Modelling Stand Variables of Beech Coppice Forest Using Spectral Sentinel-2A Data and the Machine Learning Approach

Background and Purpose: Coppice forests have a particular socio-economic and ecological role in forestry and environmental management. Their production sustainability and spatial stability become imperative for forestry sector as well as for local and global communities. Recently, integrated forest inventory and remotely sensed data analysed with non-parametrical statistical methods have enabled more detailed insight into forest structural characteristics. The aim of this research was to estimate forest attributes of beech coppice forest stands in the Sarajevo Canton through the integration of inventory and Sentinel S2A satellite data using machine learning methods. Materials and Methods: Basal area, mean stand diameter, growing stock and total volume data were determined from the forest inventory designed for represented stands of coppice forests. Spectral data were collected from bands of Sentinel S2A satellite image, vegetation indices (difference, normalized difference and ratio vegetation index) and biophysical variables (fraction of absorbed photosynthetically active radiation, leaf area index, fraction of vegetation cover, chlorophyll content in the leaf and canopy water content). Machine learning rule-based M5 model tree (M5P) and random forest (RF) methods were used for forest attribute estimation. Predictor subset selection was based on wrapping assuming M5P and RF learning schemes. Models were developed on training data subsets (402 sample plots) and evaluations were performed on validation data subsets (207 sample plots). Performance of the models was evaluated by the percentage of the root mean squared error over the mean value (rRMSE) and the square of the correlation coefficient between the observed and estimated stand variables. Results and Conclusions: Predictor subset selection resulted in a varied number of predictors for forest attributes and methods with their larger contribution in RF (between 8 and 11). Spectral biophysical variables dominated in subsets. The RF resulted in smaller errors for training sets for all attributes than M5P, while both methods delivered very high errors for validation sets (rRMSE above 50%). The lowest rRMSE of 50% was obtained for stand basal area. The observed variability explained by the M5P and RF models in training subsets was about 30% and 95% respectively, but those values were lower in test subsets (below 12%) but still significant. Differences of the sample and modelled forest attribute means were not significant, while modelled variability for all forest attributes was significantly lower (p<0.01). It seems that additional information is needed to increase prediction accuracy, so stand information (management classes, site class, soil type, canopy closure and others), new sampling strategy and new spectral products could be integrated and examined in further more complex modelling of forest attributes.

Download Full-text

Recognition of Maize Phenology in Sentinel Images with Machine Learning

Sensors ◽

10.3390/s22010094 ◽

2021 ◽

Vol 22 (1) ◽

pp. 94

Author(s):

Alvaro Murguia-Cozar ◽

Antonia Macedo-Cruz ◽

Demetrio Salvador Fernandez-Reynoso ◽

Jorge Arturo Salgado Transito

Keyword(s):

Machine Learning ◽

Satellite Image ◽

Spatial Association ◽

Machine Learning Algorithms ◽

Support Vector ◽

Classification Models ◽

Phenological Stages ◽

Area Index ◽

Stage Of Development ◽

Sentinel 2

The scarcity of water for agricultural use is a serious problem that has increased due to intense droughts, poor management, and deficiencies in the distribution and application of the resource. The monitoring of crops through satellite image processing and the application of machine learning algorithms are technological strategies with which developed countries tend to implement better public policies regarding the efficient use of water. The purpose of this research was to determine the main indicators and characteristics that allow us to discriminate the phenological stages of maize crops (Zea mays L.) in Sentinel 2 satellite images through supervised classification models. The training data were obtained by monitoring cultivated plots during an agricultural cycle. Indicators and characteristics were extracted from 41 Sentinel 2 images acquired during the monitoring dates. With these images, indicators of texture, vegetation, and colour were calculated to train three supervised classifiers: linear discriminant (LD), support vector machine (SVM), and k-nearest neighbours (kNN) models. It was found that 45 of the 86 characteristics extracted contributed to maximizing the accuracy by stage of development and the overall accuracy of the trained classification models. The characteristics of the Moran’s I local indicator of spatial association (LISA) improved the accuracy of the classifiers when applied to the L*a*b* colour model and to the near-infrared (NIR) band. The local binary pattern (LBP) increased the accuracy of the classification when applied to the red, green, blue (RGB) and NIR bands. The colour ratios, leaf area index (LAI), RGB colour model, L*a*b* colour space, LISA, and LBP extracted the most important intrinsic characteristics of maize crops with regard to classifying the phenological stages of the maize cultivation. The quadratic SVM model was the best classifier of maize crop phenology, with an overall accuracy of 82.3%.

Download Full-text

Landslide Susceptibility Evaluation and Management Using Different Machine Learning Methods in The Gallicash River Watershed, Iran

Remote Sensing ◽

10.3390/rs12030475 ◽

2020 ◽

Vol 12 (3) ◽

pp. 475 ◽

Cited By ~ 19

Author(s):

Alireza Arabameri ◽

Sunil Saha ◽

Jagabandhu Roy ◽

Wei Chen ◽

Thomas Blaschke ◽

...

Keyword(s):

Machine Learning ◽

Landslide Susceptibility ◽

Learning Methods ◽

Validation Data ◽

Area Index ◽

Study Region ◽

River Watershed ◽

Machine Learning Methods ◽

Natural Break ◽

Very High

This analysis aims to generate landslide susceptibility maps (LSMs) using various machine learning methods, namely random forest (RF), alternative decision tree (ADTree) and Fisher’s Linear Discriminant Function (FLDA). The results of the FLDA, RF and ADTree models were compared with regard to their applicability for creating an LSM of the Gallicash river watershed in the northern part of Iran close to the Caspian Sea. A landslide inventory map was created using GPS points obtained in a field analysis, high-resolution satellite images, topographic maps and historical records. A total of 249 landslide sites have been identified to date and were used in this study to model and validate the LSMs of the study region. Of the 249 landslide locations, 70% were used as training data and 30% for the validation of the resulting LSMs. Sixteen factors related to topographical, hydrological, soil type, geological and environmental conditions were used and a multi-collinearity test of the landslide conditioning factors (LCFs) was performed. Using the natural break method (NBM) in a geographic information system (GIS), the LSMs generated by the RF, FLDA, and ADTree models were categorized into five classes, namely very low, low, medium, high and very high landslide susceptibility (LS) zones. The very high susceptibility zones cover 15.37% (ADTree), 16.10% (FLDA) and 11.36% (RF) of the total catchment area. The results of the different models (FLDA, RF, and ADTree) were explained and compared using the area under receiver operating characteristics (AUROC) curve, seed cell area index (SCAI), efficiency and true skill statistic (TSS). The accuracy of models was calculated considering both the training and validation data. The results revealed that the AUROC success rates are 0.89 (ADTree), 0.92 (FLDA) and 0.97 (RF) and predication rates are 0.82 (ADTree), 0.79 (FLDA) and 0.98 (RF), which justifies the approach and indicates a reasonably good landslide prediction. The results of the SCAI, efficiency and TSS methods showed that all models have an excellent modeling capability. In a comparison of the models, the RF model outperforms the boosted regression tree (BRT) and ADTree models. The results of the landslide susceptibility modeling could be useful for land-use planning and decision-makers, for managing and controlling the current and future landslides, as well as for the protection of society and the ecosystem.

Download Full-text

Application of allometric systems for compatible area-based LiDAR-assisted estimation in the Province of Nova Scotia

Canadian Journal of Forest Research ◽

10.1139/cjfr-2021-0070 ◽

2021 ◽

pp. 1-10

Author(s):

Ting-Ru Yang ◽

John A. Kershaw ◽

Elizabeth McGarrigle ◽

Mark J. Ducey ◽

Dhirendra Shukla

Keyword(s):

Nova Scotia ◽

Forest Inventory ◽

Field Data ◽

Systems Approach ◽

Basal Area ◽

Average Height ◽

Permanent Sample Plots ◽

Forest Attributes ◽

Quadratic Mean Diameter

Light detection and ranging (LiDAR) is used to estimate tree, stand, and forest characteristics across large geographic areas. In the province of Nova Scotia, an enhanced forest inventory (EFI) was developed to provide high-resolution spatial forest inventory estimates across the landscape. For various forest attributes, independent LiDAR-based relationships were built leading to mathematical and biological inconsistency among forest attribute estimates. A systems approach, composed of allometric equations describing the relationships between volume per unit area, Lorey’s average height, basal area, quadratic mean diameter, and density, is developed to address these inconsistencies. Previous results showed that applying the systems approach provided reasonable and compatible estimates and eliminated inconsistency issues among forest attributes. This study evaluates application of the systems approach applied to eastern Nova Scotia using field data from a network of permanent sample plots and recent LiDAR acquisitions. The independent EFI estimates had inconsistencies of greater than 100% for basal area and implied stand-level form factor. These inconsistencies were eliminated using the systems approach. Results show that the systems approach can be scaled to larger landscape areas and that long-term field data can be leveraged to fit the allometric systems producing mathematically and biologically consistent estimates.

Download Full-text

Reuse of field data in ALS-assisted forest inventory

Silva Fennica ◽

10.14214/sf.10272 ◽

2020 ◽

Vol 54 (5) ◽

Author(s):

Ana de Lera Garrido ◽

Terje Gobakken ◽

Hans Ørka ◽

Erik Næsset ◽

Ole Bollandsås

Keyword(s):

Forest Inventory ◽

Laser Scanning ◽

Field Data ◽

Reference Data ◽

Growth Models ◽

Basal Area ◽

Tree Height ◽

Cost Component ◽

Forest Inventories ◽

Forest Attributes

Forest inventories assisted by wall-to-wall airborne laser scanning (ALS), have become common practice in many countries. One major cost component in these inventories is the measurement of field sample plots used for constructing models relating biophysical forest attributes to metrics derived from ALS data. In areas where ALS-assisted forest inventories are planned, and in which the previous inventories were performed with the same method, reusing previously acquired field data can potentially reduce costs, either by (1) temporally transferring previously constructed models or (2) projecting field reference data using growth models that can serve as field reference data for model construction with up-to-date ALS data. In this study, we analyzed these two approaches of reusing field data acquired 15 years prior to the current ALS acquisition to estimate six up-to-date forest attributes (dominant tree height, mean tree height, stem number, stand basal area, volume, and aboveground biomass). Both approaches were evaluated within small stands with sizes of approximately 0.37 ha, assessing differences between estimates and ground reference values. The estimates were also compared to results from an up-to-date forest inventory relying on concurrent field- and ALS data. The results showed that even though the reuse of historical information has some potential and could be beneficial for forest inventories, systematic errors may appear prominent and need to be overcome to use it operationally. Our study showed systematic trends towards the overestimation of lower-range ground references and underestimation of the upper-range ground references.

Download Full-text

Contribution to the automation of the calculations involving the forest inventory with the aid of an office computer

Silva Gandavensis ◽

10.21825/sg.v20i0.1001 ◽

1970 ◽

Vol 20 ◽

Author(s):

R. Goossens

Keyword(s):

Tree Species ◽

Forest Inventory ◽

Basal Area ◽

Annual Increment ◽

Mixed Stands ◽

Mean Values ◽

Standing Trees ◽

General Program ◽

Magnetic Card ◽

The Right

Contribution to the automation of the calculations involving the forest inventory with the aid of an office computer - In this contribution an attempt was made to perform the calculations involving the forest inventory by means of an office computer Olivetti P203. The general program (flowchart 1), identical for all tree species except for the values of the different parameters, occupies the tracks A and B of a magnetic card used with this computer. For each tree species one magnetic card is required, while some supplementary cards are used for the subroutines. The first subroutine (flowchart 1) enables us to preserve temporarily the subtotals between two tree species (mixed stands) and so called special or stand cards (SC). After the last tree species the totals per ha are calculated and printed on the former, the average trees occuring on the line below. Appendix 1 gives an example of a similar form resulting from calculations involving a sampling in a mixed stand consisting of Oak (code 11), Red oak (code 12), Japanese larch (code 24) and Beech (code 13). On this form we find from the left to the right: the diameter class (m), the number of trees per ha, the basal area (m2/ha), the current annual increment of the basal area (m2/year/ha), current annual volume increment (m3/year/ha), the volume (m3/ha) and the money value of the standing trees (Bfr/ha). On the line before the last, the totals of the quantities mentioned above and of all the tree species together are to be found. The last line gives a survey of the average values dg, g, ig, ig, v and w. Besides this form each stand or plot has a so-called 'stand card SC' on wich the totals cited above as well as the area of the stand or the plot and its code are stored. Similar 'stand card' may replace in many cases completely the classical index cards; moreover they have the advantage that the data can be entered directly into the computer so that further calculations, classifications or tabling can be carried out by means of an appropriate program or subroutine. The subroutine 2 (flowchart 2) illustrates the use of similar cards for a series of stands or eventually a complete forest, the real values of the different quantities above are calculated and tabled (taking into account the area). At the same time the general totals and the general mean values per ha, as well as the average trees are calculated and printed. Appendix 2 represents a form resulting from such calculations by means of subroutine 2.

Download Full-text

An approximate point-based alternative for the estimation of variance under big BAF sampling

Forest Ecosystems ◽

10.1186/s40663-021-00304-0 ◽

2021 ◽

Vol 8 (1) ◽

Author(s):

Thomas B. Lynch ◽

Jeffrey H. Gove ◽

Timothy G. Gregoire ◽

Mark J. Ducey

Keyword(s):

Forest Inventory ◽

Variance Estimation ◽

Basal Area ◽

Volume Estimation ◽

Delta Method ◽

Sampling Point ◽

Variance Estimator ◽

Computer Simulation Studies ◽

New Formula ◽

Estimation Of Variance

Abstract Background A new variance estimator is derived and tested for big BAF (Basal Area Factor) sampling which is a forest inventory system that utilizes Bitterlich sampling (point sampling) with two BAF sizes, a small BAF for tree counts and a larger BAF on which tree measurements are made usually including DBHs and heights needed for volume estimation. Methods The new estimator is derived using the Delta method from an existing formulation of the big BAF estimator as consisting of three sample means. The new formula is compared to existing big BAF estimators including a popular estimator based on Bruce’s formula. Results Several computer simulation studies were conducted comparing the new variance estimator to all known variance estimators for big BAF currently in the forest inventory literature. In simulations the new estimator performed well and comparably to existing variance formulas. Conclusions A possible advantage of the new estimator is that it does not require the assumption of negligible correlation between basal area counts on the small BAF factor and volume-basal area ratios based on the large BAF factor selection trees, an assumption required by all previous big BAF variance estimation formulas. Although this correlation was negligible on the simulation stands used in this study, it is conceivable that the correlation could be significant in some forest types, such as those in which the DBH-height relationship can be affected substantially by density perhaps through competition. We derived a formula that can be used to estimate the covariance between estimates of mean basal area and the ratio of estimates of mean volume and mean basal area. We also mathematically derived expressions for bias in the big BAF estimator that can be used to show the bias approaches zero in large samples on the order of $\frac {1}{n}$ 1 n where n is the number of sample points.

Download Full-text

Corn Nitrogen Status Diagnosis with an Innovative Multi-Parameter Crop Circle Phenom Sensing System

Remote Sensing ◽

10.3390/rs13030401 ◽

2021 ◽

Vol 13 (3) ◽

pp. 401

Author(s):

Cadan Cummings ◽

Yuxin Miao ◽

Gabriel Dias Paiao ◽

Shujiang Kang ◽

Fabián G. Fernández

Keyword(s):

Machine Learning ◽

Chlorophyll Content ◽

Vegetation Index ◽

Soil Drainage ◽

Management Information ◽

Area Index ◽

Sensing System ◽

Extreme Gradient Boosting ◽

Split Plot ◽

N Status

Accurate and non-destructive in-season crop nitrogen (N) status diagnosis is important for the success of precision N management (PNM). Several active canopy sensors (ACS) with two or three spectral wavebands have been used for this purpose. The Crop Circle Phenom sensor is a new integrated multi-parameter proximal ACS system for in-field plant phenomics with the capability to measure reflectance, structural, and climatic attributes. The objective of this study was to evaluate this multi-parameter Crop Circle Phenom sensing system for in-season diagnosis of corn (Zea mays L.) N status across different soil drainage and tillage systems under variable N supply conditions. The four plant metrics used to approximate in-season N status consist of aboveground biomass (AGB), plant N concentration (PNC), plant N uptake (PNU), and N nutrition index (NNI). A field experiment was conducted in Wells, Minnesota during the 2018 and the 2019 growing seasons with a split-split plot design replicated four times with soil drainage (drained and undrained) as main block, tillage (conventional, no-till, and strip-till) as split plot, and pre-plant N (PPN) rate (0 to 225 in 45 kg ha−1 increment) as the split-split plot. Crop Circle Phenom measurements alongside destructive whole plant samples were collected at V8 +/−1 growth stage. Proximal sensor metrics were used to construct regression models to estimate N status indicators using simple regression (SR) and eXtreme Gradient Boosting (XGB) models. The sensor derived indices tested included normalized difference vegetation index (NDVI), normalized difference red edge (NDRE), estimated canopy chlorophyll content (eCCC), estimated leaf area index (eLAI), ratio vegetation index (RVI), canopy chlorophyll content index (CCCI), fractional photosynthetically active radiation (fPAR), and canopy and air temperature difference (ΔTemp). Management practices such as drainage, tillage, and PPN rate were also included to determine the potential improvement in corn N status diagnosis. Three of the four replicated drained and undrained blocks were randomly selected as training data, and the remaining drained and undrained blocks were used as testing data. The results indicated that SR modeling using NDVI would be sufficient for estimating AGB compared to more complex machine learning methods. Conversely, PNC, PNU, and NNI all benefitted from XGB modeling based on multiple inputs. Among different approaches of XGB modeling, combining management information and Crop Circle Phenom measurements together increased model performance for predicting each of the four plant N metrics compared with solely using sensing data. The PPN rate was the most important management metric for all models compared to drainage and tillage information. Combining Crop Circle Phenom sensor parameters and management information is a promising strategy for in-season diagnosis of corn N status. More studies are needed to further evaluate this new integrated sensing system under diverse on-farm conditions and to test other machine learning models.

Download Full-text

Rapid Assessment of Tree Damage Resulting from a 2020 Windstorm in Iowa, USA

Forests ◽

10.3390/f12050555 ◽

2021 ◽

Vol 12 (5) ◽

pp. 555

Author(s):

Thomas C. Goff ◽

Mark D. Nelson ◽

Greg C. Liknes ◽

Tivon E. Feeley ◽

Scott A. Pugh ◽

...

Keyword(s):

Forest Inventory ◽

Damage Zone ◽

Satellite Image ◽

Rapid Assessment ◽

Aerial Survey ◽

Inventory Data ◽

Tree Damage ◽

Forest Inventory Data ◽

Damage Susceptibility ◽

The Impact

A need to quantify the impact of a particular wind disturbance on forest resources may require rapid yet reliable estimates of damage. We present an approach for combining pre-disturbance forest inventory data with post-disturbance aerial survey data to produce design-based estimates of affected forest area and number and volume of trees damaged or killed. The approach borrows strength from an indirect estimator to adjust estimates from a direct estimator when post-disturbance remeasurement data are unavailable. We demonstrate this approach with an example application from a recent windstorm, known as the 2020 Midwest Derecho, which struck Iowa, USA, and adjacent states on 10–11 August 2020, delivering catastrophic damage to structures, crops, and trees. We estimate that 2.67 million trees and 1.67 million m3 of sound bole volume were damaged or killed on 23 thousand ha of Iowa forest land affected by the 2020 derecho. Damage rates for volume were slightly higher than for number of trees, and damage on live trees due to stem breakage was more prevalent than branch breakage, both likely due to higher damage probability in the dominant canopy of larger trees. The absence of post-storm observations in the damage zone limited direct estimation of storm impacts. Further analysis of forest inventory data will improve understanding of tree damage susceptibility under varying levels of storm severity. We recommend approaches for improving estimates, including increasing spatial or temporal extents of reference data used for indirect estimation, and incorporating ancillary satellite image-based products.

Download Full-text

Identifying Spatiotemporal Patterns in Land Use and Cover Samples from Satellite Image Time Series

Remote Sensing ◽

10.3390/rs13050974 ◽

2021 ◽

Vol 13 (5) ◽

pp. 974

Author(s):

Lorena Alves Santos ◽

Karine Ferreira ◽

Michelle Picoli ◽

Gilberto Camara ◽

Raul Zurita-Milla ◽

...

Keyword(s):

Machine Learning ◽

Land Use ◽

Time Series ◽

Satellite Image ◽

Spatiotemporal Patterns ◽

Supervised Machine Learning ◽

Learning Methods ◽

Self Organizing Maps ◽

Machine Learning Methods ◽

Land Use And Cover

The use of satellite image time series analysis and machine learning methods brings new opportunities and challenges for land use and cover changes (LUCC) mapping over large areas. One of these challenges is the need for samples that properly represent the high variability of land used and cover classes over large areas to train supervised machine learning methods and to produce accurate LUCC maps. This paper addresses this challenge and presents a method to identify spatiotemporal patterns in land use and cover samples to infer subclasses through the phenological and spectral information provided by satellite image time series. The proposed method uses self-organizing maps (SOMs) to reduce the data dimensionality creating primary clusters. From these primary clusters, it uses hierarchical clustering to create subclusters that recognize intra-class variability intrinsic to different regions and periods, mainly in large areas and multiple years. To show how the method works, we use MODIS image time series associated to samples of cropland and pasture classes over the Cerrado biome in Brazil. The results prove that the proposed method is suitable for identifying spatiotemporal patterns in land use and cover samples that can be used to infer subclasses, mainly for crop-types.

Download Full-text

A Novel Hybrid Method for Landslide Susceptibility Mapping-Based GeoDetector and Machine Learning Cluster: A Case of Xiaojin County, China

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi10020093 ◽

2021 ◽

Vol 10 (2) ◽

pp. 93

Author(s):

Wei Xie ◽

Xiaoshuang Li ◽

Wenbin Jian ◽

Yang Yang ◽

Hongwei Liu ◽

...

Keyword(s):

Machine Learning ◽

Hybrid Method ◽

Roc Curve ◽

Landslide Susceptibility ◽

Susceptibility Mapping ◽

Assessment Model ◽

Landslide Susceptibility Mapping ◽

Support Vector ◽

Susceptibility Map ◽

Area Index

Landslide susceptibility mapping (LSM) could be an effective way to prevent landslide hazards and mitigate losses. The choice of conditional factors is crucial to the results of LSM, and the selection of models also plays an important role. In this study, a hybrid method including GeoDetector and machine learning cluster was developed to provide a new perspective on how to address these two issues. We defined redundant factors by quantitatively analyzing the single impact and interactive impact of the factors, which was analyzed by GeoDetector, the effect of this step was examined using mean absolute error (MAE). The machine learning cluster contains four models (artificial neural network (ANN), Bayesian network (BN), logistic regression (LR), and support vector machines (SVM)) and automatically selects the best one for generating LSM. The receiver operating characteristic (ROC) curve, prediction accuracy, and the seed cell area index (SCAI) methods were used to evaluate these methods. The results show that the SVM model had the best performance in the machine learning cluster with the area under the ROC curve of 0.928 and with an accuracy of 83.86%. Therefore, SVM was chosen as the assessment model to map the landslide susceptibility of the study area. The landslide susceptibility map demonstrated fit with landslide inventory, indicated the hybrid method is effective in screening landslide influences and assessing landslide susceptibility.

Download Full-text