandering streams, and channel migration in aerial imagery. Students should also be able to navigate to a latitude/longitude in Google Earth. They should be familiar with searching online for geologic information, such as information about tectonic uplift rates, climate, and geoscience research articles.

This exercise serves as an inquiry-based project in a laboratory setting. It takes one to two 3 hour lab periods to complete. It serves as an introduction to river dynamics, and can stand alone without too much pre-cursor material.

After completing this exercise, students will be able to:

• Recognize geomorphic features of meandering rivers
• Appreciate erosion and depositional processes of meandering rivers
• Access online data sources of land use change
• Collect and analyze data to develop and test hypotheses about controls on channel migration rates

Students are introduced to time series satellite imagery (1984-2012) for an area in the western Amazon basin, and are asked to document river features, and identify channels which migrate laterally and change location (meander cutoffs). Further, they must identify the most laterally active river, and determine, if they can, why that section of the basin has such a dynamic river. Thus, students must collect data, analyze the data, generate hypotheses, and test the hypotheses. They will be confronted with coupled tectonic/climate/erosion systems, and forced to deal with limited information.

Students develop skills in collaboration (this is best implemented as a small group activity). They also develop skills working with spatial data, generating charts and figures, and writing a scientific report.

Students access Google's Earth Engine, which hosts a global data set of Landsat imagery from 1984-2012 (the links to rivers is in the Data section). They must then find the most actively migrating river(s) by zooming in and out and navigating around the region. They must describe aspects of channel change (meander migration, avulsion, cutoffs, oxbow generation and evolution). Once they are comfortable with documenting channel features and behavior, it is time to ask the question, why is the Rio Ucayali so active? Is it bigger, or smaller, than other rivers? Is there deforestation in the area? Is it in a tectonically active area which might uplift the stream, or uplift mountains, which erode faster and deliver more sediment to this part of the river system? Students need to describe what they will measure, and then test hypotheses by making the measurements, analyzing the results, and writing a summary of their findings.

Students can work singly or as a group on this project. Actually, a group activity would be ideal, as hypothesis generation and testing is a challenging task, and collaboration broadens the scope of questions and approaches to testing hypothetical relationships.

Students are likely to find the hypothesis-generation part of this exercise to be very challenging. You can supply a couple hypotheses which might explain why a given river is so much more laterally active, such as: higher uplift rates in the upstream region deliver excessive sediment loads to this river, and cause increased lateral migration; or, clearcutting of forests around this part of the river has increased sediment loads, which has forced the river to migrate more. Or, perhaps mining activity or a new reservoir has altered sediment supply to the river. You can also invoke the role that the kind of material carried by the river and deposited in banks might control how fast the river can migrate. In all cases, students need to identify data they need to test the hypothesis (uplift rates, evidence of more clearcutting near this river than others, a type of geology upstream which would produce a different material for suspended load for a stretch of active river, and the like).

Students generate the following items in this exercise:

• Image captures of stream location, with annotations of meandering stream features;
• Statement of hypotheses for why the Rio Ucayali is so much more active;
• Table of data collected as a means of testing the hypothesis;
• Analysis of the data;
• Summary of the test of the hypothesis.

These items form the basis for an evaluation of their efforts.

• Figures should have descriptive titles and captions. Maps need directional indicator, scale, and annotations to highlight key features.
• Hypotheses must pose a relationship: for instance, the Rio Ucayali is more active because of a region of very active uplift just upstream. The test is evaluated based on how well it can be measured, and by the strength of the connection between the driving force and response of the river. For instance, higher uplift results in faster erosion, which forces rivers to carry more sediment, which leads to more sediment movement and storage, and thus more migration. One could determine if uplift is more rapid in this section of the Andes, based on earthquake activity, or other scientific publications on uplift rate, or simply the height and ruggedness of landscapes nearby. A comparative study could be implemented between adjacent streams. Deforestation could also be tested, based on estimates of disturbance over time, or just the presence of deforestation along the Rio Ucayali. Size of the river could play a role, if larger rivers carried more sediment and have greater power, they might be more prone to migrate. The causal relationships could be ranked as strong, moderate, weak, or absent a relationship.
• Data collection and analysis can be ranked according to excellent, good, fair, poor, unacceptable.
• Summary statement can be ranked according to excellent, good, fair, poor, unacceptable.

Earth Engine Views of Rivers and Land Use Around the World

• Irrawaddy, braided river https://earthengine.google.org/#timelapse/v=20.37647,94.70562,7.924,latLng&t=2.79
  
• Irrawaddy, distributary section https://earthengine.google.org/#timelapse/v=17.0467,95.56949,8.532,latLng&t=2.05
  
• Brahmaputra, braided river https://earthengine.google.org/#timelapse/v=25.91176,89.59344,7.924,latLng&t=1.74
  
• Amazon River, distributary section (STABLE!) https://earthengine.google.org/#timelapse/v=-1.06135,-51.26929,7.772,latLng&t=2.01
  
• Arkansas-Mississippi River junction (both meandering rivers) https://earthengine.google.org/#timelapse/v=33.84989,-91.16697,9.748,latLng&t=0.00
  
• Mine waste runoff effects on Ok Tedi River, Papua New Guinea https://earthengine.google.org/#timelapse/v=-5.23546,141.19554,10.356,latLng&t=1.06
  
• Rio Madre de Dios in Bolivia, with effects of deforestation visible: https://earthengine.google.org/#timelapse/v=-12.80292,-70.45737,8.783,latLng&t=0.00
  

• Google Earth Engine home: https://earthengine.google.org/#intro
  
• Google Earth Engine, Global Forest Change, 2000-2012: http://earthenginepartners.appspot.com/science-2013-global-forest
  
• Google Earth, focus on lat/lon: -8.50, -74.40. In the layers, enable Earthquakes, and under More, enable Water Body Outlines (shows which rivers have moved since the creation of this layer).
  
• As an alternative or as additional information on earthquakes, see Earthquake size, location, and time for Peru (a Google Fusion Table), with data from NOAA's National Geophysical Data Center for damaging earthquakes, along with focal depths and magnitudes (http://www.ngdc.noaa.gov/nndc/struts/form?t=101650&s=1&d=1 ), and http://maps.ngdc.noaa.gov/viewers/hazards/?layers=2&extent=-180,70,180,-70# for a searchable map.
  

The Problem River

• Rio Ucayali in eastern Peru in Earth Engine: https://earthengine.google.org/#timelapse/v=-8.36311,-74.56119,5.892,latLng&t=0.57

Geoscience Articles

Aalto, Rolf et al. "Episodic sediment accumulation on Amazonian flood plains influenced by El Nino/Southern Oscillation." Nature 425.6957 (2003): 493-497.

Aalto, Rolf, Thomas Dunne, and Jean Loup Guyot. "Geomorphic controls on Andean denudation rates." The Journal of Geology 114.1 (2006): 85-99.

Aalto, Rolf, et al. "Fluvial transport of sediment across a pristine tropical foreland basin: channel-flood plain interaction and episodic flood plain deposition." FJ Dyer, MC Thoms, & JM Olley. International Association of Hydrological Sciences, Publication 276 (2002): 339-344.

Aalto, Rolf, and Charles A Nittrouer. "210Pb geochronology of flood events in large tropical river systems." Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 370.1966 (2012): 2040-2074.

Giardino, John R., and Adam Lee, 2011, Rates of Channel Migration on the Brazos River, http://www.twdb.texas.gov/publications/reports/contracted_reports/doc/0904830898_Brazos.pdf.

Howard, Alan, 2009, How to make a meandering river, Proc Natl Acad Sci U S A. Oct 13, 2009; 106(41): 17245–17246. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2762660/ . Check out the references in this article!!!

Safran, Elizabeth B et al. "Erosion rates driven by channel network incision in the Bolivian Andes." Earth Surface Processes and Landforms 30.8 (2005): 1007-1024. https://onlinelibrary.wiley.com/doi/full/10.1002/esp.1259