SlideShare une entreprise Scribd logo
1  sur  41
Estimation of Total Above-Ground Biomass Using
High Resolution UAS Imagery
David Kramar1, Valquiria Quirino1, Asami Minei1, Alison Wallace, and Breanna Huynh2
Minnesota State University Moorhead
1Department of Anthropology and Earth Sciences
2Department of Biological Sciences
Objectives
• Collect and Classify High-Resolution
Imagery from Two Sites using an
Off-The-Shelf UAS
• Compare multiple Non-NIR
Vegetation Indices to Determine
which Produces the Best Results
• Determine Which Index Best
Estimates Above Ground Biomass
Source: https://www.mnstate.edu/sciencecenter/
The Regional Science Center
• The Minnesota State University
Moorhead Regional Science Center
(RSC) is located 16 miles east of
Moorhead, adjacent to Buffalo River
State Park (BRSP)1.*
• There are 25 rare plants and animal
species within the complex BRSP,
RSC, and Bluestem Prairie2.**
Source: https://www.mnstate.edu/sciencecenter/
The Regional Science Center
• Currently several projects are underway in the Regional Science Center
o One is a prairie restoration project funded under the Minnesota Environment
and Natural Resources Trust Fund (ENRTF).
o A second project supporting the Nutrient Network (NutNet)* – which is a
coordinated research network that is quantifying the impacts of human activities
on ecological systems (see http://www.nutnet.org/home for additional
information about NutNet) - is also underway.
o There is a rich archeology history here.
• It provides programs for over 7000 Pre-K-12 students every year.
• It is an excellent “Living Laboratory”.
Thus, there is an INTEREST IN MONITORING VEGETATION CHANGES in
this area.
Nutrient Network Overview
Nutrient Network Overview
• At Nutrient Network ecologists and scientists are collecting data in a
standardized experimental design, across a broad range of site types.
• Designed to quantify community and ecosystem responses across a range of
vegetative ecosystems.
• Comprised of more than 40 grassland sites worldwide.
• Grassroots effort.
• We have two sites that we are currently monitoring.
Plot Design of the Northern Study Area
Software, UAS Platform, and Tools
• Software: (GIS/Remote Sensing)
o ArcGIS, QGIS, ORFEO TB
• Software: (Statistics)
o R (RATTLE Package)
o R (random forest, party, tree, etc)
o Statistica
• Software: (UAS/Processing)
o Pix4D Desktop
o Pix4D Capture
o Drone Harmony
• UAS Platform:
• Phantom 4 Advanced
• Example of Tools (GIS)
• Spatial Analyst Extension
• Segment Mean Shift
• Nibble
• Region
• Etc
Vegetation Indices (Minei 2019)
• We focused on two sites.
• Imagery was collected every 2
weeks from mid-May until the
end of September.
• We wanted to determine
which index did the best job
identifying healthy versus
unhealthy vegetation, and
which index would best
predict total above-ground
biomass.
• All vegetation indexes were
generated and interpreted by
student Asami Minei. The
sheer volume of data she has
analyzed is overwhelming.
• Modified VVI
• 𝑉𝑉𝐼 = 𝐿𝑜𝑔 10 1 −
𝑅−𝑅𝑜
𝑅+𝑅𝑜
+ 10 1 −
𝐺−𝐺𝑜
𝐺+𝐺𝑜
+ 10 1 −
𝐵−𝐵𝑜
𝐵+𝐵𝑜
+ 10
• Modified Global Environmental Monitoring Index
• 𝐺𝐸𝑀𝐼 = 𝑒𝑡𝑎 1 − 0.25 ∗ 𝑒𝑡𝑎 −
𝑝𝑅𝑒𝑑−0.125
1−𝑝𝑅𝑒𝑑
• where 𝑒𝑡𝑎 =
2∗ 𝑝𝐺𝑟𝑒𝑒𝑛2−𝑝𝑅𝑒𝑑2 +1.5 ∗𝑝𝐺𝑟𝑒𝑒𝑛+0.5 ∗𝑝𝑅𝑒𝑑
𝑝𝐺𝑟𝑒𝑒𝑛+𝑝𝑅𝑒𝑑+0.5
• Green Leaf Index
• GLI =
2 ∗ρ𝐺𝑟𝑒𝑒𝑛−𝑝𝑅𝑒𝑑 −ρ𝐵𝑙𝑢𝑒
2 ∗ρ𝐺𝑟𝑒𝑒𝑛+𝑝𝑅𝑒𝑑+ρ𝐵𝑙𝑢𝑒
• VariGreen
• 𝑉𝐴𝑅𝐼 =
𝑝𝐺𝑟𝑒𝑒𝑛−𝑝𝑅𝑒𝑑
𝑝𝐺𝑟𝑒𝑒𝑛+𝑝𝑅𝑒𝑑−𝑝𝐵𝑙𝑢𝑒
• Green-Red Vegetation Index
• 𝐺𝑅𝑉𝐼 =
ρ𝐺𝑟𝑒𝑒𝑛 −𝑝𝑅𝑒𝑑
ρ𝐺𝑟𝑒𝑒𝑛+ρ𝑅𝑒𝑑
• We also calculated a soil-adjusted index (modified version of SAVI) which
appeared to perform well. Due to time limitations, we do not present that here.
North Pond
2018
2019
Minei, A. 2019
Vegetation Indices
VVI – Greater Detail GLI – More Washed Out
Minei, A. 2019
Minei, A. 2019
HoustonMinei, A. 2019
Minei, A. 2019
Minei, A. 2019
Biomass Estimates
• Biomass was collected in 10 cm wide x 100 cm long strips from two
locations in each plot.
• This resulted in 60 sub-plots. During the first collection, the sub-plots
were combined (DOH!), which reduced the total sample size to 30.
Moving forward these will be kept separate.
• We extracted the min, max, mean, and standard deviation of the
different vegetation indices as they related to the biomass sub-plots.
• Models were developed using polynomial regression, random
forests/CART, and neural networks.
• Today I will present results from Decision Trees (CART), Random Forests,
and Neural Networks.
Biomass Estimates
Biomass Estimates
1. Collected
2. Sorted
3. Identified
4. Weighed
Biomass Estimates
• Initial Models were Developed in R using both command line and
Rattle.
• Methods were implemented into an R-Package that will be distributed
via the R CRAN in the future (later talk at conference).
• Models were implemented spatially in both ArcGIS and R (due to R-
package development).
• Results were cross-validated. This year we have training AND validation
biomass plots that allow us to assess the models in areas outside of the
Nutrient Network proper.
• While we present the estimates as “High, Medium, Low”, actual
estimates were in grams, and then converted to grams per meter
squared.
Biomass Estimates
Biomass-Decision Tree
Biomass-Decision Tree (VVI)
Generally predicted total biomass well based
off the modified VVI.
Biomass-Decision Tree (VVI)
Generally predicted grass biomass well based
off the modified VVI.
Biomass-Decision Tree (VVI)
Prediction of forbes biomass was not as robust
as grass and total biomass.
Biomass-Decision Tree (GLI)
GLI performed better than GEMI, but not as
well as the VVI.
Biomass-Decision Tree (GLI)
GLI performed better than GEMI, but not as
well as the VVI. Total Biomass
Biomass-Decision Tree (GLI)
Similar to forbes estimates for the VVI,
estimates from the GLI were also lower
Biomass-Decision Tree (VariGreen)
VariGreen also performed better than GEMI,
But not as well as the VVI and not as well as
the GLI. As such – No Map For You!
Come Back 1 Month!
Biomass-Decision Tree (GEMI)
GEMI did not perform well in terms of
biomass estimates. As a result, the map is a
HOT MESS and not presented!
Biomass-Random Forests
Biomass-Random Forests(VVI)
Random Forests provided a better solution to
Biomass estimates than a traditional
regression tree.
Biomass-Random Forests(VVI)
Random Forests provided a better solution to
Biomass estimates of total grass than a
traditional regression tree.
Biomass-Random Forests(VVI)
Random Forests provided a better solution to
Biomass estimates of total forbes than a
traditional regression tree.
Biomass-Random Forests(GLI)
GLI did not Perform as well as the VVI, however, the
results are still acceptable. GEMI and Varigreen did
not perform well, and are not presented here. Total
Biomass.
Biomass-Random Forests(GLI)
Forbes Biomass.
Biomass - Neural Networks
Biomass - Neural Networks (VVI)
Neural Networks provided a good solution to
Biomass estimates using the VVI. Whereas
the relationship was not as strong as the
Random Forests, the fit was adequate.
Biomass - Neural Networks (GLI)
Neural Networks also provided a good
solution to Biomass estimates using the GLI.
Whereas the relationship was not as strong as
the Random Forests, the fit was adequate,
and performs on par with the VVI.
Varigreen and GEMI performed poorly with
the resulting pseudo R2 values ~ R = 0.23
Conclusions
• The Modified VVI worked well, as did the GLI. Overall, the VVI was a
better predictor of total above ground biomass, however GLI also
performed at an acceptable rate.
• I am really on the fence with object oriented versus traditional ISO
classification. I have had great luck with object-oriented classification
using imagery such as NAIP, or even resolution down to about 6”. In
some respects object oriented worked better, in other cases it did not.
I am not aware of this being done at sub ¼” pixel resolution.
• Random Forests, CART, and Neural Networks work well in predicting
total above ground biomass based on this study.
• The project was a lot of fun!
Moving Forward
• Biomass collection from each sub-plot will not be collapsed into a single
sample. Samples for 2019 were processed a couple weeks ago, models are
in progress – promising!
• There is a continued interest in determining if we can estimate biomass for
certain plant types (equisitem, forbes, big bluestem, etc). To that end,
biomass has been separated based on type. Currently grass estimates are
robust.
• While we have been able to get reasonable results in classifying down to
species (not presented here), additional work is warranted.
• Focus efforts more heavily on issues pertaining to non-native grass
encroachment.
• We recently acquired a Sentera NDVI sensor mounted on a DJI Mavic. We
are anxious to see how this performs in comparison.
Questions?

Contenu connexe

Tendances

Gis application on forest management
Gis application on forest managementGis application on forest management
Gis application on forest managementprahladpatel6
 
Forecasting Wheat Yield and Production for Punjab Province, Pakistan from Sat...
Forecasting Wheat Yield and Production for Punjab Province, Pakistan from Sat...Forecasting Wheat Yield and Production for Punjab Province, Pakistan from Sat...
Forecasting Wheat Yield and Production for Punjab Province, Pakistan from Sat...CIMMYT
 
Module 5 - EN - Promoting data use III: Most frequent data analysis techniques
Module 5 - EN - Promoting data use III: Most frequent data analysis techniques Module 5 - EN - Promoting data use III: Most frequent data analysis techniques
Module 5 - EN - Promoting data use III: Most frequent data analysis techniques Alberto González-Talaván
 
Maximize the value of Earth Observation Data in a Big Data World
Maximize the value of Earth Observation Data in a Big Data WorldMaximize the value of Earth Observation Data in a Big Data World
Maximize the value of Earth Observation Data in a Big Data WorldBYTE Project
 
Mapping trees species in geographical and environmental space: a comparison o...
Mapping trees species in geographical and environmental space: a comparison o...Mapping trees species in geographical and environmental space: a comparison o...
Mapping trees species in geographical and environmental space: a comparison o...World Agroforestry (ICRAF)
 

Tendances (8)

Gis application on forest management
Gis application on forest managementGis application on forest management
Gis application on forest management
 
Forecasting Wheat Yield and Production for Punjab Province, Pakistan from Sat...
Forecasting Wheat Yield and Production for Punjab Province, Pakistan from Sat...Forecasting Wheat Yield and Production for Punjab Province, Pakistan from Sat...
Forecasting Wheat Yield and Production for Punjab Province, Pakistan from Sat...
 
Module 5 - EN - Promoting data use III: Most frequent data analysis techniques
Module 5 - EN - Promoting data use III: Most frequent data analysis techniques Module 5 - EN - Promoting data use III: Most frequent data analysis techniques
Module 5 - EN - Promoting data use III: Most frequent data analysis techniques
 
Sonti's Article
Sonti's ArticleSonti's Article
Sonti's Article
 
Grace Villamor - Bundling And Negotiating Ecosystem Services For Small Scale ...
Grace Villamor - Bundling And Negotiating Ecosystem Services For Small Scale ...Grace Villamor - Bundling And Negotiating Ecosystem Services For Small Scale ...
Grace Villamor - Bundling And Negotiating Ecosystem Services For Small Scale ...
 
MEDECOSXII_CMostert_Final
MEDECOSXII_CMostert_FinalMEDECOSXII_CMostert_Final
MEDECOSXII_CMostert_Final
 
Maximize the value of Earth Observation Data in a Big Data World
Maximize the value of Earth Observation Data in a Big Data WorldMaximize the value of Earth Observation Data in a Big Data World
Maximize the value of Earth Observation Data in a Big Data World
 
Mapping trees species in geographical and environmental space: a comparison o...
Mapping trees species in geographical and environmental space: a comparison o...Mapping trees species in geographical and environmental space: a comparison o...
Mapping trees species in geographical and environmental space: a comparison o...
 

Similaire à NDGeospatialSummit2019 - Estimation of Total Above Ground Biomass from High-Resolution UAS Imagery

2015 05 Scaling from seeds to ecosystems
2015 05 Scaling from seeds to ecosystems2015 05 Scaling from seeds to ecosystems
2015 05 Scaling from seeds to ecosystemsTimeScience
 
Presentation1 - Basis of application of Ecogeography in PGR
Presentation1 - Basis of application of Ecogeography in PGRPresentation1 - Basis of application of Ecogeography in PGR
Presentation1 - Basis of application of Ecogeography in PGRMauricio Parra Quijano
 
Ausplots Training - Session 2
Ausplots Training - Session 2Ausplots Training - Session 2
Ausplots Training - Session 2bensparrowau
 
PhD defense Julien Troudet (29/11/2017)
PhD defense Julien Troudet (29/11/2017)PhD defense Julien Troudet (29/11/2017)
PhD defense Julien Troudet (29/11/2017)Julien Troudet
 
GB20 Nodes Training Course 2013, module 5B: Latest trends in data analysis
GB20 Nodes Training Course 2013, module 5B: Latest trends in data analysisGB20 Nodes Training Course 2013, module 5B: Latest trends in data analysis
GB20 Nodes Training Course 2013, module 5B: Latest trends in data analysisDag Endresen
 
15 energy from_biomass_p_paul
15 energy from_biomass_p_paul15 energy from_biomass_p_paul
15 energy from_biomass_p_paulanandonliner
 
Module 4B - EN - Promoting data use II: use in key scientific and policy areas
Module 4B - EN - Promoting data use II: use in key scientific and policy areasModule 4B - EN - Promoting data use II: use in key scientific and policy areas
Module 4B - EN - Promoting data use II: use in key scientific and policy areasAlberto González-Talaván
 
NACCB 2016 Madison WI
NACCB 2016 Madison WINACCB 2016 Madison WI
NACCB 2016 Madison WIPeter Solymos
 
Deforestation diagnostics
Deforestation diagnosticsDeforestation diagnostics
Deforestation diagnosticsCIFOR-ICRAF
 
2015-08-13 ESA: NextGen tools for scaling from seeds to traits to ecosystems
2015-08-13 ESA: NextGen tools for scaling from seeds to traits to ecosystems2015-08-13 ESA: NextGen tools for scaling from seeds to traits to ecosystems
2015-08-13 ESA: NextGen tools for scaling from seeds to traits to ecosystemsTimeScience
 
TraitCapture: NextGen phenomics tools for lab and field [ComBio2015]
TraitCapture: NextGen phenomics tools for lab and field [ComBio2015]TraitCapture: NextGen phenomics tools for lab and field [ComBio2015]
TraitCapture: NextGen phenomics tools for lab and field [ComBio2015]TimeScience
 
Spatio-­‐temporal Sensor Integration, Analysis, Classification or Can Exascal...
Spatio-­‐temporal Sensor Integration, Analysis, Classification or Can Exascal...Spatio-­‐temporal Sensor Integration, Analysis, Classification or Can Exascal...
Spatio-­‐temporal Sensor Integration, Analysis, Classification or Can Exascal...Joel Saltz
 
The evolution of the Sentinel Landscape initiative
The evolution of the Sentinel Landscape initiativeThe evolution of the Sentinel Landscape initiative
The evolution of the Sentinel Landscape initiativeWorld Agroforestry (ICRAF)
 
Discovering the impact of climate change on the marine species, Aquamaps
Discovering the impact of climate change on the marine species, AquamapsDiscovering the impact of climate change on the marine species, Aquamaps
Discovering the impact of climate change on the marine species, AquamapsiMarine283644
 

Similaire à NDGeospatialSummit2019 - Estimation of Total Above Ground Biomass from High-Resolution UAS Imagery (20)

2014 nyu-bio-talk
2014 nyu-bio-talk2014 nyu-bio-talk
2014 nyu-bio-talk
 
Presentation1 ecogeographic basis
Presentation1 ecogeographic basisPresentation1 ecogeographic basis
Presentation1 ecogeographic basis
 
2015 05 Scaling from seeds to ecosystems
2015 05 Scaling from seeds to ecosystems2015 05 Scaling from seeds to ecosystems
2015 05 Scaling from seeds to ecosystems
 
Presentation1 - Basis of application of Ecogeography in PGR
Presentation1 - Basis of application of Ecogeography in PGRPresentation1 - Basis of application of Ecogeography in PGR
Presentation1 - Basis of application of Ecogeography in PGR
 
Ausplots Training - Session 2
Ausplots Training - Session 2Ausplots Training - Session 2
Ausplots Training - Session 2
 
PhD defense Julien Troudet (29/11/2017)
PhD defense Julien Troudet (29/11/2017)PhD defense Julien Troudet (29/11/2017)
PhD defense Julien Troudet (29/11/2017)
 
GB20 Nodes Training Course 2013, module 5B: Latest trends in data analysis
GB20 Nodes Training Course 2013, module 5B: Latest trends in data analysisGB20 Nodes Training Course 2013, module 5B: Latest trends in data analysis
GB20 Nodes Training Course 2013, module 5B: Latest trends in data analysis
 
15 energy from_biomass_p_paul
15 energy from_biomass_p_paul15 energy from_biomass_p_paul
15 energy from_biomass_p_paul
 
Development of the Global Exposure Database (GED)
Development of the Global Exposure Database (GED)Development of the Global Exposure Database (GED)
Development of the Global Exposure Database (GED)
 
Module 4B - EN - Promoting data use II: use in key scientific and policy areas
Module 4B - EN - Promoting data use II: use in key scientific and policy areasModule 4B - EN - Promoting data use II: use in key scientific and policy areas
Module 4B - EN - Promoting data use II: use in key scientific and policy areas
 
(gis)
 (gis) (gis)
(gis)
 
NACCB 2016 Madison WI
NACCB 2016 Madison WINACCB 2016 Madison WI
NACCB 2016 Madison WI
 
Deforestation diagnostics
Deforestation diagnosticsDeforestation diagnostics
Deforestation diagnostics
 
2015-08-13 ESA: NextGen tools for scaling from seeds to traits to ecosystems
2015-08-13 ESA: NextGen tools for scaling from seeds to traits to ecosystems2015-08-13 ESA: NextGen tools for scaling from seeds to traits to ecosystems
2015-08-13 ESA: NextGen tools for scaling from seeds to traits to ecosystems
 
TraitCapture: NextGen phenomics tools for lab and field [ComBio2015]
TraitCapture: NextGen phenomics tools for lab and field [ComBio2015]TraitCapture: NextGen phenomics tools for lab and field [ComBio2015]
TraitCapture: NextGen phenomics tools for lab and field [ComBio2015]
 
Public Meeting Presentation 121509 Final
Public Meeting Presentation 121509 FinalPublic Meeting Presentation 121509 Final
Public Meeting Presentation 121509 Final
 
Landrace gap analysis for the CGIAR, AGM 2017 Brussels
Landrace gap analysis for the CGIAR, AGM 2017 BrusselsLandrace gap analysis for the CGIAR, AGM 2017 Brussels
Landrace gap analysis for the CGIAR, AGM 2017 Brussels
 
Spatio-­‐temporal Sensor Integration, Analysis, Classification or Can Exascal...
Spatio-­‐temporal Sensor Integration, Analysis, Classification or Can Exascal...Spatio-­‐temporal Sensor Integration, Analysis, Classification or Can Exascal...
Spatio-­‐temporal Sensor Integration, Analysis, Classification or Can Exascal...
 
The evolution of the Sentinel Landscape initiative
The evolution of the Sentinel Landscape initiativeThe evolution of the Sentinel Landscape initiative
The evolution of the Sentinel Landscape initiative
 
Discovering the impact of climate change on the marine species, Aquamaps
Discovering the impact of climate change on the marine species, AquamapsDiscovering the impact of climate change on the marine species, Aquamaps
Discovering the impact of climate change on the marine species, Aquamaps
 

Plus de North Dakota GIS Hub

NDGeospatialSummit2022 - ArcGIS Solutions: A tour for Water Utilities
NDGeospatialSummit2022 - ArcGIS Solutions: A tour for Water UtilitiesNDGeospatialSummit2022 - ArcGIS Solutions: A tour for Water Utilities
NDGeospatialSummit2022 - ArcGIS Solutions: A tour for Water UtilitiesNorth Dakota GIS Hub
 
NDGeospatialSummit2022 - SaaS GIS for Local Government
NDGeospatialSummit2022 - SaaS GIS for Local GovernmentNDGeospatialSummit2022 - SaaS GIS for Local Government
NDGeospatialSummit2022 - SaaS GIS for Local GovernmentNorth Dakota GIS Hub
 
NDGeospatialSummit2022 - Carpe Geo A Field Guide To Enjoying Your Career Journey
NDGeospatialSummit2022 - Carpe Geo A Field Guide To Enjoying Your Career JourneyNDGeospatialSummit2022 - Carpe Geo A Field Guide To Enjoying Your Career Journey
NDGeospatialSummit2022 - Carpe Geo A Field Guide To Enjoying Your Career JourneyNorth Dakota GIS Hub
 
NDGeospatialSummit2022 - STARK COUNTY, ND: POWERING PROCESSES WITH MODERN GIS
NDGeospatialSummit2022 - STARK COUNTY, ND: POWERING PROCESSES WITH MODERN GISNDGeospatialSummit2022 - STARK COUNTY, ND: POWERING PROCESSES WITH MODERN GIS
NDGeospatialSummit2022 - STARK COUNTY, ND: POWERING PROCESSES WITH MODERN GISNorth Dakota GIS Hub
 
NDGeospatialSummit2022 - USGS The National Map
NDGeospatialSummit2022 - USGS The National MapNDGeospatialSummit2022 - USGS The National Map
NDGeospatialSummit2022 - USGS The National MapNorth Dakota GIS Hub
 
NDGeospatialSummit2022 - Geospatial Program Update
NDGeospatialSummit2022 - Geospatial Program UpdateNDGeospatialSummit2022 - Geospatial Program Update
NDGeospatialSummit2022 - Geospatial Program UpdateNorth Dakota GIS Hub
 
NDGeospatialSummit2022 - The North Dakota Parcel Program
NDGeospatialSummit2022 - The North Dakota Parcel ProgramNDGeospatialSummit2022 - The North Dakota Parcel Program
NDGeospatialSummit2022 - The North Dakota Parcel ProgramNorth Dakota GIS Hub
 
NDGeospatialSummit2022 - GIS FOR NEXT GENERATION 9-1-1 The Value Add Proposition
NDGeospatialSummit2022 - GIS FOR NEXT GENERATION 9-1-1 The Value Add PropositionNDGeospatialSummit2022 - GIS FOR NEXT GENERATION 9-1-1 The Value Add Proposition
NDGeospatialSummit2022 - GIS FOR NEXT GENERATION 9-1-1 The Value Add PropositionNorth Dakota GIS Hub
 
NDGeospatialSummit2022 - UAS Lidar Mission Planning and Applications
NDGeospatialSummit2022 - UAS Lidar Mission Planning and ApplicationsNDGeospatialSummit2022 - UAS Lidar Mission Planning and Applications
NDGeospatialSummit2022 - UAS Lidar Mission Planning and ApplicationsNorth Dakota GIS Hub
 
NDGeospatialSummit2022 - StoryMaps : Let’s Talk About Science!
NDGeospatialSummit2022 - StoryMaps : Let’s Talk About Science!NDGeospatialSummit2022 - StoryMaps : Let’s Talk About Science!
NDGeospatialSummit2022 - StoryMaps : Let’s Talk About Science!North Dakota GIS Hub
 
NDGeospatialSummit2022 - Learning on the Fly: The MN GIS Response to Bird Flu...
NDGeospatialSummit2022 - Learning on the Fly: The MN GIS Response to Bird Flu...NDGeospatialSummit2022 - Learning on the Fly: The MN GIS Response to Bird Flu...
NDGeospatialSummit2022 - Learning on the Fly: The MN GIS Response to Bird Flu...North Dakota GIS Hub
 
NDGeospatialSummit2022 - GIS Applications in Right of Way: A Case Study on th...
NDGeospatialSummit2022 - GIS Applications in Right of Way: A Case Study on th...NDGeospatialSummit2022 - GIS Applications in Right of Way: A Case Study on th...
NDGeospatialSummit2022 - GIS Applications in Right of Way: A Case Study on th...North Dakota GIS Hub
 
NDGeospatialSummit2022 - Soil carbon in the Red River Valley: towards precisi...
NDGeospatialSummit2022 - Soil carbon in the Red River Valley: towards precisi...NDGeospatialSummit2022 - Soil carbon in the Red River Valley: towards precisi...
NDGeospatialSummit2022 - Soil carbon in the Red River Valley: towards precisi...North Dakota GIS Hub
 
NDGeospatialSummit2022 - Well Rounded with GIS Managing the Flow of Informati...
NDGeospatialSummit2022 - Well Rounded with GIS Managing the Flow of Informati...NDGeospatialSummit2022 - Well Rounded with GIS Managing the Flow of Informati...
NDGeospatialSummit2022 - Well Rounded with GIS Managing the Flow of Informati...North Dakota GIS Hub
 
NDGeospatialSummit2022 - KEEPING IT FRESH: ASSET COLLECTOR FOR ND STATE PARKS
NDGeospatialSummit2022 - KEEPING IT FRESH: ASSET COLLECTOR FOR ND STATE PARKSNDGeospatialSummit2022 - KEEPING IT FRESH: ASSET COLLECTOR FOR ND STATE PARKS
NDGeospatialSummit2022 - KEEPING IT FRESH: ASSET COLLECTOR FOR ND STATE PARKSNorth Dakota GIS Hub
 
NDGeospatialSummit2022 - A Geospatial Analysis of the Association Between Chr...
NDGeospatialSummit2022 - A Geospatial Analysis of the Association Between Chr...NDGeospatialSummit2022 - A Geospatial Analysis of the Association Between Chr...
NDGeospatialSummit2022 - A Geospatial Analysis of the Association Between Chr...North Dakota GIS Hub
 
NDGeospatialSummit2022 - Using Machine Learning and Quasi Binomial Model to P...
NDGeospatialSummit2022 - Using Machine Learning and Quasi Binomial Model to P...NDGeospatialSummit2022 - Using Machine Learning and Quasi Binomial Model to P...
NDGeospatialSummit2022 - Using Machine Learning and Quasi Binomial Model to P...North Dakota GIS Hub
 
NDGeospatialSummit2022 - Replacing the Clipboard with ESRI’s Survey123
NDGeospatialSummit2022 - Replacing the Clipboard with ESRI’s Survey123NDGeospatialSummit2022 - Replacing the Clipboard with ESRI’s Survey123
NDGeospatialSummit2022 - Replacing the Clipboard with ESRI’s Survey123North Dakota GIS Hub
 
NDGeospatialSummit2022 - A Look Underneath the Climax Slump
NDGeospatialSummit2022 - A Look Underneath the Climax SlumpNDGeospatialSummit2022 - A Look Underneath the Climax Slump
NDGeospatialSummit2022 - A Look Underneath the Climax SlumpNorth Dakota GIS Hub
 
NDGeospatialSummit2022 - NDDOT ROADWAY DATA FROM PAST TO PRESENT A TALE OF IN...
NDGeospatialSummit2022 - NDDOT ROADWAY DATA FROM PAST TO PRESENT A TALE OF IN...NDGeospatialSummit2022 - NDDOT ROADWAY DATA FROM PAST TO PRESENT A TALE OF IN...
NDGeospatialSummit2022 - NDDOT ROADWAY DATA FROM PAST TO PRESENT A TALE OF IN...North Dakota GIS Hub
 

Plus de North Dakota GIS Hub (20)

NDGeospatialSummit2022 - ArcGIS Solutions: A tour for Water Utilities
NDGeospatialSummit2022 - ArcGIS Solutions: A tour for Water UtilitiesNDGeospatialSummit2022 - ArcGIS Solutions: A tour for Water Utilities
NDGeospatialSummit2022 - ArcGIS Solutions: A tour for Water Utilities
 
NDGeospatialSummit2022 - SaaS GIS for Local Government
NDGeospatialSummit2022 - SaaS GIS for Local GovernmentNDGeospatialSummit2022 - SaaS GIS for Local Government
NDGeospatialSummit2022 - SaaS GIS for Local Government
 
NDGeospatialSummit2022 - Carpe Geo A Field Guide To Enjoying Your Career Journey
NDGeospatialSummit2022 - Carpe Geo A Field Guide To Enjoying Your Career JourneyNDGeospatialSummit2022 - Carpe Geo A Field Guide To Enjoying Your Career Journey
NDGeospatialSummit2022 - Carpe Geo A Field Guide To Enjoying Your Career Journey
 
NDGeospatialSummit2022 - STARK COUNTY, ND: POWERING PROCESSES WITH MODERN GIS
NDGeospatialSummit2022 - STARK COUNTY, ND: POWERING PROCESSES WITH MODERN GISNDGeospatialSummit2022 - STARK COUNTY, ND: POWERING PROCESSES WITH MODERN GIS
NDGeospatialSummit2022 - STARK COUNTY, ND: POWERING PROCESSES WITH MODERN GIS
 
NDGeospatialSummit2022 - USGS The National Map
NDGeospatialSummit2022 - USGS The National MapNDGeospatialSummit2022 - USGS The National Map
NDGeospatialSummit2022 - USGS The National Map
 
NDGeospatialSummit2022 - Geospatial Program Update
NDGeospatialSummit2022 - Geospatial Program UpdateNDGeospatialSummit2022 - Geospatial Program Update
NDGeospatialSummit2022 - Geospatial Program Update
 
NDGeospatialSummit2022 - The North Dakota Parcel Program
NDGeospatialSummit2022 - The North Dakota Parcel ProgramNDGeospatialSummit2022 - The North Dakota Parcel Program
NDGeospatialSummit2022 - The North Dakota Parcel Program
 
NDGeospatialSummit2022 - GIS FOR NEXT GENERATION 9-1-1 The Value Add Proposition
NDGeospatialSummit2022 - GIS FOR NEXT GENERATION 9-1-1 The Value Add PropositionNDGeospatialSummit2022 - GIS FOR NEXT GENERATION 9-1-1 The Value Add Proposition
NDGeospatialSummit2022 - GIS FOR NEXT GENERATION 9-1-1 The Value Add Proposition
 
NDGeospatialSummit2022 - UAS Lidar Mission Planning and Applications
NDGeospatialSummit2022 - UAS Lidar Mission Planning and ApplicationsNDGeospatialSummit2022 - UAS Lidar Mission Planning and Applications
NDGeospatialSummit2022 - UAS Lidar Mission Planning and Applications
 
NDGeospatialSummit2022 - StoryMaps : Let’s Talk About Science!
NDGeospatialSummit2022 - StoryMaps : Let’s Talk About Science!NDGeospatialSummit2022 - StoryMaps : Let’s Talk About Science!
NDGeospatialSummit2022 - StoryMaps : Let’s Talk About Science!
 
NDGeospatialSummit2022 - Learning on the Fly: The MN GIS Response to Bird Flu...
NDGeospatialSummit2022 - Learning on the Fly: The MN GIS Response to Bird Flu...NDGeospatialSummit2022 - Learning on the Fly: The MN GIS Response to Bird Flu...
NDGeospatialSummit2022 - Learning on the Fly: The MN GIS Response to Bird Flu...
 
NDGeospatialSummit2022 - GIS Applications in Right of Way: A Case Study on th...
NDGeospatialSummit2022 - GIS Applications in Right of Way: A Case Study on th...NDGeospatialSummit2022 - GIS Applications in Right of Way: A Case Study on th...
NDGeospatialSummit2022 - GIS Applications in Right of Way: A Case Study on th...
 
NDGeospatialSummit2022 - Soil carbon in the Red River Valley: towards precisi...
NDGeospatialSummit2022 - Soil carbon in the Red River Valley: towards precisi...NDGeospatialSummit2022 - Soil carbon in the Red River Valley: towards precisi...
NDGeospatialSummit2022 - Soil carbon in the Red River Valley: towards precisi...
 
NDGeospatialSummit2022 - Well Rounded with GIS Managing the Flow of Informati...
NDGeospatialSummit2022 - Well Rounded with GIS Managing the Flow of Informati...NDGeospatialSummit2022 - Well Rounded with GIS Managing the Flow of Informati...
NDGeospatialSummit2022 - Well Rounded with GIS Managing the Flow of Informati...
 
NDGeospatialSummit2022 - KEEPING IT FRESH: ASSET COLLECTOR FOR ND STATE PARKS
NDGeospatialSummit2022 - KEEPING IT FRESH: ASSET COLLECTOR FOR ND STATE PARKSNDGeospatialSummit2022 - KEEPING IT FRESH: ASSET COLLECTOR FOR ND STATE PARKS
NDGeospatialSummit2022 - KEEPING IT FRESH: ASSET COLLECTOR FOR ND STATE PARKS
 
NDGeospatialSummit2022 - A Geospatial Analysis of the Association Between Chr...
NDGeospatialSummit2022 - A Geospatial Analysis of the Association Between Chr...NDGeospatialSummit2022 - A Geospatial Analysis of the Association Between Chr...
NDGeospatialSummit2022 - A Geospatial Analysis of the Association Between Chr...
 
NDGeospatialSummit2022 - Using Machine Learning and Quasi Binomial Model to P...
NDGeospatialSummit2022 - Using Machine Learning and Quasi Binomial Model to P...NDGeospatialSummit2022 - Using Machine Learning and Quasi Binomial Model to P...
NDGeospatialSummit2022 - Using Machine Learning and Quasi Binomial Model to P...
 
NDGeospatialSummit2022 - Replacing the Clipboard with ESRI’s Survey123
NDGeospatialSummit2022 - Replacing the Clipboard with ESRI’s Survey123NDGeospatialSummit2022 - Replacing the Clipboard with ESRI’s Survey123
NDGeospatialSummit2022 - Replacing the Clipboard with ESRI’s Survey123
 
NDGeospatialSummit2022 - A Look Underneath the Climax Slump
NDGeospatialSummit2022 - A Look Underneath the Climax SlumpNDGeospatialSummit2022 - A Look Underneath the Climax Slump
NDGeospatialSummit2022 - A Look Underneath the Climax Slump
 
NDGeospatialSummit2022 - NDDOT ROADWAY DATA FROM PAST TO PRESENT A TALE OF IN...
NDGeospatialSummit2022 - NDDOT ROADWAY DATA FROM PAST TO PRESENT A TALE OF IN...NDGeospatialSummit2022 - NDDOT ROADWAY DATA FROM PAST TO PRESENT A TALE OF IN...
NDGeospatialSummit2022 - NDDOT ROADWAY DATA FROM PAST TO PRESENT A TALE OF IN...
 

Dernier

Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Drew Madelung
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersThousandEyes
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024The Digital Insurer
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 3652toLead Limited
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Paola De la Torre
 
Google AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGGoogle AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGSujit Pal
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Alan Dix
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountPuma Security, LLC
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsMaria Levchenko
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitecturePixlogix Infotech
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Igalia
 
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking MenDelhi Call girls
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure servicePooja Nehwal
 

Dernier (20)

Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101
 
Google AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGGoogle AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAG
 
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...Swan(sea) Song – personal research during my six years at Swansea ... and bey...
Swan(sea) Song – personal research during my six years at Swansea ... and bey...
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC Architecture
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
 
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
 

NDGeospatialSummit2019 - Estimation of Total Above Ground Biomass from High-Resolution UAS Imagery

  • 1. Estimation of Total Above-Ground Biomass Using High Resolution UAS Imagery David Kramar1, Valquiria Quirino1, Asami Minei1, Alison Wallace, and Breanna Huynh2 Minnesota State University Moorhead 1Department of Anthropology and Earth Sciences 2Department of Biological Sciences
  • 2. Objectives • Collect and Classify High-Resolution Imagery from Two Sites using an Off-The-Shelf UAS • Compare multiple Non-NIR Vegetation Indices to Determine which Produces the Best Results • Determine Which Index Best Estimates Above Ground Biomass Source: https://www.mnstate.edu/sciencecenter/
  • 3. The Regional Science Center • The Minnesota State University Moorhead Regional Science Center (RSC) is located 16 miles east of Moorhead, adjacent to Buffalo River State Park (BRSP)1.* • There are 25 rare plants and animal species within the complex BRSP, RSC, and Bluestem Prairie2.** Source: https://www.mnstate.edu/sciencecenter/
  • 4. The Regional Science Center • Currently several projects are underway in the Regional Science Center o One is a prairie restoration project funded under the Minnesota Environment and Natural Resources Trust Fund (ENRTF). o A second project supporting the Nutrient Network (NutNet)* – which is a coordinated research network that is quantifying the impacts of human activities on ecological systems (see http://www.nutnet.org/home for additional information about NutNet) - is also underway. o There is a rich archeology history here. • It provides programs for over 7000 Pre-K-12 students every year. • It is an excellent “Living Laboratory”. Thus, there is an INTEREST IN MONITORING VEGETATION CHANGES in this area.
  • 6. Nutrient Network Overview • At Nutrient Network ecologists and scientists are collecting data in a standardized experimental design, across a broad range of site types. • Designed to quantify community and ecosystem responses across a range of vegetative ecosystems. • Comprised of more than 40 grassland sites worldwide. • Grassroots effort. • We have two sites that we are currently monitoring.
  • 7. Plot Design of the Northern Study Area
  • 8. Software, UAS Platform, and Tools • Software: (GIS/Remote Sensing) o ArcGIS, QGIS, ORFEO TB • Software: (Statistics) o R (RATTLE Package) o R (random forest, party, tree, etc) o Statistica • Software: (UAS/Processing) o Pix4D Desktop o Pix4D Capture o Drone Harmony • UAS Platform: • Phantom 4 Advanced • Example of Tools (GIS) • Spatial Analyst Extension • Segment Mean Shift • Nibble • Region • Etc
  • 9. Vegetation Indices (Minei 2019) • We focused on two sites. • Imagery was collected every 2 weeks from mid-May until the end of September. • We wanted to determine which index did the best job identifying healthy versus unhealthy vegetation, and which index would best predict total above-ground biomass. • All vegetation indexes were generated and interpreted by student Asami Minei. The sheer volume of data she has analyzed is overwhelming. • Modified VVI • 𝑉𝑉𝐼 = 𝐿𝑜𝑔 10 1 − 𝑅−𝑅𝑜 𝑅+𝑅𝑜 + 10 1 − 𝐺−𝐺𝑜 𝐺+𝐺𝑜 + 10 1 − 𝐵−𝐵𝑜 𝐵+𝐵𝑜 + 10 • Modified Global Environmental Monitoring Index • 𝐺𝐸𝑀𝐼 = 𝑒𝑡𝑎 1 − 0.25 ∗ 𝑒𝑡𝑎 − 𝑝𝑅𝑒𝑑−0.125 1−𝑝𝑅𝑒𝑑 • where 𝑒𝑡𝑎 = 2∗ 𝑝𝐺𝑟𝑒𝑒𝑛2−𝑝𝑅𝑒𝑑2 +1.5 ∗𝑝𝐺𝑟𝑒𝑒𝑛+0.5 ∗𝑝𝑅𝑒𝑑 𝑝𝐺𝑟𝑒𝑒𝑛+𝑝𝑅𝑒𝑑+0.5 • Green Leaf Index • GLI = 2 ∗ρ𝐺𝑟𝑒𝑒𝑛−𝑝𝑅𝑒𝑑 −ρ𝐵𝑙𝑢𝑒 2 ∗ρ𝐺𝑟𝑒𝑒𝑛+𝑝𝑅𝑒𝑑+ρ𝐵𝑙𝑢𝑒 • VariGreen • 𝑉𝐴𝑅𝐼 = 𝑝𝐺𝑟𝑒𝑒𝑛−𝑝𝑅𝑒𝑑 𝑝𝐺𝑟𝑒𝑒𝑛+𝑝𝑅𝑒𝑑−𝑝𝐵𝑙𝑢𝑒 • Green-Red Vegetation Index • 𝐺𝑅𝑉𝐼 = ρ𝐺𝑟𝑒𝑒𝑛 −𝑝𝑅𝑒𝑑 ρ𝐺𝑟𝑒𝑒𝑛+ρ𝑅𝑒𝑑 • We also calculated a soil-adjusted index (modified version of SAVI) which appeared to perform well. Due to time limitations, we do not present that here.
  • 12. Vegetation Indices VVI – Greater Detail GLI – More Washed Out Minei, A. 2019
  • 17. Biomass Estimates • Biomass was collected in 10 cm wide x 100 cm long strips from two locations in each plot. • This resulted in 60 sub-plots. During the first collection, the sub-plots were combined (DOH!), which reduced the total sample size to 30. Moving forward these will be kept separate. • We extracted the min, max, mean, and standard deviation of the different vegetation indices as they related to the biomass sub-plots. • Models were developed using polynomial regression, random forests/CART, and neural networks. • Today I will present results from Decision Trees (CART), Random Forests, and Neural Networks.
  • 19. Biomass Estimates 1. Collected 2. Sorted 3. Identified 4. Weighed
  • 20. Biomass Estimates • Initial Models were Developed in R using both command line and Rattle. • Methods were implemented into an R-Package that will be distributed via the R CRAN in the future (later talk at conference). • Models were implemented spatially in both ArcGIS and R (due to R- package development). • Results were cross-validated. This year we have training AND validation biomass plots that allow us to assess the models in areas outside of the Nutrient Network proper. • While we present the estimates as “High, Medium, Low”, actual estimates were in grams, and then converted to grams per meter squared.
  • 23. Biomass-Decision Tree (VVI) Generally predicted total biomass well based off the modified VVI.
  • 24. Biomass-Decision Tree (VVI) Generally predicted grass biomass well based off the modified VVI.
  • 25. Biomass-Decision Tree (VVI) Prediction of forbes biomass was not as robust as grass and total biomass.
  • 26. Biomass-Decision Tree (GLI) GLI performed better than GEMI, but not as well as the VVI.
  • 27. Biomass-Decision Tree (GLI) GLI performed better than GEMI, but not as well as the VVI. Total Biomass
  • 28. Biomass-Decision Tree (GLI) Similar to forbes estimates for the VVI, estimates from the GLI were also lower
  • 29. Biomass-Decision Tree (VariGreen) VariGreen also performed better than GEMI, But not as well as the VVI and not as well as the GLI. As such – No Map For You! Come Back 1 Month!
  • 30. Biomass-Decision Tree (GEMI) GEMI did not perform well in terms of biomass estimates. As a result, the map is a HOT MESS and not presented!
  • 32. Biomass-Random Forests(VVI) Random Forests provided a better solution to Biomass estimates than a traditional regression tree.
  • 33. Biomass-Random Forests(VVI) Random Forests provided a better solution to Biomass estimates of total grass than a traditional regression tree.
  • 34. Biomass-Random Forests(VVI) Random Forests provided a better solution to Biomass estimates of total forbes than a traditional regression tree.
  • 35. Biomass-Random Forests(GLI) GLI did not Perform as well as the VVI, however, the results are still acceptable. GEMI and Varigreen did not perform well, and are not presented here. Total Biomass.
  • 37. Biomass - Neural Networks
  • 38. Biomass - Neural Networks (VVI) Neural Networks provided a good solution to Biomass estimates using the VVI. Whereas the relationship was not as strong as the Random Forests, the fit was adequate.
  • 39. Biomass - Neural Networks (GLI) Neural Networks also provided a good solution to Biomass estimates using the GLI. Whereas the relationship was not as strong as the Random Forests, the fit was adequate, and performs on par with the VVI. Varigreen and GEMI performed poorly with the resulting pseudo R2 values ~ R = 0.23
  • 40. Conclusions • The Modified VVI worked well, as did the GLI. Overall, the VVI was a better predictor of total above ground biomass, however GLI also performed at an acceptable rate. • I am really on the fence with object oriented versus traditional ISO classification. I have had great luck with object-oriented classification using imagery such as NAIP, or even resolution down to about 6”. In some respects object oriented worked better, in other cases it did not. I am not aware of this being done at sub ¼” pixel resolution. • Random Forests, CART, and Neural Networks work well in predicting total above ground biomass based on this study. • The project was a lot of fun!
  • 41. Moving Forward • Biomass collection from each sub-plot will not be collapsed into a single sample. Samples for 2019 were processed a couple weeks ago, models are in progress – promising! • There is a continued interest in determining if we can estimate biomass for certain plant types (equisitem, forbes, big bluestem, etc). To that end, biomass has been separated based on type. Currently grass estimates are robust. • While we have been able to get reasonable results in classifying down to species (not presented here), additional work is warranted. • Focus efforts more heavily on issues pertaining to non-native grass encroachment. • We recently acquired a Sentera NDVI sensor mounted on a DJI Mavic. We are anxious to see how this performs in comparison. Questions?