Download as PDF, PPTX
![20
SMAR-EnKF
optimization
and prediction
Root mean square errors
ranging from 0.014 -
0.049 [cm3 cm-3].
Semi-arid Highlands
Temperate Forests
Temperate Forests
North American Deserts
Great PlainsTropical Wet Forests
Forested Mountains Northern Forests
(Baldwin et al., J. Hydr., 2017)](https://image.slidesharecdn.com/2019manfredaperugia1-190130201755/75/DRONES-IN-HYDROLOGY-20-2048.jpg)
![29
Surface Flow Velocity
Pixels
500 600 700 800 900 1000 1100
Pixels
400
500
600
700
800
900
[m/s]
0
0.5
1
1.5
2
2.5
3
Charcoal
y = 1.0849x + 0.0869
R² = 0.9
0
0.5
1
1.5
2
2.5
0 1 2 3
UAVderivedsurface
velocity(m/s)
Surface Velocity measure with
traditional techniques (m/s)](https://image.slidesharecdn.com/2019manfredaperugia1-190130201755/75/DRONES-IN-HYDROLOGY-29-2048.jpg)
![32
Original Video File Name: [River]_[Country][ddmmyearhhmmUTC].mov
Camera Model:
Platform used (gaugecam, drone, mobile, etc.):
Camera setting (autofocus, field of view, ISO, stabilization, …):
Video resolution (4000x2000, …)
Video frequency (Hz):
Presence of tracers and type:
Optional Info
Lumen:
Wind speed and orientation:
Case Study
River Name:
River Basin Drainage Area (km2
):
Cross-Section Coordinates (Lat, Long WGS84):
Flow regime (low, medium, high):
Ground-true availability (yes or not):
File Format (mov, avi, mp4, etc.):
Reference paper:
Processed Data
File Name of Processed Frames: [River]_[Country][ddmmyearhhmmUTC].zip
Number of frames:
Frame rate (Hz):
Pixel dimension:
Pre-processing actions (contrast correction, channel used, orthorectification,
stabilization, etc.):
Stream Flow Monitoring – Data Collection](https://image.slidesharecdn.com/2019manfredaperugia1-190130201755/75/DRONES-IN-HYDROLOGY-32-2048.jpg)
Uploaded bySalvatore Manfreda
DRONES IN HYDROLOGY
A network of scientists is cooperating within the COST Action "Harmonious" to promote UAS monitoring strategies and establish harmonized practices. The action has 36 partner countries and aims to transfer advances in UAS methods through a global network. It has five working groups focused on data processing, vegetation monitoring, soil moisture, streamflow monitoring, and harmonization. The overall goal is to increase the capacity to describe water resources and assist management through consistent UAS monitoring of key variables.
More Related Content
DRONES IN HYDROLOGY
- 1. DRONES IN HYDROLOGY: InternationalWinter School on Hydrology, Perugia, 28 Jan. - 1 Feb. 2019. Prof. Salvatore Manfreda Associate Professor of Water Management and Ecohydrology - http://www2.unibas.it/manfreda Chair of the COST Action Harmonious - http://www.costharmonious.eu
- 2. 2 A network ofscientists is currently cooperating within the framework of a COST (European Cooperation in Science and Technology) Action named “Harmonious”. The intention 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. COST Action HARMONIOUS
- 3. 3 HARMONIOUS Partners COST Countries 36Partners HARMONIOUS Network
- 4. 4 Contrast Enhancement Geometric Correction and imagecalibration WG3 Soil Moisture Content Leader Zhongbo Su Vice leader David Helman Stream flow River morphology WG2 Vegetation Status Leader Antonino Maltese Vice leader Felix Frances Harmonization of different procedures and algorithms in different environments WG1: UAS data processing Leader Pauline Miller Vice leader Victor Pajuelo Madrigal WG5: Harmonization of methods and results Leader Eyal Ben Dor Vice leader Flavia Tauro WG4 Leader Matthew Perks Vice leader Marko Kohv Action Chair Salvatore Manfreda Vice Chair Brigitta Toth Science Communications Manager: Guiomar Ruiz Perez STSM coordinator: Isabel De Lima Training School Coordinator: Giuseppe Ciraolo HARMONIOUS Action
- 5. 5 WG5: Harmonization of different proceduresand algorithms in different environments WG4: River and Streamflow monitoring WG3: Soil Moisture Monitoring WG2: Vegetation Monitoring WG1: Data Collection, Processing and Limitations b) Identification of the shared problems a) Peculiarities and specificity of each topic c) Identification of possible common strategies for the four WGs d) Definition of the correct protocol fro UAS Environmental Monitoring
- 6. 6 The Home Page- https://www.costharmonious.eu
- 7.
- 8. 8 Facebook Harmonious-European-COST- Action 352 followerson facebook https://www.facebook.com/Harmonious-European-COST- Action-485412205186817/
- 9. 9 Examples of Commonimage artifacts (Whitehead and Hugenholtz, 2014) 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. WG1: UAS data processing
- 10. 10 Comparison between aCubeSat 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. WG2 Vegetation Status
- 11. 11 UAS thermal surveyover an Aglianico vineyard in the Basilicata region (southern Italy) WG2 Vegetation Status (Manfreda et al., R.S. 2018a)
- 12. 12 How to detectwater stress from an UAV? From Xurxo Gago
- 13. 13 Aerial thermography forwater stress detection (Berni et al., 2009)
- 14. 14 Aerial thermography forwater stress detection (González-Dugoetal., 2012)
- 15. 15 How to detectthe drought from an UAV? Thermal indexes… an attempt to normalize the environment (Idso et al., 1980; Jones, 1999) § CWSI = T canopy – Twet / Tdry – Twet § IG = T dry – Tcanopy / Tcanopy – Twet § I3= T canopy – Twet / Tdry – Tcanopy § And the leaf energy balance: !" = −%&'!() * +, − +- + / 0 +, − +- %&' − !()123 − !-4 +, − +-= 567 589: 5; <)=>?'@A567B '@A < 589: 5; :"567
- 16.
- 17. 17 Relationship existing betweensurface and root-zone soil moisture Manfreda et al. (AWR - 2007) Developing a relationship between the relative soil moisture at the surface to that in deeper layers of soil would be very useful for remote sensing applications. This implies that prediction of soil moisture in the deep layer given the superficial soil moisture, has an uncertainty that increases with a reduced near surface estimate.
- 18. 18 Soil Moisture AnalyticalRelationship (SMAR) The schematization proposed assumes the soil composed of two layers, the first one at the surface of a few centimeters and the second one below with a depth that may be assumed coincident with the rooting depth of vegetation (of the order of 60–150 cm). This may allow the derivation of a function of the soil moisture in one layer as a function of the other one. Manfreda et al. (HESS - 2014) s1(t) s2(t) First layer Second layer Zr2 Zr1 !! !! = !! + (!! !!!! − !!)!!!! !!!!!!! + 1 − !! !!! !! !! − !!!!
- 19. 19 Sensitivity of SMAR’sparameters The derived root zone soil moisture (SRZ) is plotted changing the soil water loss coefficient (A), the depth of the second soil layer (B), and the soil textures (C). Manfreda et al. (HESS - 2014)
- 20. 20 SMAR-EnKF optimization and prediction Root meansquare errors ranging from 0.014 - 0.049 [cm3 cm-3]. Semi-arid Highlands Temperate Forests Temperate Forests North American Deserts Great PlainsTropical Wet Forests Forested Mountains Northern Forests (Baldwin et al., J. Hydr., 2017)
- 21. 21 Field Site ofMonteforte (SA)
- 22. 22 Optic/thermal sensors Radarsensors ( ) Me e U U M M M 1 1 )( max - - ==F )" = $ % − %' ((% − %' di Non contact equipments Advantages 1) High spatial and temporal resolution 2) Relatively low costs 3) Applicable inaccessible sections
- 23. 23 Stream flow monitoringwith UAS Particle Tracking Velocimetry (PTV) Image processing WG4: Stream Monitoring Lagrangian method (Tauro et al., 2016)
- 24. 24 Particle Tracking Particle detectionVelocity vectors
- 25. 25 Monitoring River Systems (DalSasso et al., E.M.A. 2018)
- 26. 26 Optimal parameter settingsfor PTV techniques Box plot of the relative error for the different densities investigated in the configurations: a ideal condition, b real condition (Dal Sasso et al., E.M.A. 2018) Particle displacement: Dx=1.5Dxp – Number of frames: 20 Ideal configuration Real configuration 0 1 2 3 4 5 6 x 10 -4 0 200 400 600 800 1000 1200 1400 Seeding density (ppp) N.frames fitted curve (y= 0.0002635x-1.514 ) fitted curve (y= 0.0001318x-1.514 ) fitted curve (y= 0.0000659x-1.514 ) Numerical experiments (Dx=1.5Dxp) Numerical experiments (Dx=3Dxp) Numerical experiments (Dx=6Dxp)
- 27.
- 28. 28 2-D flow velocityfield 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). Image Velocimetry
- 29. 29 Surface Flow Velocity Pixels 500600 700 800 900 1000 1100 Pixels 400 500 600 700 800 900 [m/s] 0 0.5 1 1.5 2 2.5 3 Charcoal y = 1.0849x + 0.0869 R² = 0.9 0 0.5 1 1.5 2 2.5 0 1 2 3 UAVderivedsurface velocity(m/s) Surface Velocity measure with traditional techniques (m/s)
- 30. 30 Field Experience with UAS Validationwith current meters
- 31. 31 Stream Flow Monitoring– Data Collection for Benchmarking Optical Techniques CASE STUDIES
- 32. 32 Original Video FileName: [River]_[Country][ddmmyearhhmmUTC].mov Camera Model: Platform used (gaugecam, drone, mobile, etc.): Camera setting (autofocus, field of view, ISO, stabilization, …): Video resolution (4000x2000, …) Video frequency (Hz): Presence of tracers and type: Optional Info Lumen: Wind speed and orientation: Case Study River Name: River Basin Drainage Area (km2 ): Cross-Section Coordinates (Lat, Long WGS84): Flow regime (low, medium, high): Ground-true availability (yes or not): File Format (mov, avi, mp4, etc.): Reference paper: Processed Data File Name of Processed Frames: [River]_[Country][ddmmyearhhmmUTC].zip Number of frames: Frame rate (Hz): Pixel dimension: Pre-processing actions (contrast correction, channel used, orthorectification, stabilization, etc.): Stream Flow Monitoring – Data Collection
- 33. 33 FRCs are generallyobtained using curve fitting methods with river stage (H) and discharge (Q) observations. The most common equation is: The Use of Discharge DATA: FRCs topographic surveys Velocity Measurements
- 34. 34 The Key Idea Localminima Parameter space Scheme 1 Scheme 3 Scheme 2 Scheme 4 Fitness Function Impact of physical information on the parameter space domain Manfreda et al. (HP - 2018) Time Q (m3/s) SURFACE RUNOFF SNOW MELT BASE FLOW Decomposing the parameter calibration according to the existing processes leads to more reliable model calibrations. Physical constrains Model Performances Including physical info Stream Flow Components
- 35. 35 The V WMethodQ(m3/s) H (m) Decoupling Streamflow measurements Classic Method V W Method H (m) topographic surveys V(m/s) Manfreda (JH - 2018)
- 36. 36 § FRCs derivedwith different permutation of the same dataset; § Comparison is made on the calibration dataset and on the data excluded from the calibration. Comparison of the two methodologies Manfreda (JH - 2018)
- 37. 37 § UAS-based remotesensing 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. § The wide range of applications testifies to the great potential of these techniques, but, at the same time, the variety of methodologies adopted is evidence that there is still need for harmonization efforts. Conclusion
- 38.
- 39.
- 40. 40 § Manfreda andMcCabe (2019). Emerging earth observing platforms offer new insights into hydrological processes, Hydrolink. § Perks, Hortobágyi, Le Coz, Maddock, Pearce, Tauro, Dal Sasso, Grimaldi, Manfreda (2019) Towards harmonization of image velocimetry techniques for determining open-channel flow, Earth system science data (in preparation). § Manfreda, Dvorak, Mullerova, Herban, Vuono, Arranz Justel, Perks (2019) Assessing the Accuracy of Digital Surface Models Derived from Optical Imagery Acquired with Unmanned Aerial Systems, Drones. § Manfreda, On the derivation of flow rating-curves in data-scarce environments, Journal of Hydrology, 2018. § Dal Sasso, Pizarro, Samela, Mita, and Manfreda (2018) Exploring the optimal experimental setup for surface flow velocity measurements using PTV, Environmental Monitoring and Assessment. § Manfreda, McCabe, Miller, Lucas, Pajuelo Madrigal, Mallinis, Ben-Dor, Helman, Estes, Ciraolo, Müllerová, Tauro, De Lima, De Lima, Frances, Caylor, Kohv, Maltese (2018), On the Use of Unmanned Aerial Systems for Environmental Monitoring, Remote Sensing. § Baldwin, Manfreda, Keller, and Smithwick, Predicting root zone soil moisture with soil properties and satellite near-surface moisture data at locations across the United States, Journal of Hydrology, 2017. § Manfreda, Brocca, T. Moramarco, F. Melone, and J. Sheffield, A physically based approach for the estimation of root-zone soil moisture from surface measurements, Hydrology and Earth System Sciences, 18, 1199-1212, 2014. § Manfreda, Lacava, Onorati, Pergola, Di Leo, Margiotta, and Tramutoli, On the use of AMSU-based products for the description of soil water content at basin scale, Hydrology and Earth System Sciences, 15, 2839-2852, 2011. Related Publications