ASSESSING REMOTE SENSING FOR MAPPING AND MONITORING OF THE NATURAL AND MAN-MADE ENVIRONMENT

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: MCINTIRE-STENNIS
• Reporting Frequency: Annual
• Accession No.: 1002755
• Project Start Date: May 19, 2014
• Project End Date: Feb 28, 2019
• Program Code: [(N/A)]- (N/A)

Recipient Organization
UNIVERSITY OF FLORIDA
G022 MCCARTY HALL
GAINESVILLE,FL 32611

Performing Department
Forest Resources and Conservation

Non Technical Summary
The project will measure the efficacy of using aerial and space-based remote sensing devices and techniques for assessing, monitoring and mapping of the natural and man-made environment. It will be done throughthree sub-projects which test a specific remote sensing instrument and technique for a spectific application.The three projecst are:1. Using a Thermal Infrared Radiometer forDetection of Burmese pythons2. Use of an Unmanned Autonmous Vehicle for Detection and Mapping of Willow Trees3. Correlating Remote Sensing Data with Soil Nutrients

Animal Health Component

100%

Research Effort Categories

Basic

(N/A)

Applied

100%

Developmental

(N/A)

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
404	7410	2060	100%

Knowledge Area
404 - Instrumentation and Control Systems;

Subject Of Investigation
7410 - General technology;

Field Of Science
2060 - Geography;

Keywords

remote sensing

satellite imagery

aerial photography

digital imagery

unmanned autonomous vehicle

Goals / Objectives
The goal of this projectis to examine and test the range of applications of airborne and satellite remote sensing for assessing, mapping and monitoring the natural and man-made environment. The remote sensing techniques and methods utiliized will include digital photography, thermal infrared remote senisng,and multispectral,scanning.

Project Methods
The project will consist ofthree sub-projects:(1). Use of Thermal Infrared Radiometers for Detetion of Burmese PythonsThe location of the testing will be the National Wildlife Research Center (NWRC) in Gainesville Florida. The NWRC houses a number of male and female Python bivittatus of various sizes. The NWRC grounds are suitable for conducting tests of range up to 100m, which is the normal flying altitude of a UAV.We will conduct tests to quantify three variables: range, time and vegetation cover. Starting at dusk on Day 1, we will image an uncovered Python and one covered with vegetation at a range of 5m. We will repeat the experiment every 10 minutes until the snake is no longer detectable (i.e., it is the same kinetic temperature as the background). We will repeat these experiments at 10m intervals for 10 days. Details on the equipement and exact procedures are given in the Project Description.(2). Detection and Mapping of Willow Tress in the St. Johns River Water Management District.Using a UAVWewill fly a UAV over areas inthe St. Johns River Water Management Districtwhere willow tree are known to exist. We will determine under which environmental conditions willow can best be detected and whether or not they can be mapped usingcomputer visionimage processing techniques.This will include object recogntion, edge-detection, shape detection and other techniques.The current sensor suite of the UAV is adigital single lens reflex camera that operates in the visual and near-infrared part of the electromagntic spctrum. We plan to add a multispectral scanner to test whether willow can be indentified more accurately than is possible with the current sensor suite.(3). Correlating Remote Sensing Data with Soil NutrientsData collection: Ground-truth data will be collected at soil observation sites of phenological cycles. We will perform an inventory of cloud-free imagery from the following IRS remote sensing satellites: RESOURCESAT-1, CARTOSAT-1 and CARTOSAT-2. (acquire images of both 2 x 10 km2 study sites for three times a year). The GLCF imagery library will be screened for freely available scenes - either IKONOS, Quickbird or GeoEye.Analysis to derive crop/vegetation specific properties: Radiometric or other corrections & supervised classification (land cover / crop type) will be performed using standard methods. Various vegetation indices (VI) will be derived from the images (satellite images and the Leica ADS40): (1) the normalized difference (NDVI), (2) perpendicular vegetation index (PVI), (3) soil adjusted vegetation index (SAVI), (4) transformed soil adjusted vegetation index (TSAVI), and (5) the Tasseled Cap. Other crop/vegetation-specific properties that will be derived include the biomass (green & dead), vegetation stress, chlorophyll concentration, fraction of absorbed photosynthetically active radiation (FAPAR), leaf area index (LAI), vegetation phenological metrics (VPM), and nutrient content in crops (leaves). Spectral data derived from images will be related to those collected with a spectrophotometer on the ground and spectral libraries (e.g. USGS and NASA) to interpret spectral signatures (crop identification and mapping of vegetation properties).An objective of this project isto fuse spatially-explicit soil and remote-sensing derived spectral data into a geospatial multi-spectral engine (GeoSCIE). Spectral data will be ground-truthed using a set of geo-referenced field observations to calibrate the engine which infers on a set of critical crop and soil signatures.Diffuse reflectance spectroscopy and remote sensing data assembled under (1) and (2) as well as soil and field observations with their respective x and y coordinates will be fused into a relational database. Although all data collected in the project are spatially-explicit they are derived at different support (grain, pixel size, or sample volume). Soil observations collected in the field and DRS data are considered "point" data, with specific x, y coordinates, due to their small sample support. In contrast, remote sensing data come in a particular IVOFs ("footprint") discretized into a fixed grain or pixel-size for processing dependent on the sensor resolution. Site-specific soil data are only available sparsely (at 500 locations across the two study sites), whereas satellite and aerial imagery and derived crop-specific products provide dense, continuous grids across the study areas. GeoSCIE will integrate point and pixel-specific soil and crop data, DRS and remotely-sensed spectral data and indices, and ancillary environmental spatial data (e.g. DEM, geology, soil maps, etc.) into a holistic engine that produces integrated soil-crop indices as output. The processing steps are the following:We will contrast model(s) derived using soil and crop/vegetation-specific properties with model(s) derived from spectral data in study area 1. To test this hypothesis the total observation set will be split (70% - model calibration; 30% - validation) and error statistics (R2, RMSE, and mean prediction error) used to compare models in study area 1. . We will repeat this analysis for different seasons using temporal sets of remote sensing data (selection of satellite images will be dependent on availability of cloud/noise free imagery. The transferability potential will be tested by employing the same model developed in study area 2. Independent validation will be performed to assess the transferability potential

Progress 10/01/18 to 02/28/19

Outputs
Target Audience: Nothing Reported Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?A paper was published in the journal Computers and Electronics in Agriculture in January 2019. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? I intiated a study to indentify and map saw plametto vegetation in northern Florida using very resolution satellite imagery and drone imagery. The objective is to differentiate high, medium and low density of saw palmetto and acertain whether the maturity of the plants can be determined with the drone imagery. I continued a study of using hyperpsectral drone-based imagery for determining the maturity of peanut plants.

Publications

• Type: Journal Articles Status: Published Year Published: 2019 Citation: Monsef, H., S. Smith, D. Rowland and N. Rasol, Use of Hyperspectral Scanner Imagery for Determining Peanut Maturity. Journal of Computers and Electronics in Agriculture, January 2019.

Progress 05/19/14 to 02/28/19

Outputs
Target Audience: Nothing Reported Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?I gave a number of lectures and published a number of journal articles on the topic. How have the results been disseminated to communities of interest?Peer-reviewd journal articles, lectures to university-level students, public school students, interest groups, professional society meetings and other forums. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? I explored a wide range of applications of remote sensing to assess, map and monitor the natural and man-made environment.

Publications

• Type: Journal Articles Status: Published Year Published: 2019 Citation: Monsef, H., S. Smith, D. Rowland and N. Rasol, Use of Hyspectral Imagery for Determining Peanut Maturity, Journal of Computers and Electronics in Agriculture, January 2019.

Progress 10/01/17 to 09/30/18

Outputs
Target Audience:Three target audiences were reached during this reporteing period: (1) academic researchers and teachers (2) the private sector and (3) government agencies. 1. Academic researchers and teachers: 2. Private Sector 3. Government Agencies Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?Publications in peer-reviewed publications. What do you plan to do during the next reporting period to accomplish the goals?We will continue to collect and assess data on peanut matruity using a UAS-based hyperspectral scanner. We will continue to fly the LIDAR, hyperspectral, multi-spectral and RGB camera USA for natureal landscapes. We will continue our research in saw palmetto.

Impacts
What was accomplished under these goals? We continued to collect and perform analysis ofdata for the peanut maturity project. One paper was published in 2018 in the journal Computersand Electronicsin Agriculture on this topic. We flew a large numb er of missions using our rotary blade UAS. The emphasis in 2018 was on calibrating a LIDAR sensor on the UAS and merging it with optical imagery. Flightswere primarily over timberland and rangeland in north central Florida. Priminary research was conducted on the feasibility of using very high resolution satellite imagery to identify and map high density patches of saw palmetto vegetation.

Publications

• Type: Journal Articles Status: Accepted Year Published: 2018 Citation: Smith, S., H. Monsef, Roland, D. and N. El Rasol, "Development of Vegetation Indicies for a Peanut Canopy Prediction Model". Computers and Electronics in Agriculture.

Progress 10/01/16 to 09/30/17

Outputs
Target Audience:The target audience were end users of Unmanned Autonomous Systems. An exapme was a presentation made in March 2017 to a group of scientists interested in using UAS's thermal infrared sensor for detecting Bumess Pythons in South Florida. We als gave a number of talks and poster sessions on the capabilites of the UAS. In the summer 2017, we were part of a major prescribed wildfire experiment with the EPA, USGS and NOAA at Tall Timbers. Changes/Problems:There are no changes to the intial mission except that we are finding more applications for our rotory UAS syste3m vis-a-vis the fixed-wing aircraft. Therefore, we are spending more time with it. What opportunities for training and professional development has the project provided?We gave demonstratiopns of the UAS to each of the course I teach as well as a new field course on UAS's which is taught during the Summer semester. During the previously mentioned prescribed wildlfire mission at Tall Timbers, we were able to train personal from the EPA, USGS and NOAA on the UAS. We also gave demostrations to UF and USDA persponal at the UF Agricultural Research Station in Citra Florida. How have the results been disseminated to communities of interest?Mainly through presentations at a number of meetings and workshops. We also published several relevant papers. What do you plan to do during the next reporting period to accomplish the goals?We will continue to fly the UAS on a number of tasked-based missions for the USGS and other Federal agencies. The peanut study will continue with a focus of early detection of disease which was an issue on 2017. We will focus more on simpliflying the deployment of UAS's so that they can be used by more agencies such as water management districts.

Impacts
What was accomplished under these goals? We made several presentations at local (Florida) meetings as well as National meetings on the use of UAS's for environmental monitoring. My focus this year was on Burmes Pythons, wild fire monitoring and assessing the maturity and drought stress to peanuts using a hyperspectral scanner mountedon the UAS. We aquired new instruments for the UAS, including a LiDAR and hyperspectral scanner. They needed to be intergrated to the other sensors and navigation system.

Publications

• Type: Journal Articles Status: Published Year Published: 2017 Citation: " Burgess M., R. Carthy, B. Wilkinson, T. White, S. Tyler, P. Ifju, S. Smith, 2017. All Things Unmanned (AUVSI), Multiple Entities Team Up for UAS/Sensor Evaluation during Fire Science Consortium: Measuring fine scale fire behavior and fire effects, 2017, http://www.auvsi.org/blogs/auvsi-news/2017/04/26/multiple-entities-team-up-for-uassensor-evaluation-during-fire-science-consortium-measuring-fine-scale-fire-behavior-and-fire-effects-event.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: " Xu, Y., S. Smith, S. Grunwald, A. Abd-Elrahman, S. Wani, 2017. Incorporation of Satellite Remote Sensing Pan-sharpened Imagery into Soil Prediction Models to Characterize Soil Property Variability in Small Agricultural Fields. ISPRS Journal of Photogrammetry and Remote Sensing. Vol. 123, pg 119.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: " Monsef H. and S. Smith, 2017, A New Approach for Estimating Mangrove Canopy Cover Using Landsat 8 Satellite Imagery. Computers and Electronics in Agriculture. 135. 183-194.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: " Szantoi, Z. L. Koh, S. Smith and S. Wich, 2017, Mapping of Orangutan Habitat Using Landsat Imagery Augmented with Unmanned Autonomous Vehicle Photography. International Journal of Remote Sensing. http://dx.doi.org/10.1080/01431161.2017.1280638.
• Type: Journal Articles Status: Awaiting Publication Year Published: 2017 Citation: " Yiming Xu, Scot E. Smith, Sabine Grunwald, Amr Abd-Elrahman and S. Wani, 2017, "Spatial Downscaling of Soil Prediction Models based on Weighted Generalized Additive Models in Smallholder Farm Settings", Environmental Monitoring and Assessment. Accepted. In Press.
• Type: Journal Articles Status: Accepted Year Published: 2017 Citation: " Xu, Y., S. Smith, S. Grunwald, A. Abd-Elrahman, S. Wani, 2017, Evaluating the Effect of Remote Sensing Image Spatial Resolution on Soil Exchangeable Potassium Prediction Models in Smallholder Farm Settings. Journal of Environmental Management, In review.
• Type: Journal Articles Status: Under Review Year Published: 2017 Citation: " Xu, Y., S. Smith, S. Grunwald, A. Abd-Elrahman, S. Wani, 2017, Effects of Image pan sharpening on soil total nitrogen prediction models in South India. Computers and Electronics in Agriculture. In Review.

Progress 10/01/15 to 09/30/16

Outputs
Target Audience:The target audience for this research are (1) wildlife biologists concerned with managing the introction of Burmese Pythons in the Florida Everglades and other wetlands in Florida (2) persons interestd in the eradication of invasive exoitc vegetation in Florida and how UAVs can be employed (3) Farmers of small plot operations. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?Publications in peer-reviewed journals Reports College-level courses What do you plan to do during the next reporting period to accomplish the goals?MOre experiments on using thrmal infrared radiometers for pythions, detection and mapping of other invasive exoitc vegetation using a RGB camera on a UAV and more experiments to explore the relation between soil nutrients and spectral reflectance.

Impacts
What was accomplished under these goals? Several experiements were conducted to test whether of not thermal infrared radiometers could be used to detect phythons after duisk. A final report was written. An experiemnt was conducted to test the efficay of using a RGB camera on a UAV to identiy Carolina willows in the St Johns River Water Management District. A study was performed on two sites in southern Inida to test whether soil nutrient concentrationcould be corelated with spectral refelctance by satellite imagery.

Publications

• Type: Journal Articles Status: Accepted Year Published: 2016 Citation: " Yiming Xu, Scot E. Smith, Sabine Grunwald, Amr Abd-Elrahman, Suhas P. Wani Incorporation of Satellite Remote Sensing Pan-sharpened Imagery into Soil Prediction Models to Characterize Soil Property Variability in Small Agricultural Fields. ISPRS Journal of Photogrammetry and Remote Sensing. Accepted 9/2016.

Progress 10/01/14 to 09/30/15

Outputs
Target Audience: Nothing Reported Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?Numerous seminars were conducted to describe the UAV and it's capabilites. What do you plan to do during the next reporting period to accomplish the goals?Two manuscripts are being prepared for submission to peer-reviewd journals.

Impacts
What was accomplished under these goals? Several flights of a fixed-wingunmammed autonomous vechicle (UAV) were made with weredesigned to accomplish the goals of the project. Construction of a rotary UAV was completed and the UAV was tested.

Publications

Progress 05/19/14 to 09/30/14

Outputs
Target Audience: Staff at the St Johns River Water Management District Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Training on UAS for staff of the St Johns River Water Management District How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals? Continue flying the UAS

Impacts
What was accomplished under these goals? UAS was flown, imagery was acquired and interpreted on 5 dates of the Upper Ocklawaha River Basin.

Publications