Use of Unmanned Aerial Systems for Hydrological Monitoring

Page created by Derrick Barker
 
CONTINUE READING
Use of Unmanned Aerial Systems for Hydrological Monitoring
Use of Unmanned Aerial
                     Systems for Hydrological Monitoring
                                        Prof. Salvatore Manfreda and HARMONIOUS TEAM
                                              Dipartimento di Ingegneria Civile, Edile e Ambientale (DICEA)
                                      Professor of Hydrology and Advanced Monitoring - www.salvatoremanfreda.it
                                            Chair of the COST Action Harmonious - www.costharmonious.eu
                                                          e-mail: salvatore.manfreda@unina.it
                                                                Twitter: @Sal_Manfreda

HARMONIOUS TEAM: Salvatore Manfreda, Brigitta Toth, David Helman, Giorgos Mallinis, Richard Lucas, Pauline Miller, Antonino Maltese, Petr
Dvořák, Sander Mücher, Giuseppe Ciraolo, Yijian Zeng, Zhongbo Su, Jana Müllerová, Nunzio Romano, Xurxo Gago, Ruodan Zhuang, Goran Tmusic,
Helge Aasen, Mike James, Gil Concalves, Eyal Ben-Dor, Anna Brook, Maria Polinova, Jose Juan Arranz, Janos Meszaros, Ruodan Zhuang, Kasper
Johansen, Yoann Malbeteau, Matthew McCabe, Isabel Pedroso de Lima, Corine Davis, Sorin Herban, Sophie Pearce, Robert Ljubicic, Salvador
Peña-Haro, Matthew Perks, Flavia Tauro, Alonso Pizarro, Silvano Fortunato Dal Sasso, Dariia Strelnikova, Salvatore Grimaldi, Ian Maddock, Gernot
Paulus, Jasna Plavšic, Dušan Prodanovic, Rafi Kent, Josef Brůna, Martynas Bučas, Joan Estrany, Adrien Michez, Martin Mokroš, Maria Tsiafouli,
João L. M. P. de Lima
Use of Unmanned Aerial Systems for Hydrological Monitoring
2        Unmanned Arial Systems (UAS) for Environmental Monitoring
    Scope: improve ability to monitor river basin processes

    Variable of interests: state of the vegetation, soil water content, streamflow (flow
    velocity and water level), flooded areas and river morphology.

    Objective: To define integrated procedures to improve hydrological/hydraulic
    monitoring capacity using UAS.

    Scale: from plot-scale to the river basin scale providing operational monitoring tools.

    Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
Use of Unmanned Aerial Systems for Hydrological Monitoring
3
                                                      Environmental Monitoring

                            Global                                            Few hectars                        Few m2

                      Up to 30cm                                              Up to 1cm                         Up to 1cm

                                                                                            Comparable Scales

    Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
Use of Unmanned Aerial Systems for Hydrological Monitoring
4                                                                   UAS vs Satellite

                                                                                       (Manfreda et al., Remote Sensing 2018)

    Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
Use of Unmanned Aerial Systems for Hydrological Monitoring
5       Number of articles extracted from the database ISI web of knowledge
                                                  3000                                                                               100%
                                                                                      automation of a single or multiple vehicles,
                                                                                      tracking and flight control systems,           90%
                                                  2500                                hardware and software innovations,
                                                                                                                                     80%

                                                                                                                                            Percentage of Environmental Studies (%)
                                                                                      tracking of moving targets,
                                                                                      image correction and mapping performance       70%
                                                  2000                                assessment
                               Number of Papers

                                                                                                                                     60%

                                                  1500                                                                               50%
                                                         UAS applications
                                                                                                                                     40%
                                                  1000   Environmental Applications
                                                                                                                                     30%
                                                         Percentage of Environmental Studies
                                                                                                                                     20%
                                                   500
                                                                                                                                     10%

                                                     0                                                                               0%
                                                         1990
                                                         1991
                                                         1992
                                                         1993
                                                         1994
                                                         1995
                                                         1996
                                                         1997
                                                         1998
                                                         1999
                                                         2000
                                                         2001
                                                         2002
                                                         2003
                                                         2004
                                                         2005
                                                         2006
                                                         2007
                                                         2008
                                                         2009
                                                         2010
                                                         2011
                                                         2012
                                                         2013
                                                         2014
                                                         2015
                                                         2016
                                                         2017
                                                                                        Year

                                                         from 1990 up to 2017 (last access 15/01/2018)
                                                                                                                      (Manfreda et al., Remote Sensing 2018)

    Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
Use of Unmanned Aerial Systems for Hydrological Monitoring
6           CA16219 - COST Action HARMONIOUS

    • A network of scientists is currently cooperating within the
      framework of a COST (European Cooperation in Science and
      Technology) Action named “Harmonious”.
    • The aim of “Harmonious” is to promote monitoring strategies,
      establish harmonized monitoring practices, and transfer most
      recent advances on UAS methodologies to others within a global
      network.
    • HARMONIOUS involves 36 Countries

    Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
Use of Unmanned Aerial Systems for Hydrological Monitoring
Action Chair Salvatore Manfreda
                                     Vice Chair Brigitta Toth
                      Science Communications Manager: László Bertalan
                                STSM coordinator: Isabel De Lima                       WG2:
7
                         Training School Coordinator: Giuseppe Ciraolo            Vegetation Status            WG5: Harmonization of
                             ITC Grant Coordinator: Martin Mokros                Leader Antonino Maltese        methods and results
                                                                                 Vice leader Felix Frances         Leader Eyal Ben Dor
                                                                                Vice leader Jana Mullerova       Vice leader Flavia Tauro
                                              WG1: UAS data
                                               processing
                                              Leader Sorin Herban                       WG3
                                            Vice leader Corine Davis             Soil Water Content
                                                                                    Leader Yijian Zeng                Harmonization
                                                                                 Vice leader Anna Brook                 of different
                                                                                                                        procedures
                                          Geometric Correction                           WG4                          and algorithms
                                          and image calibration                   Leader Matthew Perks                  in different
                                                                              Vice leader Dariia Strelnikova          environments

                                                                                River morphology
                                                 Contrast
                                               Enhancement
                                                                                    Stream flow

    Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
Use of Unmanned Aerial Systems for Hydrological Monitoring
8                                                        Stay Connected with Us

                                Link                                                 Link

                                      Link                                           Link

                                                                              Link

    Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
Use of Unmanned Aerial Systems for Hydrological Monitoring
WG1
                                                                                                    UAS data processing
9         Examples of Common Image Artifacts

                                                                              a) saturated image;
                                                                              b) vignetting;
                                                                              c) chromatic aberration;
                                                                              d) mosaic blurring in overlap area;
                                                                              e) incorrect colour balancing;
                                                                              f) hotspots on mosaic due to
                                                                                 bidirectional reflectance effects;
                                                                              g) relief displacement (tree lean)
                                                                                 effects in final image mosaic;
                                                                              h) Image distortion due to DSM
                                                                                 errors;
                                                                              i) mosaic gaps caused by
                                                                                 incorrect orthorectification or
                                                                                 missing images.
(Whitehead and Hugenholtz, 2014)

    Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
Use of Unmanned Aerial Systems for Hydrological Monitoring
WG2
                                                                                                    Vegetation Status
10

       Comparison between a Satellite, CubeSat and UAS NDVI map

         Multi-spectral false colour (near infrared, red, green) imagery collected over the RoBo Alsahba date palm farm
         near Al Kharj, Saudi Arabia. Imagery (from L-R) shows the resolution differences between: (A) UAV mounted
         Parrot Sequoia sensor at 50 m height (0.05 m); (B) a WorldView-3 image (1.24 m); and (C) Planet CubeSat data
         (approx. 3 m), collected on the 13th, 29° and 27th March 2018, respectively.

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
WG3
                                                                               Soil Water Content Monitoring
11

a) Soil moisture downscaling
   workflow based on random
   forest regression (RF);
b) The importance of land
   surface features for the RF
   model;
c) RF-based soil moisture
   products at 15cm;
d) Comparison       of   the
   downscaled soil moisture
   with in-situ measurements.

                                                                                         (Su et al., Water 2020)

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
WG4
                                                                                                     River Monitoring
12

     Image Velocimetry
     2-D flow velocity field derived using
     an optical camera mounted on a
     quadcopter hovering over a portion
     of the Bradano river system in
     southern Italy. One of the images
     used for the analysis is shown as a
     background, where surface
     features used by flow tracking
     algorithms are highlighted in the
     insets (a, b).

                                                                               (Manfreda and McCabe, Hydrolink 2019)

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
13
                            Stream Flow Monitoring – Data Collection for
                                 Benchmarking Optical Techniques

 Image velocimetry data available for 13 case studies                          Article: (Perks et al., ESSD 2019)
                                                                                 Data: (Perks et al., ESSD 2019)

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
WG4
                                                                                                                        River Monitoring
14

                                                                               Comparison among the most recent Image Velocimetry Techniques

                                                                                                      (Pearce et al., Remote Sensing 2020)

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
WG4
                                                                                                                                                                                                                                                                                                                River Monitoring
15        UAS data Collection: How to improve DSM accuracy
                                                                                                         10 0                                                                                                                10 1

                                                                                                                                                                                                                                                                                            Fig. 1: Comparison of results obtained changing
                                                                                                                                                                                 (A)                                                                                                (B)
                          (A)                       (B)                     (C)
                                                                                                                                                              Single Flight                                                                                      Single Flight

                                                                                                                                                                                                                                                                                            the number of GCPs and adopting a single flight
                                                                                                                                                              Combined Flights                                                                                   Combined Flights
                                                                                                   10 -1
                                                                                                                                                                                                                             10 0

                                                                                  RMSEX,Y (m)

                                                                                                                                                                                              RMSEZ (m)
                                                                                                   10 -2
                                                                                                                                                           RMSE X,Y= 0.2cm                                                                                           RMSE Z= 10cm
                                                                                                                                                                                                                                                                                            or a two flights dataset on the plane (A) and z-
                          (D)                       (E)                     (F)                    10 -3
                                                                                                                                                                                                                            10 -1
                                                                                                                                                                                                                                                                                            axes (B).
                                                                                                                                                                                                                                             RMSE Z= 3.5cm
                                                                                                   10 -4                                                                                                                    10 -2
                                                                                                                                        3        4    5       6      7       8    9                                                          3      4        5   6     7       8       9
                                                                                                                                                       N. GCPs                                                                                               N. GCPs

                                                                                                                                                                                   Fig. 1                                                                                                   Fig. 2: RMSE of the 3D model as a function of the
                                                                                                                                                                                                                                                                                            mean distance between GCPs obtained for the

                                                                                           Relative change of RMSEX,Y (%)
                                                                                                                            500                                                                                             300

                                                                                                                                                                                             Relative change of RMSEZ (%)
                                                                                                                                                                                                                                       (B)

                                                                                                                                                                                                                                                                                            flight N.2 (A, B) and for the combination of
                                                                                                                                       (A)                                       3 GCP                                                                                         3 GCP
                                                                                                                            400                                                  4 GCP                                                                                         4 GCP
                                                                                                                                                                                 5 GCP                                      200                                                5 GCP
        Fligth   Fligth plan      Level      Camera       Avg GSD    Images                                                 300                                                  6 GCP                                                                                         6 GCP

                                                                                                                                                                                                                                                                                            flights N.1 and N.4 (C, D).
                                                                                                                                                                                 7 GCP                                                                                         7 GCP
                                Above the      Tilt       (cm/px )                                                                                                               8 GCP                                      100                                                8 GCP
                                                                                                                            200                                                  9 GCP                                                                                         9 GCP
                                 Ground     (degree)                                                                                     R2=0.15
                                   (m)                                                                                      100                                                                                               0
         N.1                       60          0°           1.9       276                                                     0
                                                                                                                                                                                                                            -100                                     R2=0.08
                                                                                                                            -100

         N.2                       60          0°           1.9       268                                                   -200                                                                                            -200
                                                                                                                                   0                  50              100              150                                         0                    5            10                15
                                                                                                                                                 Mean Planar Distance (m)                                                                        Mean Vertical Distance (m)

         N.3                       60         70°            -        271
                                                                                           Relative change of RMSEX,Y (%)

                                                                                                                            600                                                                                             500

                                                                                                                                                                                             Relative change of RMSEZ (%)
                                                                                                                                       (C)                                       3 GCP                                                 (D)                                     3 GCP
                                                                                                                            500                                                  4 GCP                                      400                                                4 GCP
         N.4                       60         20°           2.0       273                                                   400
                                                                                                                                                                                 5 GCP                                                                                         5 GCP
                                                                                                                                                                                 6 GCP                                      300                                                6 GCP
                                                                                                                                                                                 7 GCP                                                                                         7 GCP
                                                                                                                            300                                                  8 GCP                                                                                         8 GCP
                                                                                                                                             2                                   9 GCP                                      200                                                9 GCP
         N.5                       60          0°           1.9       257                                                   200             R =0.16
                                                                                                                                                                                                                            100                                        R2=0.00

                                                            3.3       85
                                                                                                                            100
                                                                                                                              0
                                                                                                                                                                                                                              0                                                                                  UAS-based tiled Model
         N.6                      120          0°                                                                                                                                                                           -100
                                                                                                                            -100
                                                                                                                            -200                                                                                            -200
                                                                                                                                   0                  50              100              150                                         0                    5            10                15
                                                                                                                                                 Mean Planar Distance (m)                                                                        Mean Vertical Distance (m)

                                                                                                                                                                                       Fig. 2                                                                                                    (Manfreda et al., Drones 2019)

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
16                                     Guidelines for UAS-based monitoring                           Action Deliverables

                                                                               (Tmušić et al., Remote Sensing 2020)

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
The influence of various study conditions on the output quality
17
                                of a UAS observation
     •   output product
     •   quality parameters
     •   software and hardware
     •   experiment design
     •   environmental conditions
     •   internal (affected) parameters

     NB: The arrow color indicates the type of relationship: gray -
     direct dependence, orange - inverse dependence.
                                                                                (Tmušić et al., Remote Sensing 2020)

      Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
18                                Activities involved in UAS-based Mapping

                                                                               (Tmušić et al., Remote Sensing 2020)

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
19                                     (Dis)advantages of Different Platforms

                                                                               (Tmušić et al., Remote Sensing 2020)

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
20
                                           Preliminary Checklist before Flying

                                                                               (Tmušić et al., Remote Sensing 2020)

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
21
              Metadata Associated with
                  each UAS-survey
 A detailed description of the flight characteristics and
 preprocessing activities carried out on the published
 data is extremely useful, not only for scientific
 reproducibility, but also to guarantee a certain quality,
 while also advancing and educating the broader field.

 All potential variables and details about a survey are
 listed in the UAS survey form reported herein.

                                                                               (Tmušić et al., Remote Sensing 2020)

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
22                     Proposed UAS Environmental Mapping Workflow

                                                                               (Tmušić et al., Remote Sensing 2020)

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
23                                                                     Final Remarks
     • UAS-based remote sensing provides new advanced procedures to monitor key variables,
       including vegetation status, soil moisture content, and stream flow.
     • The detailed description of such variables will increase our capacity to describe water
       resource availability and assist agricultural and ecosystem management.
     • HARMONIOUS COST Action is supporting the definition of standardized protocols for
       UAS-based applications to improve reliability of such technology.
     • Aim of the next activities is to specialize these guidelines on specific aspects and build
       new tools to support the use of UAS for environmental monitoring.

      Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
24              Publications produced by the COST Action HARMOIOUS
 Tmušić, G.; Manfreda, S.; Aasen, H.; James, M.R.; Gonçalves, G.; Ben-Dor, E.; Brook, A.; Polinova, M.; Arranz, J.J.; Mészáros, J.; Zhuang, R.;
 Johansen, K.; Malbeteau, Y.; de Lima, I.P.; Davids, C.; Herban, S.; McCabe, M.F. Current Practices in UAS-based Environmental
 Monitoring, Remote Sensing, 12, 1001, 2020. [pdf]
 Pearce, S.;R.; R. Ljubičić; S. Peña-Haro; M. Perks; F. Tauro; A. Pizarro; S.F. Dal Sasso; D. Strelnikova; S. Grimaldi; I. Maddock; G. Paulus; J.
 Plavšić; D. Prodanović; S. Manfreda, An Evaluation of Image Velocimetry Techniques under Low Flow Conditions and High Seeding
 Densities Using Unmanned Aerial Systems, Remote Sensing, 12, 232, 2020. [pdf]
 Perks, M. T., S. Fortunato Dal Sasso, A. Hauet, S. Pearce, S. Peña-Haro, F. Tauro, S. Grimaldi, B. Hortobágyi, M. Jodeau, J. Le Coz, I.
 Maddock, L. Pénard, and S. Manfreda, Towards harmonization of image velocimetry techniques for river surface velocity
 observations, Earth System Science Data Discussion, 2019. [pdf]
 Manfreda, S., S. F. Dal Sasso, A. Pizarro, F. Tauro, Chapter 10: New Insights Offered by UAS for River Monitoring, Applications of Small
 Unmanned Aircraft Systems: Best Practices and Case Studies, CRC Press, Taylor & Francis Groups, 211, 2019.
 Manfreda, S., M.F. McCabe. Emerging earth observing platforms offer new insights into hydrological processes, Hydrolink, 1, 8-9,
 2019. [pdf]
 Manfreda, S., P. Dvorak, J. Mullerova, S. Herban, P. Vuono, J.J. Arranz Justel, M. Perks, Assessing the Accuracy of Digital Surface Models
 Derived from Optical Imagery Acquired with Unmanned Aerial Systems, Drones, 3(1), 15, 2019. [pdf]
 Manfreda, S., M. F. McCabe, P. E. Miller, R. Lucas, V. Pajuelo Madrigal, G. Mallinis, E. Ben-Dor, D. Helman, L. Estes, G. Ciraolo, J. Müllerová,
 F. Tauro, M. I. de Lima, J. L. M. P. de Lima, A. Maltese, F. Frances, K. Caylor, M. Kohv, M. Perks, G. Ruiz-Pérez, Z. Su, G. Vico, and B. Toth, On
 the Use of Unmanned Aerial Systems for Environmental Monitoring, Remote Sensing, 10(4), 641; 2018. [pdf]

     Prof. Salvatore Manfreda – Università degli Studi di Napoli Federico II
You can also read