Mastering GIS

What is an operator? Describe and give examples of each of the following: arithmetic operators logical operators spatial operators Boolean operators

1. Arithmetic operators like +, -, /, and * allow for the combination of multiple attributes mathematically to help make queries scale based on these combinations. 2. Logical operators are operators like <,>, and = that help test if logical equivalencies or inequalities hold true for some features. 3. Spatial operators like containment, intersection and proximity test whether features meet some specified geospatial qualifications. 4. Boolean operators like and & or help add specificity and more complexity to queries to allow for more precise selection of features based on multiple queries.

For each of the following queries, state whether the syntax is correct or incorrect. If incorrect, explain why. ZONE = 'COM' AND ZONE = 'RES' COVTYPE = 'SPRUCE' AND CROWNCOV > 50 POP2000 > 2000 OR POP2000 < 9000 INCOME < 10000 AND INCOME > 50000

1. Incorrect, Zone cannot be both of those values at once 2. Correct 3. Technically Correct, but this will select all of the features. If that's what you're going for, it's ok. However, it is likely that an AND operator would provide a more interesting subset 4. Incorrect, a value cannot be simultaneously less than 10000 and more than 50000

What is the cardinality of each of the following relationships? students to college classes states to govenors students to grades counties to states

1. Many-to-many 2. One-to-one 3. Many-to-one 4. Many-to-one

A 1:20,000,000 scale map of the United States displays the interstates with a line symbol that is 3.4 points wide. There are 72 points to an inch. What is the uncertainty in the location of the road due to the width of the line used to represent it? Give the answer in feet and miles.

3.4/72 = .0472 map inches .0472 map inches * 20,000,000 = 944,444.44 inches, which is 78703.70 feet and 14.91 miles.

Explain the difference among the terms feature, feature class, and feature data set.

A feature is a basic vector object, like a point or a line. A feature class is a collection of features that contains only features of a certain type, like a class of transmission lines or highways. They can only be one type of feature. A feature dataset contains multiple different feature classes.

Which would take more storage space, a layer file showing all the US counties or a layer file showing all the US states?

A layer file only contains the path to accessing geospatial data. As a result, it is unlikely that they would be markedly different from one another.

What is a query? Give an example. Why might you use one when exporting data to a geodatabase?

A query is an option given to a GIS user to allow them to specify what subset of a dataset should be imported or exported elsewhere. For example, I could choose to only export route 2 from the Williamstown roads shapefile to the U.S. highways file. One might do this to remove extraneous data or to limit file sizes.

Can you identify reasons that aerial photography might yet retain a role, even in competition with the strengths of lidar data?

Aerial photography is still aesthetically preferable and requires far less processing to make many determinations. Aerial photography can help provide a sanity check for the findings of LIDAR data as well, ensuring that what is visually observed matches what is produced by LIDAR.

Describe the difference between using Statistics and using Summarize functions on a field.

Calculating statistics is a good first step for analyzing the data, which involves calculating some basic, general stats about the data in your sample, generally without ties directly to its geographic distribution. Summarize, on the other hand, allows you to create tables that give valuable statistics based on a categorical attribute field in a way to better understand how the data are distributed based on a specific field.

Why might change detection using post-classified images instead of pre-classified images be problematic?

Change detection using post-classified images might be problematic because shifts in colors, patterns, and spectral ranges might intensify or minimize differences between data sets, affecting final interpretation. Processing pre-classified images through change detection generally presents a more trustworthy analysis between data sets. Post-classified images may be used for change detection, though analysts must keep these external variables in mind.

What two approaches are helpful in digitizing complexly related polygons?

Divide and Conquer, which involves digitizing a larger area, then manually specifying the boundaries of polygons using the Cut Polygons tool. Adding Territory, which uses the Auto Complete polygons tool to build the polygons from a blank slate.

What are the differences between using map topology and using planar topology during editing? How are they similar?

Editing using map topology uses cluster tolerance that helps ensure that boundaries have integrity, and do not overlap or have gaps. To accomplish this, features have edges and nodes that other features can snap to. Editing map topology involves editing these edges and nodes. Editing with planar topology, on the other hand, automatically snaps together clusters that are within a certain distance from one another, and then marks errors that show where that distance could not be adhered to. While editing with both map topology and planar topology can fundamentally alter the shapes of feature classes, using planar topology can do so without the specific intent of the user. Both use a cluster tolerance, but map topology uses far more user input to make edits.

True of False? Because you can Undo edits, it is not necessary to make a backup copy of any data set before editing. Explain your answer.

False. If the ArcMap system is overloaded or a glitch occurs, there is potential that the data set could become corrupted, which could disrupt data or restrict access to that information. As a result, even if user-made edits can be undone, unintended system errors could lead to losses. Thus, make a backup copy somewhere.

Why do you think the Mappa Mundi and other ancient maps are considered such treasures? What can we learn from the some of the earliest attempts to map space?

First and foremost, early maps like this offer a rare insight into the worldview of the people of the past. Take the Mappa Mundi for example. As Garfield notes, the message of the map is that "we've done our work in this place, for the inhabitable world is laid down on the back of a calf." (57). These sorts of shortcomings point to the hubris in the past, but also put perspective on modern claims of complete and absolute knowledge. These ancient maps reflect a worldview that has proven to be false, which in itself casts doubt on fundamental truths that our contemporaries hold dear. Additionally, while they are abstractions, they are still somehow familiar. Even though they represent falsehoods, we can connect with the similarities. For example, the wear and tear on the Hereford portion of the Mappa Mundi (56) show that people have an interest in where these past projections of the world connect with their own lives. Although there is a gaping canyon of space, time, and knowledge that separates the map-viewer from the map-maker, the shared realities of their world and the modern world build a strong connection to the past. In evaluating the evolution of maps through time, we find an evolution of human consideration in their place in the world, and at all time, we are acutely aware that we fall somewhere on that continuum.

Describe 2 reasons that you would pursue an unsupervised classification instead of a supervised classification.

First, an unsupervised classification requires far less human work time, as the machine is doing most of the processing. Another potential reason that one would pursue an unsupervised classification is because the lack of human involvement removes potential biases. In a supervised classification, pre-conceived notions held by the analyst may skew the results of the classification.

The image below is a map featuring the Great Trigonometrical Survey of India of the early 19th century. What were the two principal geographical breakthroughs associated with the survey? And how was triangulation used in map making at this time?

First, the Great Trigonometrical Survey of India yielded the best evidence to date of the curvature of the Earth. The measurement of India from top to bottom spanned 1600 miles, and had significant implications for future map makers as well, as they based their maps on this impressive meridian. Additionally, the measurement of the height of Mount Everest, while ill-conceived and deeply unnecessary, was impressively accurate, as the estimate was well within the variation seen in the height of Everest today. This measurement was a testament to the momentous efforts made by the British to cement their hold over India with a cartographic fist. As for the general uses of triangulation, it made a better map-making strategy for creating strategically useful maps, as it was highly effective at tracking distances between landmarks. Although labor intensive, these surveys created accurately scaled maps that reflected geographic information in a manner that could be directly converted into actionable distance information. By taking triangulated measurements, map-makers could both verify the accuracy of their maps mathematically and measure areas that were geographically challenging by creating triangles around impasses that make direct measurement impossible.

Explain the steps used to create a digital elevation model (DEM) from lidar data.

First, the LIDAR creates a representation of the surface that is similar to an image. However, the returns from this initial path are likely improperly placed, so a grid is interpolated to the image, which allows the data points to accurately represent both position and distance. The resulting image is a DEM.

The image below is a photograph of Fra Mauro's 15th century map of the world, shown in the orientation that Mauro intended. Why is the map considered to be a cartographic masterpiece? What do you find significant about the work? What do you find troubling?

Fra Mauro's map is considered a masterpiece certainly in part because of its intricacy. Drawing extensively from the work of Marco Polo, the map is known for the level of detail of the text on the map, as it contained almost 3000 place names (76). This intricacy of the world is significant because it was indicative of a world getting to know itself better. The travels of Marco Polo were only made possible by the establishment of trade routes across massive distances, which brought goods and knowledge to people a world away. The same zeal that led Fra Mauro to label thousands of places likely also contributed to the explosive exploration that would occur in the following centuries. However, given that Fra Mauro drew heavily on the work of Marco Polo, he left himself open to the inaccuracies of Marco Polo's stories and depictions of much of the Eastern world. Not only was leaning heavily on an unreliable source a problem for accuracy, it also may have introduced bias into the map itself. If Marco Polo misrepresented a place in "Travels," that misrepresentation could be potentially disseminated as fact in one of these maps. Additionally, the depiction of what appear to be places of worship on the map, while perhaps useful to contemporaries, could skew apparent power to areas of note or knowledge to Fra Mauro, and may not depict the most culturally significant places for locals. Although this is an almost completely unsolvable problem, given that Fra Mauro was not embedded in all of the communities he was mapping, he is likely unable to depict all of the locales he is mapping with the cultural understanding necessary to determine which buildings are of the most importance.

The image below shows the panels of Martin Waldseemüller's 16th century map of the world. Describe the two elements of the map that make it cartographically and historically important (the top row of panels has a hint)? And if you could undo Waldseemüller's mistake, what would you have chosen as a more appropriate name and why?

From a cartographic perspective, one of the most important contributions of Waldseemuller's map was the depiction of the Pacific Ocean between the New World and Asia. Its existence was not confirmed until 1522 (113), but it appeared on Waldseemuller's map anyway, both in the map itself and in the miny globes that appear at the top of the map. Additionally, it also was the first to fully depict something close to the Western Hemisphere (save for perhaps Vinland). While there are some mistakes, what we know as the world is beginning to take shape in Waldseemuller's map. However, the most colloquially significant happening that came about through Waldseemuller's map was the naming of the New World as "America" after explorer Amerigo Vespucci. Not only are the new continents referred to as America, but there is also a depiction of Vespucci at the top of the map, indicating that he is the explorer responsible for much of the information, along with Ptolemy. While Waldseemuller later corrected this mistake, the damage was done, and America became the name for this New World. If I could undo this mistake, I would certainly not name the New World after Columbus, who was both navigationally challenged (people knew the size of the Earth, and there was no way he could have made it to Asia with the supplies on his ships) and a murderous colonizer. Instead, I'd follow Waldseemuller's later lead and name it Terra Papagalli (land of parrots) because that is far more innocuous and respects the unique qualities of the land itself.

What is geoprocessing? In what different ways can commands be executed?

Geoprocessing is manipulating geospatial data to produce more interesting or specific results. It utilizes the tools inherent in ArcGIS to create more informative geospatial datasets. These tools can be used through a model builder, which provides a graphical user interface to use geoprocessing tools. This is a great option for research because it allows for testing of different parameters, and keeps high quality records of the processes performed. ArcToolbox can also be used, and is great for geoprocessing on the fly. Scripts allow for more sophisticated and replicable geoprocesses, but require coding experience to work effectively.

Lidar has often been used to estimate the height of forests. Explain how this is possible.

Given that LIDAR can interpret different kinds of returns from the illumination it can send downwards, it would identify the primary returns as the tops of the forest, because that is what is returned first. With that elevation recorded, the secondary returns could help reveal the elevation of the ground. Subtracting the two values would yield the forest height.

What extra step is performed when projecting rasters that is not needed when projecting vector data? What happens during this step?

Given that rasters in two coordinate systems will have a different shape, the grid needs to be resampled to maintain accuracy. In this step, a user needs to specify a new cell size to project the raster.

What two options may be used to handle one-to-many relationships in a spatial join?

Given that there will be many potential matches for each datum on the left side, a summarized join must be employed. This can take the form of a summarized inside join, wherein many features within a certain area are summarized by some variable and then appended to a larger table. Alternatively, a summarized distance join involves summarize many features based on their distances to a smaller set of features and allocating a summarized values to the smaller set.

Describe the difference between the terms grid and raster.

Grid is a type of raster file that is traditionally used in conjunction with the spatial analyst tool and is unique to ArcGIS. Rasters, on the other hand, can come in many other forms and are loosely defined as files that share geospatial data using individualized pixels that can vary smoothly across a continuous set of values.

Explain what you would need to do for map labels to increase or decrease in font size when the user zoomed into or out of your map.

I would need to set a reference scale. When I do that, the labels will increase in size when I zoom in from the reference scale and decrease in size when I zoom out.

You have a shapefile with a UTM Zone 10 NAD 1983 coordinate system, and you want to bring it into your sate database, which uses the Oregon Statewide Lambert coordinate system. Explain your next steps.

I would use the project tool to project the shape file using the UTM into the Oregon Statewide Lambert Coordinate system. This should ensure that there is limited bias in the analysis.

You have a shapefile with an Unknown coordinate system, but a file on the web site says the coordinate system is UTM Zone 13 NAD 1983. Explain your next steps.

If the area of interest for the shapefile lies in only one UTM projection zone, and that zone is 13, then you're in great shape! If it lies in multiple zones, I would work to determine if the map would be better suited in a different zone (if more of the interest area falls in another zone). If it spans many zones, I would likely project it to a more global coordinate system. Regardless, I should ensure to use define projection to delineate what coordinate system was ultimately chosen.

What's happens during digital image classification?

Image classification uses image processing techiques, often digital, to extract meaning from an image and attempt to make sense of what appears in an image. In its simplest form, each pixel is classified based on its brightness into a certain class in the hopes of classifying pixels in a manner that is practical. More complicated classification processes involve concepts like image texture, wherein multiple pixels are classified based on how they relate to one another.

Below is a portion of John Snow's map of Soho (London), drawn in 1854. Why do many consider the work to be the first GIS analysis ever conducted? What methods would you employ in ArcGIS to confirm Snow's finding?

In his map, John Snow uses geospatial data to compare the site of a contaminated water pump to the number of deaths from cholera in the surrounding area. Given that people were more likely to draw water from the nearest pump, he essentially did a simple distance join on the dead patients, and determined that a large percentage of them had the broad street pump as their nearest water source. He, of course, did not have access to the same geospatial analysis tools that we have today, but his geospatial analysis very much mirrors modern GIS work. As a result, performing Snow's amazing and groundbreaking analysis using this technology would not be all that difficult. With a feature class containing points of water pumps, and another containing the addresses of those who died from cholera, I would do a simple distance join of the pumps to the addresses, assigning each afflicted individual with their nearest water pump. I would then symbolize (color) the patients by the nearest pump, and hopefully, determine that the broad street pump was the culprit. To add more detail, I would potentially search for the contamination using a water flow feature class that would show where groundwater fed each pump. That, in turn, would allow for targeted analysis of what was contaminating the water pump.

Explain what it means to produce a map with good balance and clear visual hierarchy.

Intentional use of negative space, prioritize the elements that you are trying to highlight through size/color. All elements should be understandable, but the most important should be the most prioritized. Unnecessary space should be minimized.

Outline the procedure for ordering Landsat images for a study area near your home.

Landsat imagery is not all that difficult to access. The USGS Earth Explorer has all of the Landsat imagery, and one can simply outline the area of interest and zoom in using the Earth Explorer's interactive tool. GLOVIS can also offer free access to Landsat images through the internet.

Explain the Landsat system. What was it designed for? How long has it been in existence? How is it currently supported?

Landsat is a series of satellites designed to create repeated, intertemporally distinct images of Earth's surface. Landsat was first launched in 1972 after being proposed for scientific and technological purposes by scientists in the U.S. government. It was designed to allow for repeated observations of the Earth's surface for scientific and civil service purposes. The U.S. government still administers the Landsat system, but it lacks a specified home and has interest from both NASA and the USGS.

Many observers believe that the increasing availability of lidar data will displace the current role of aerial photography for many applications. What are some of the reasons that might lead people to believe that lidar could replace many of the remote sensing tasks fulfilled by aerial photographs?

Lidar data can give very detailed views of topography, surface elevation and vegetation layers, as well as urban landscape views often captured by aerial photography. However, LIDAR is not sensitive to weather to the degree that aerial photography is. As a result, LIDAR could replace aerial photography for these functions.

Review some of the strengths of lidar data relative to Landsat data.

Lidar is particularly adept at evaluating terrain, as it can separate vegetation from terrain. Given that lidar can return precise surface elevation values, it can also be valuable in building 3-d models of urban settings. This same elevation sensing can also help build floodplain models and help route roads and pipelines.

The image below is a map of the world according to Gerardus Mercator's famous projection, still used by Google Maps. What was Mercator's motivation for the projection? After some research, what do you think is a better alternative to the projection and why?

Mercator was interested in accurately depicting the latitude and longitude lines on the map. Inspired primarily by a desire to help navigators more effectively track their courses, he wanted to create a map that maintained an accurate map that did not distort latitude and longitude. Unfortunately, while this was an excellent addition to the field of map-making from a practical, navigational perspective, the spread of the map had some unfortunate unintended consequences. Primarily, to achieve a robust latitude-longitude grid, Mercator's map distorted the sizes of the continents. Unfortunately, this led to an overrepresentation of the size of places near the poles and an underrepresentation of the size of landmasses near the Equator. Given that much of the power in the colonial world was focused in the North Atlantic, this led to problematic oversizing of the places that also had concentrated wealth and power. As a result, for casual viewing, a map like the Gall-Peters, in spite of its horrendousness, is likely the best possible option, as it depicts the size of the continents far more accurately. However, given the aesthetic challenges of the Gall-Peters, a compromise map like the Laskowski projection likely strikes a good balance between aesthetics, shapes, and size accuracy.

Explain how single color images of satellite data are made.

Multiple images are taken in different portions of the electromagnetic spectrum. Individually, these images emphasize different features of Earth's surface. When combined, these layers become a single-color image that balances these emphases into a single image.

Write a valid SQL expression to select all counties whose names begin with the letter Q.

SELECT*FROM counties where Name = `Q%'

Why should distance joins always be performed on layers with a projected coordinate system? What kind of projection should be used?

Projected Coordinate Systems preserve distance. Also, the projection used needs to preserve distance over the area in question. If for example, the distance between a site in Colorado and a site in New York is being calculated, a projection that only preserves distance in the mountain time zone should not be used.

Describe some of the advantages and disadvantages of using satellite observations of the Earth instead of aerial photography (taken from an aircraft).

Satellite observations were more systematic than aerial photographs and allowed for apples-to-apples inter-temporal comparison of images over time. Satellite observations also led to the marriage of aerial imagery with computers and digital analysis. However, in the beginning, satellite imagery also lacked detail and required expensive launches into space.

What characters should be avoided when naming GIS files, folders, and map documents? Explain why these restrictions are necessary?

Spaces and special characters like $ and # should be omitted. If these rules are not followed, the program may crash, and folders must be completely renamed from scratch or the data may be lost.

List the types of data sources from which tables may display data.

Tables can display data from a shapefile or geodatabase, but also from data sources that are not geographic in nature, like a database or an Excel spreadsheet.

What attribute fields will be present in a layer resulting from a dissolve?

The only resulting attribute will be the attribute upon which the dissolve was performed, and any others that are chosen

Explain the difference between thematic rasters and image rasters.

Thematic rasters are representations of data in a raster format. They could take the form of a raster showing elevation or land use polygons. Image rasters, on the other hand, are taken directly from an image, like an aerial photograph or a transmission from a sattelite.

Imagine you are looking at a geodatabase that contains 50 states, 500 cities, and 100 rivers. How may feature classes are there? How many features? How many attribute tables? How many total records in all the attribute tables?

There are 3 feature classes, 650 features, 3 attribute tables, and 650 total records in the attribute tables.

Below is a modern redrawing of the 1807 version of the Commissioners' map for Manhattan, a few years before it was adopted in 1811. What was significant about this map? What criticisms did it face and why?

This map was notable primarily because of its link to New York's urban planning at large. Its depiction of the grid system, which would become iconic and foundational to the city's identity and development, was so clear and prominent on this map. Some contend that the perspective, viewing the city from on high, was indicative of the city's drive to be more and do more. The map, however, was not universally popular, and it faced criticism for lacking open space and for being concerningly uniform. It valued pragmatic planning over pesky aesthetics and free-flow, which made some observers uneasy. Some even criticized it simply for being too bold. In all, despite these criticisms, Manhattan's grid, and by proxy, the map that ambitiously charted its future, has become a critical part of the fabric of the city, and this map's ties to the roots of the grid make it a fascinating piece.

What's the point of training areas in supervised classification?

Training areas are specified by the analyst to be of a certain type desired by the classification. Once these areas are identified, the classifying program searches for similar areas and classifies them as laid out by the analyst.

What's the difference between unsupervised and supervised classifications?

Unsupervised classifications rely heavily on the digital processing algorithm, whereas supervised classifications involve significant input from the individual running the classification to clarify the objectives and identify the areas of interest.

What are ways that remotely sensed images differ from maps?

Whereas maps are purely abstracted versions of a true geospatial reality, remotely sensed images are direct imagery of the Earth acquired without aerial photography. Remotely sensed images are pure data, whereas maps have an inherent human element.

In these 2 chapters, Simon Garfield argues that even with the prevalence of digital maps (e.g., Google Maps), there's an aesthetic associated with paper maps and travel guides that will consistently draw us to them. Given the readings and your own experience, construct a brief argument in support of or against Garfield's view.

While I agree that there is something that holds aesthetic value for paper maps, that aesthetic is unlikely to cause people to voluntarily use them instead of digital maps. Digital maps are just too practical, and while many people on the older side may still prefer unfolding an enormous roadmap, the growing population of "digital natives" will simply never need to understand how to handle these cumbersome maps. That does not mean they will not have a place, but it is unlikely that place will be in a car or on a city street. Paper maps could still have roles in wilderness areas where cellular service is weak, or for mapping very specific features in a small area. As a result, the paper map is not dead but is simply moving on to a semi-retirement, where it can be used for pursuits where its shortcomings are camouflaged. There also is and will continue to be a market for older, collectible maps for museums or private collections. The new digital age will allow these maps to be observed not only for their individual significance but also for the foundationally different time that they represented. These maps can be taken in and cherished as what they are: art.

What primary characteristics distinguishes a spatial join from an attribute join?

While attribute joins use a text-based or number-based common field to join two classes into one, spatial joins combine the two information sets based on spatial characteristics. Also, while attribute tables add the new information to an existing table, spatial joins create an entirely new feature class.

List and describe the five Environment settings that can be helpful in managing raster analysis. When should these be checked and set?

Workspace: This specifies where ArcGIS will place output rasters from the geospatial analysis. These should be checked and set at any time, as specifying a temporary storage location for intermediate rasters will make identifying and potentially deleting unnecessary temporary files far easier in the future. Output Coordinates: This specifies the coordinate system of the output. This should only be undertaken very intentionally and is thus likely best left as the default. Processing Extant: This specifies the area that will be processed by a tool. By default, it will process everything, but if one sets it, only that area will be processed. For our purposes, we will likely use the intersection of inputs option. Cell Size: This sets the raster output resolution. This should be checked if there are worries about processing constraints, or if the default suggestions are creating a raster with too high or low resolution. Mask: Using a mask specifies which cells will be processed, but it creates a unique raster that has nodata values outside of the raster. If one wants to limit the extent of their processing to a subset of the existing raster, this will be useful.

You are adding three rasters with cell resolutions of 30 meters, 50 meters, and 90 meters. List the four available options that would determine the cell size of the output raster and state what the output resolution would be in this case.

Your Answer: Maximum of Inputs: This will create a 90-meter resolution raster Minimum of Inputs: This will create a raster with 30-meter resolution As specified Below: This allows the user to specify a specific cell size and resolution Same as layer geogrid: This allows the user to set the raster resolution to the same resolution as an existing layer. If that existing layer has a 50-meter resolution, then the output raster would also have a 50-meter resolution

Which type of snapping (end, vertex, or edge) would work best in the following situations? digitizing streams digitizing parcels digitizing streets digitizing traffic lights at intersections digitizing stream gauges on streams

a) End snapping to connect the streams. (b) Vertex or edge snapping. (c) End snapping or edge snapping. (d) End snapping. (e) Edge snapping.