Uneingeschränkter Zugang

An indirect approach to predict deadwood biomass in forests of Ukrainian Polissya using Landsat images and terrestrial data

Maksym Matsala
Maksym Matsala
, 
Viktor Myroniuk
Viktor Myroniuk
, 
Andrii Bilous
Andrii Bilous
, 
Andrii Terentiev
Andrii Terentiev
, 
Petro Diachuk
Petro Diachuk
 und 
Roman Zadorozhniuk
Roman Zadorozhniuk
   | 11. März 2021
Forestry Studies's Cover Image
Über diesen Artikel
Vorheriger Artikel
Nächster Artikel
Zitieren
Article Category: Research paper
Online veröffentlicht: 11. März 2021
Seitenbereich: 107 - 124
Eingereicht: 16. Sept. 2020
Akzeptiert: 16. Nov. 2020
DOI: https://doi.org/10.2478/fsmu-2020-0018
Schlüsselwörter
© 2020 Maksym Matsala et al., published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1

Location of the study site. The Chernihiv region (green) and study area (red) are coloured on the context map in the top. White lines refer to the edges of local forest management polygons used in this study, while the background is a raw Landsat image given in the band combination Near Infrared-Red-Green.
Location of the study site. The Chernihiv region (green) and study area (red) are coloured on the context map in the top. White lines refer to the edges of local forest management polygons used in this study, while the background is a raw Landsat image given in the band combination Near Infrared-Red-Green.

Figure 2

Workflow of the study. The flowchart illustrates utilization of the data, and the modelling process for classification and imputation purposes.
Workflow of the study. The flowchart illustrates utilization of the data, and the modelling process for classification and imputation purposes.

Figure 3

Ancillary datasets used for spatial modelling. The DEM variable is given in meters above sea level.
Ancillary datasets used for spatial modelling. The DEM variable is given in meters above sea level.

Figure 4

Scatterplot of the relationship between mean GSV and deadwood biomass stocks within the study area. Coloured gradient represents the density of points.
Scatterplot of the relationship between mean GSV and deadwood biomass stocks within the study area. Coloured gradient represents the density of points.

Figure 5

Importance of predictors for specific tree species and in total given by the RF classification model.
Importance of predictors for specific tree species and in total given by the RF classification model.

Figure 6

Land cover classes map (a) and tree species map within the forest mask (b) produced by the respective RF model.
Land cover classes map (a) and tree species map within the forest mask (b) produced by the respective RF model.

Figure 7

Agreement between predicted and observed values of deadwood biomass: imputed by the best k-NN (a, k = 1) and GBM (b) models.
Agreement between predicted and observed values of deadwood biomass: imputed by the best k-NN (a, k = 1) and GBM (b) models.

Figure 8

Test landscape with predicted deadwood biomass stock: imputed by the k-NN model (a) or estimated by the GBM model (b).
Test landscape with predicted deadwood biomass stock: imputed by the k-NN model (a) or estimated by the GBM model (b).

Figure 9

Subsets of predicted forest cover within the study area vs. actual linear stands across water bodies (a–b) and pathways (c–d). The green mask is produced by the RF model and the purple mask is created using GFC data.
Subsets of predicted forest cover within the study area vs. actual linear stands across water bodies (a–b) and pathways (c–d). The green mask is produced by the RF model and the purple mask is created using GFC data.

Comparison of mean and total values of deadwood biomass stock for the test landscape at polygon level.

Type of modelMean ± CI, t·ha−1Total stock, tMean difference with reference per polygon, t·ha−1
Reference7.8 ± 0.38011–
k-NN9.0 ± 0.19107+4.5
GBM8.7 ± 0.18996+4.1

Confusion matrix of RF tree species classification.

PredictedObserved
ALGLBEPEOTHERPISYPOTR
ALGL121543
BEPE012370
OTHERS00312
PISY3721170
POTR00012

Confusion matrix of RF land cover classification.

PredictedObserved
BogCroplandsForestGrasslandsOthersSettlementShrublandWater
Bog130100010
Croplands03481110191270
Forest65261601110
Grasslands122211101191
Others00000100
Settlement042021510
Shrubland1661400141
Water100000027

Hyperparameter grid for tuning GBM models in this study.

Parameter to tuneHyperparameter values
Shrinkage0.010.050.020.001
Interaction depth246810
Number of trees1000500010000
Number of observations in the terminal nodes51015
Bagging fraction0.50.751.0

Comparison of mean and total values of deadwood biomass stock within the study area produced by the best performing models.

Type of modelTree species
ALGLBEPEOTHERPISYPOTR
Mean ± SD, t·ha−1
k-NN7.0 ± 3.48.5 ± 1.98.1 ± 2.88.6 ± 1.98.1 ± 2.1
GBM6.8 ± 2.38.0 ± 1.78.1 ± 1.78.4 ± 1.58.1 ± 2.1
Total, thousands t
k-NN39.034.819.9237.64.9
GBM37.932.819.9232.04.9
eISSN:
1736-8723
Sprache:
Englisch
Zeitrahmen der Veröffentlichung:
2 Hefte pro Jahr
Fachgebiete der Zeitschrift:
Biologie, Botanik, Ökologie, andere
Zeitschrift RSS Feed