FOR375 EXAM #2 STUDY SESSION SPRING Lecture 14 Exam #2 Study Session

Transcription

1 FOR375 EXAM #2 STUDY SESSION SPRING 2016 Lecture 14 Exam #2 Study Session

2 INTRODUCTION TO REMOTE SENSING

3 TYPES OF REMOTE SENSING Ground based platforms Airborne based platforms Space based platforms

4 TYPES OF RESOLUTION Spatial resolution spatial detail & geographic extent Spectral resolution colors or spectral differentiation (wavelength differentiation) Temporal resolution when, or how often (monitoring) Radiometric resolution how many gray levels, or levels of brightness differentiation

5 TYPES OF RESOLUTION Spatial resolution what size can I observe? Spectral resolution what wavelengths can I observe? Temporal resolution how often do you observe? Radiometric resolution degree of detail observed?

6 REMOTE SENSORS Image Extent the area covered by an image, and depends on the physical size of the sensing area (sd), the camera focal length (h) and the flying height (H). GD/Image Extent = sd * H /h The ground resolution on aerial images can be determined by substituting the cell dimension for sensor dimension Ground Resolution = cd * H /h

7 As a general rule of thumb, in order to detect and/or map spatial objects on the ground, the pixel size should be about 1/3 the size of the object Spatial resolution: refers to the pixel size, cell size, or instantaneous field of view (IFOV) Typically the lower the flying height the greater the detail (larger scale), smaller the individual pixel size Trade-offs exist between high spatial resolution and overall ground area encompassed by an image High spatial resolution Intermediate spatial resolution Low spatial resolution

8 ELECTROMAGNETIC SPECTRUM (EM) AND WAVELENGTHS Light can be divided up into bands that have differing wavelengths. Objects of differing composition (trees, grass, rocks) reflect the varied wavelengths of light in differing ways. Nanometer meter, or one billionth of a meter.

9 SPECTRAL RESOLUTION Spectral resolution: refers to the number of colors or spectral bands we can acquire with a particular sensor or camera Traditional cameras capture visible wavelengths of light (B,G,R), and many newer sensors can capture or image wavelength bands (channels) outside of the visible spectrum (NIR, MIR, TIR). A band refers to a range along the electromagnetic spectrum. (Red band = ~.6-.7 µmeters)

10 Normalized Difference Vegetation Index: NDVI = NIR - visible NIR + visible The normalized difference vegetation index (NDVI) is a simple graphical indicator that can be used to analyze remote sensing measurements, typically but not necessarily from a space platform, and assess whether the target being observed contains live green vegetation or not.

11 NORMALIZED DIFFERENCE VEGETATION INDEX (NDVI) The pigment in plant leaves, chlorophyll, strongly absorbs visible light (from 0.4 to 0.7 µm) for use in photosynthesis. The cell structure of the leaves, on the other hand, strongly reflects nearinfrared light (from 0.7 to 1.1 µm). The more leaves a plant has, the more these wavelengths of light are absorbed, respectively. By exploiting the strong differences in plant reflectance, we can determine their spatial distribution NIR - visible NIR + visible NDVI can be used to quantify the photosynthetic capacity of plant canopies.

12 SPATIAL ANALYSIS TECHNIQUES

13 SPATIAL AUTOCORRELATION A measure of the degree to which a set of spatial features and their associated data values tend to be clustered together in space (positive spatial autocorrelation) or dispersed (negative spatial autocorrelation).

14 SPATIAL ANALYSIS TECHNIQUES Buffer: create a new area around an object. Clip: create new files with the shape of one layer and the attributes of another. Intersect: create a file with the shape of the shared areas and attributes of both layers. Dissolve: remove the boundaries between adjacent areas that have the same values for attributes. Spatial Join: add attributes of one file to another based on location. Union: Joins two layers together visually with the new attribute table consisting of shared/overlapping areas.

15 COMBINING LAYERS CAN LEAD TO POSSIBLE PROBLEMS Parcels Overlay Impervious/Pervious Yikes! Zillions of polygons! In this instance, output layer combines spatial polygons and attribute tables Illustrative material courtesy of Leslie Morrissey

16 COMBINING LAYERS CAN LEAD TO POSSIBLE PROBLEMS Can result in polygon slivers (small polygon errors created by misalignment of overlay data) Output file type (shapefile vs. feature class) Shapefile output: no recalculation of area, perimeter, or length fields Output as GDB feature class for accurate area, perimeter, and length calculations! Input layers must have matching projection/datum (spatial reference) No automatic recalculation of numeric attributes for polygons that depend on a spatial unit! Slide content courtesy of Leslie Morrissey

17 COMBINING GEOPROCESSING TOOLS Involve multiple tasks performed in sequence, such as those that clip, buffer, intersect, union, then select datasets. Step by step Build and run a model Create and run a script

18 SPATIAL ANALYSIS TECHNIQUES SUMMARY Union Intersect Identity Clip Dissolve Merge Buffer

19 SPATIAL ANALYSIS TECHNIQUES WHAT YOU SHOULD KNOW Spatial autocorrelation what is it Spatial analysis techniques buffer, intersect, union, dissolve, merge, clip

20 GIS TABLE OPERATIONS

21 BASICS OF SPATIAL ANALYSIS: SPATIAL TABLE OPERATIONS Querying geospatial data Aspatial, or Attribute, or Tabular - querying data based on tabular relationships Example: select all the counties that have a population per square mile less than 14. Spatial querying data based on a geospatial relationships Example: select all the counties that are adjacent to those counties that have a population per square mile less than 14.

22 Data Data is numerical, character or other symbols which can be recorded in a form suitable for processing by a computer. (e.g. names and addresses of students enrolling onto a university course). Schema The schema is the structure of data, whereas the data are the facts. Schema basically indicates the rules which the data must obey. Such rules can be enforced by a database. and the more rules there is the harder it is to enter poor quality data. Database A Database is a collection of related data (such as an enrolling students data) arranged for speedy search and retrieval. Database Management System A Database Management System is a collection of programs that allows users to specify the structure of a database, to create, query and modify the data in the database and to control access to it. (e.g. limit access to the database so that only relevant staff can access details of enrolling students).

23 TABLE OPERATIONS Identify Near Spatial Queries Join Spatial Joins

24 TABLE JOINS SIMPLE YET POWERFUL!

25 PRIMARY VS FOREIGN KEY A primary key uniquely identifies a record in the table. A foreign key is a field in the table that is primary key in another table. A primary key can't accept null values. A foreign key can accept multiple null values. By default, a primary key is a clustered index, and data in the database table is physically organized in the sequence of this index.a foreign key does not automatically create an index, clustered or non-clustered. You can manually create an index on a foreign key. We can have only one primary key in a table. We can have more than one foreign key in a table.

26 INTRODUCTION TO RASTER ANALYSIS

27 TYPES OF RASTER SPATIAL ANALYSIS Statistics Cell statistics Neighborhood statistics Zonal statistics Reclassification Raster Calculator (map algebra) Conversion of vector and raster

28 MAPPING DISTANCE USING RASTERS The Straight Line Distance function measures the straight line distance from each cell to the closest source The Allocation function allows you to identify which cells belong to which source based on straight line distance The Cost Weighted Distance function modifies the Straight Line Distance by some other factor, which is a cost to travel through any given cell. For example, it may be shorter to climb over the mountain to the destination, but it is faster to walk around it. The Distance and Direction raster datasets are normally created from Cost Weighted Distance function to serve as inputs to the pathfinding function, the shortest (or leastcost) path.

29 MAPPING DISTANCE USING RASTERS The Straight Line Distance function measures the straight line distance from each cell to the closest source The Allocation function allows you to identify which cells belong to which source based on straight line distance The Cost Weighted Distance function modifies the Straight Line Distance by some other factor, which is a cost to travel through any given cell. For example, it may be shorter to climb over the mountain to the destination, but it is faster to walk around it. The Distance and Direction raster datasets are normally created from Cost Weighted Distance function to serve as inputs to the pathfinding function, the shortest (or leastcost) path.

30

31 ALLOCATION FUNCTION Allows you to identify which cells belong to which source based on straight line distance function or cost weighted distance function. Straight line distance Why use the Allocation function? Use the allocation function to perform analysis such as: Identifying crop yield observations supported by local grain harvest locations Find out which water source is closest Finding areas within a forest inventory that have a shortage of a particular species of tree Cost weighted distance 1,2 are shopping centers

32 COST WEIGHTED DISTANCE Functions that perform cost weighted distance mapping are similar to the Straight Line Distance functions, but instead of calculating the actual distance from one point to another, they compute the accumulative cost of traveling from each cell to the nearest source, based on the cell s distance from each source and the cost to travel through it. For example, its easier to walk thru a meadow than a swamp.

33 CELL STATISTICS (LOCAL FUNCTION) a statistic for each cell in an output raster is based on the values of each cell of multiple input rasters. for instance, to analyze the average crop yield over a 10-year period Majority, maximum, mean, median, minimum, minority, range, standard deviation, sum, variety

34 MAP ALGEBRA If any of the input is NODATA, the output is NODATA

35 NEIGHBORHOOD STATISTICS (FOCAL) A statistic for each cell in an output raster is based on the values of cells within a specified neighborhood: rectangle, circle, annulus, and wedge Majority, maximum, mean, median, minimum, minority, range, standard deviation, sum, variety Focal Minority the least frequent value in a neighborhood Focal Majority the most frequent value in a neighborhood Focal Minimum the minimum value in a neighborhood Focal Maximum the maximum value in a neighborhood Focal Sum the total of all values in a neighborhood Focal Mean the average of all values in a neighborhood Focal Variety the number of different values in a neighborhood Sum of 3 x 3 cell neighborhood Range = max-min

36 ZONAL STATISTICS Computer statistics for each zone of a zone dataset based on the information in a value raster. zone dataset can be feature or raster, the value raster must be a raster.

37 6. RECLASSIFICATION

38 RASTER ANALYSIS WHAT YOU NEED TO KNOW Straight line function, allocation function, cost weighted function Neighborhood and cell statistics Map algebra Zonal statistics Hillshade, aspect, slope, density

39 STEROSCOPY

40 STEREOSCOPY Stereoscopy the science of perceiving depth using two eyes / vantage points parrallactic angles basis for depth perception as distance to objects increase, parrallactic angle gets smaller and we lose depth perception/stereo ability (~ 1000m) Stereoscopy Bolstad p

41 SPATIAL MODELING I & 2

42 SPATIAL MODELING 2 Brief summary of discussion from Spatial Modeling 1 Example focus: Erosion Modeling Example focus: Livestock Distribution Modeling

43 ACCESSING CLIMATE DATA Download/ftp Using RESTful web services Using a web browser Access web service data directly in ArcGIS (more on this in a later lecture and lab)

44 CARTOGRAPHIC MODELING Cartographic modeling is common Landuse planning,transportation route and corridor studies, modeling disease Temporally static. Useful for suitability analysis

45 SIMPLE SPATIAL MODELING Typically generates a statistical model for predictive efforts (ie. regression or classification techniques kmeans, nearest neighbor, linear and quadratic discriminant analysis, hiearchical clustering, etc.) Useful when there is a well-established model that is based on discrete values Examples: NDVI calculation and predictive assessment, Revised USLE calculation and prediction.

46 SPATIAL-TEMPORAL MODELING Used when you have data varying by time and space Uses spatially explicit inputs to calculate or predict spatially-explicit outputs. An advanced form of spatial modeling

47 SPATIAL-TEMPORAL MODELING Other types of spatial-temporal modeling techniques: cell based modeling agent based modeling

48 HOW SPATIAL MODELING WORKS USING ARCGIS MODEL BUILDER Drag layers you want to participate into the model Drag tools you want to use into the model Output layers, tables, objects shown in green Connect the features using arrows Order matters to certain tools (Clip)

49 SPATIAL MODELING I - WHAT YOU SHOULD KNOW Define cartographic simple spatial, and spatial-temporal models; Know what is cell based modeling; Understand how each of the model types generates data thru spatial operations (cartographic), statistical functions (simple spatial), or predictive analyses (spatial-temporal).

50 SPATIAL MODELING, PART II WHAT YOU SHOULD KNOW Understand how natural resource systems can be modeled geospatially (hydrology, erosion, forestry) Begin to formulate ideas on how you might apply modeling techniques to alternative natural resource systems. Fisheries Habitat populations Could you draw a basic conceptual model of a natural resource model, such as erosion or hydrology?

51 OPEN SOURCE GIS

52 OPEN SOURCE GIS Open source history Active open source vector and raster software you can use on your own Why would you use OS software? Discussion on python and R - two powerful open source tools for geospatial analysis (really any type of data analysis)

53 OPEN SOURCE GIS TIMELINE

54

55 WHO USES OPEN SOURCE GIS? ArcGIS uses GDAL python library United Nations extensive open source GIS use UCSB Marine Map Many case studies here: More case studies: Big projects like Linux, Apache, Mozilla Firefox and OpenOffice are supported by Fortune 500 companies like IBM and Sun. OSGeo is supported by Autodesk.

56 OPEN SOURCE GIS WHAT YOU SHOULD KNOW Be able to list and describe the functionality of three or four open source software packages Be able to narratively describe the advantages and disadvantages of OS software Have a very basic understanding of python and R what they are