Lab 8: Spectral Signature Analysis & Resource Monitoring

Introduction:

This lab is set to help familiarize students with spectral reflectance signatures of various objects found on the Earth's surface that can be made visible in aerial photographs. Student will first collect these signatures from remotely sensed images before graphing and analyzing their outputs in order to confirm their validity using the spectral separability test. In analyzing the combination of these remotely sensed images and object signatures, students will ultimately determine and further monitor the health conditions of various vegetation and soils.

Methods:

Part One: Spectral Signature Analysis-

For Part One of the lab, students used the Polygon and Raster Supervised Signature Editor tools synced with Google Earth in Erdas to collect spectral signatures of various feature objects. From a satellite image of Eau Claire county, students collected a total of 12 signatures, including those from standing water, still water, deciduous forests, coniferous forests, riparian environments, crops, wet soil, dry soil, rock, asphalt, an airport runway, and concrete.

After collecting each of the signatures, students were able to make an analysis comparing the signatures similarities and differences in terms of their reflectance levels portrayed between each of the 6 bands visible in the Spectral Signature Plot Box. Students drew upon their knowledge from lecture material and their observations from the Plot Box to make inferences as to why each of the object reflectance behavior was symbolized as it was.


Part Two: Resource Monitoring-

For part two, students were able to monitor the health of Eau Claire and Chippewa counties' vegetation and soils. Ultimately, students were able to assemble a series of two maps which explored the spatial distribution of iron contents in soils in one and abundant areas of green vegetation in the other.

Results:

Part One: Spectral Signature Analysis- 

The combination of the spectral signatures collected from each of the 12 features is featured below in Figure 1. The figure shows the most variance between signatures occurring in the last 4 bands sequenced, bands 3, 4, 5 and 6. These bands are good tools for noticing key differences in the identity of a feature detected from satellite imagery.

Figure 1: Combination of Spectral Signatures

Figure 2: Dry vs Wet Soils

Some key signature differences are worth observing between some of the observed features. For example, Figure 2 shows that there is a significant difference in the way in which dry and wet soil reflectance can be observed in band 5. This is due to the level of moisture found in the soil. Moisture and reflectance, in this sense, are have an inverse relationship. When there is more moisture in soil (as in the case of the dark brown, wet soil), there is then less reflectance.





Part 2: Monitoring Vegetation and Soil Health-

The first map featured in Figure 3 below is an example of Vegetation Health Monitoring for Chippewa and Eau Claire Counties. This map is classified using equal intervals classification method for distinguishing between features in the map. The map's lightest areas in white are representative of healthy, green vegetation. Black spots indicate areas of water, and the areas in dark grey symbolize areas of no vegetation at all. 

Figure 3: Map of Vegetation Health Monitoring

Figure 4, then, shows a map of soil monitoring, accounting for areas which are highly concentrated in iron. In this instance, the darkest areas account for those that are highly vegetated, or show minimally exposed soils. The lighter three shades, then, distinguish between areas which have been deemed concentrated with either low, moderate, or high ferrous minerals. Areas in white are indicative of soils with the highest levels of iron minerals. Though there are fewer spaces of these, a few can be seen concentrated to the east of Lake Altoona. 

Figure 4: Soil Health Monitoring Map

Conclusion:

This lab was extremely helpful in introducing students to the use of spectral signatures and the benefits of using satellite imagery for soil and vegetation health monitoring in various communities. These are extremely resourceful skill sets to capitalize on that can be easily implemented in future satellite imagery analysis taken on in real world scenarios. Overall, this lab has better prepared students for real world applications involving the monitoring of environmental health and individual features as they exist in the real world through the much more easily accessible use of satellite imagery rather than the same analysis being run through field survey techniques.

Sources: 

Satellite image is from Earth Resources Observation and Science Center, United States Geological Survey.