Evaluating Remote Sensing Data Or How to Avoid Making Great Discoveries by Misinterpreting Data Richard Kleidman ARSET-AQ Applied Remote Sensing Education and Training A project of NASA Applied Sciences

It may take a few seconds to really get the idea behind this comic strip. Interpreting data can be a tricky business.

AOD at 550 nm Area Average Plot Giovanni MODIS Daily Instance June 15 - 30, 2008 What are the two biggest differences between the results for these two sensors and time periods? Terra Daily Overpass ~ 10:30 AM local time Aqua Daily Overpass ~ 1:30 PM local time

Real or Not Real ? You Decide! A Potential Discovery! AOD at 550 nm Area Average Plot Giovanni MODIS Daily Instance June 15 - 30, 2008 A Potential Discovery! Real or Not Real ? You Decide! What are the two biggest differences between the results for these two sensors and time periods? Terra Daily Overpass ~ 10:30 AM local time Aqua Daily Overpass ~ 1:30 PM local time

AOD at 550 nm Area Average Plot Giovanni MODIS Daily Instance June 15 - 30, 2008 Hypothesis 1: The differences in aerosol concentration represent a Diurnal Cycle with different amounts of aerosol being produced or transported afternoon and morning. Hypothesis 2: The differences in aerosol concentration are either not real or not significant and could be caused by Sensor error Algorithm error Sampling error ? What are the two biggest differences between the results for these two sensors and time periods? Terra Daily Overpass ~ 10:30 AM local time Aqua Daily Overpass ~ 1:30 PM local time

Possible ways to interpret these differences Real world differences Leads to a conclusion(s) about aerosols Can lead to further research about aerosols Differences due to other factors Can lead to false conclusion about aerosols Need to be explored and understood to avoid similar problems in the future - Can lead to advances in remote sensing capabilities! These are two very broad categories. Note from the last point that even discoveries that don’t reflect real world phenomena can be useful and should not be ignored.

Remote Sensing Process Energy Source or Illumination (A) Recording of Energy by the Sensor (D)  Transmission, Reception, and Processing (E)  Interpretation and Analysis (F) Radiation and the Atmosphere (B)  Energy Source or Illumination (A) the first requirement for remote sensing is to have an energy source which illuminates or provides electromagnetic energy to the target of interest. Radiation and the Atmosphere (B) - as the energy travels from its source to the target, it will come in contact with and interact with the atmosphere it passes through. This interaction may take place a second time as the energy travels from the target to the sensor.  Interaction with the Target (C) - once the energy makes its way to the target through the atmosphere, it interacts with the target depending on the properties of both the target and the radiation Application (G) - the final element of the remote sensing process is achieved when we apply the information we have been able to extract from the imagery about the target in order to better understand it, reveal some new information, or assist in solving a particular problem Recording of Energy by the Sensor (D) - after the energy has been scattered by, or emitted from the target, we require a sensor (remote - not in contact with the target) to collect and record the electromagnetic radiation. Transmission, Reception, and Processing (E) - the energy recorded by the sensor has to be transmitted, often in electronic form, to a receiving and processing station where the data are processed into an image (hardcopy and/or digital) Interpretation and Analysis (F) - the processed image is interpreted, visually and/or digitally or electronically, to extract information about the target which was illuminated. (G) Application  Interaction with the Target (C)  Reference: CCRS/CCT

Important Factors to Understand Analysis Tool Giovanni – provides data in 1 degree resolution Data Products– Aqua and Terra Level 2 Products are from a single overpass 10 KM resolution Aqua and Terra Level 3 Products are global composites in 1 Degree resolution Black circles - The same region may have very different average values of AOD for this time period. Maroon squares – There are differences in where there are no values returned for this time period.

Important Factors to Understand Mid – latitude 1° x 1° is about 85 Km x 110 Km ~ 90 10 Km MODIS retrievals possible ~ 45 10 Km MODIS retrievals possible Swath Nadir Edge Black circles - The same region may have very different average values of AOD for this time period. Maroon squares – There are differences in where there are no values returned for this time period.

Important Factors to Understand One MODIS 10 Km Retrieval Begins life as 400 .5 km (at nadir) pixels And ends as a product composed of 12 – 120 pixels Black circles - The same region may have very different average values of AOD for this time period. Maroon squares – There are differences in where there are no values returned for this time period.

Information Necessary to Understand the Results Data Products– Aqua and Terra Level 2 Products are in 10 KM resolution Aqua and Terra Level 3 Products are in 1 Degree resolution Giovanni provides Level 3 Data in 1 Degree resolution. Sensor Characteristics – Aqua and Terra are identical designs There are some small differences in sensor performance Black circles - The same region may have very different average values of AOD for this time period. Maroon squares – There are differences in where there are no values returned for this time period. Algorithm Details– Aqua and Terra use the same algorithm.

AOD at 550 nm Area Average Plot June 15 - 22, 2008 Examining other features in this data set and using our knowledge of the sensor and products can help us to understand the cause of the differences in mean aerosol. A blank (white) square has no retrievals for the entire time period. Aqua Terra

Evaluating Data Understand the sensor characteristics Understand the product details Understand the data visualization tools and outputs.