Using R as an environment for automatic extraction of forest growth - PowerPoint PPT Presentation

Waldwachstum Systemanalyse Using R as an environment for automatic extraction of forest growth parameters from terrestrial laser scanner data useR! 2008 Dortmund 13.08.2008 1/ 17 Dr. Hans-Joachim Klemmt

Waldwachstum Systemanalyse Contents • Introduction • Relevant parameters for forest inventory purposes • Determination of stem positions • Calculation of tree heights • Estimation of diameters in breast height (DBH) • Calculation of volume of stem axis • Performance of the developed system (case study „Selb“) • Summary and perspectives 2/ 17

Terrestrial Laserscanner m m 2 1 - 6 60 - 200 m 90° Systemanalyse Waldwachstum 3/ 17

R-Package „RLaserForest“ Systemanalyse Waldwachstum 4/ 17

Waldwachstum Systemanalyse R-Package RLaserForest: Determination of stem positions (slide 1) 5/ 17

Waldwachstum Systemanalyse R-Package RLaserForest: Determination of stem positions (slide 2) 6/ 17

Waldwachstum Systemanalyse R-Package RLaserForest : Height calculation and Determination of diameters in breast height (DBH) 7/ 17

R-Package RLaserForest : Calculation of stem volume Systemanalyse Waldwachstum 8/ 17

case study „Selb“ 50°09’12‘‘ N 12°11’55‘‘ O Systemanalyse Waldwachstum 9/ 17

Images of case study stand Systemanalyse Waldwachstum 10/ 17

Waldwachstum Systemanalyse description of data measurement in field • 37 Norway spruce trees + 13 Scots pine trees Applicated in RLaserForest 37 Norway spruce trees + 9 Scots pine trees Norway spruce: mean DBH 38,54 cm (20,65-61,25cm); mean height: 30,9m (21,35-38,32) Scots pine: mean DBH 38,07 cm (31,25-49,9cm); mean height: 32,10m (26,08-34,22) 11/ 17

Waldwachstum Systemanalyse Results (case study „Selb“) here: length of stems Manual measured stem length vs. Terrestrial laserscanner stem length 40 Deviation between manual and terrestrial laser stem length measurement 1.0 stem length by terrestrial laserscanning [m] mean: -0.169 mean: -0.046 35 0.5 deviation [m] 30 0.0 25 -0.5 20 Norway spruce Scots pine Scots pine Norway spruce 15 12/ 17 15 20 25 30 35 40 manual measured stem length [m]

Waldwachstum Systemanalyse Results (case study „Selb“) here: volume of stem axis calculated in sections vs. volume by quadrature of cubic spline 4 volume of manual measured sections [m³] 3 2 1 Scots pine Norway spruce 13/ 17 0 0 1 2 3 4 volume by cubic spline interpolation of laser data [m³]

Waldwachstum Systemanalyse Results here: attempt to explain the deviation between real and calculated volume Tree no. 2, Norway spruce Tree no. 21, Norway spruce 30 30 0. 0. 0.25 0.25 0.20 0.20 radius [m] radius [m] 0.15 0.15 0.10 0.10 0.05 0.05 stemprofile by cubic spline interpolation stemprofile by cubic spline interpolation manual caliper measured section radii manual caliper measured section radii 0.00 0.00 0 5 10 15 20 25 30 0 10 20 30 stem length [m] stem length [m] Tree no. 8, Scots pine Tree no. 12, Scots pine 0.30 0.30 0.25 0.25 0.20 0.20 radius [m] radius [m] 0.15 0.15 14/ 17 0.10 0.10 0.05 0.05 stemprofile by cubic spline interpolation stemprofile by cubic spline interpolation manual caliper measured section radii manual caliper measured section radii 0.00 0.00 0 5 10 15 20 25 30 5 10 15 20 25 stem length [m] stem length [m]

Waldwachstum Systemanalyse Summary and perspectives TODOs: improvement of volume calculation improvement of diameter estimation for excentric stems afterwards: automated determination of tree species (classification) afterwards: automated separation of crown parameters (spectral clustering) Objective:modular built R-Package „RLaserForest“ for automatic extraction of forest growth relevant inventory parameters by the use of the statistic programming language R 15/ 17

Waldwachstum Systemanalyse I want to say thank you to: • BaySF: Forstbetrieb Selb (insbes. Herrn Michael Grosch und Herrn Hubert Fellermeyer) for enabling case study in field • LfWwk: Herrn Stefan Seifert, Herrn Thomas Seifert, Herrn Istvan Pal, Herrn Gerhard Schütze, Frau Andrea Oumeddah as well as Herrn Sebastian Seibold and Herrn Martin Stary 16/ 17 • colleagues from FMI UHUL from Czech republic (cooperation within a common Interreg IIIa-project)

Waldwachstum Systemanalyse Thank you very much for your attention! I am looking forward to a fruitful discussion … 17/ 17