Pine Forest Management and Disturbance in Northern Poland: Combining High-Resolution 100-Year-Old Paleoecological and Remote Sensing Data

Global warming has compelled to strengthen the resilience of European forests. Due to repeated droughts and heatwaves, weakened trees become vulnerable to insect outbreaks, pathogen invasions, and strong winds. This study combines high-resolution analysis of a 100-year-old high-resolution peat archive synthesized from the Martwe peatland in Poland with remote sensing data. We present the first REVEALS based vegetation reconstruction in a tornado-hit area from Poland on the background of previous forest management in monocultural even-aged stands – Tuchola Pinewoods. During the 20th century, the pine monocultures surrounding the peatland were affected by clear-cutting and insect outbreaks. In 2012, a tornado, destroyed ca. 550 ha of pine forest around the peatland. The palynological record reflects these major events of the past 100 years as well as changes in forest practices. Our study showed the strong relationships between the decrease of Pinus sylvestris (Scots pine) in palynological record as well as planting patterns after the tornado. Moreover, past forestry practices [such as domination of Pinus sylvestris, the collapse of Picea abies (Norway spruce), low share of Betula spec. (birch) due to Pinus sylvestris promotion and probable also to a lesser by removal of Betula as a “forest weed,” and low plant coverage of tree species due to clear-cutting and cutting after insect outbreaks] were well identified in the proxy record. In monocultures managed over decades, the reconstruction of vegetation may be challenging due to changes in the age composition of the Pinus sylvestris stands. We found that through historical, remote sensing, and paleoecological data, the dynamics of disturbances such as insect outbreaks and tornadoes, as well as the changing perceptions of local society about forests, can be determined.

Download Full-text

Forest Damage Assessment Using Deep Learning on High Resolution Remote Sensing Data

Remote Sensing ◽

10.3390/rs11171976 ◽

2019 ◽

Vol 11 (17) ◽

pp. 1976 ◽

Cited By ~ 18

Author(s):

Zayd Mahmoud Hamdi ◽

Melanie Brandmeier ◽

Christoph Straub

Keyword(s):

Neural Network ◽

Remote Sensing ◽

Deep Learning ◽

Forest Management ◽

High Resolution ◽

Damage Assessment ◽

Remote Sensing Data ◽

Forest Area ◽

Forest Damage ◽

Sensing Data

Storms can cause significant damage to forest areas, affecting biodiversity and infrastructure and leading to economic loss. Thus, rapid detection and mapping of windthrows are crucially important for forest management. Recent advances in computer vision have led to highly-accurate image classification algorithms such as Convolutional Neural Network (CNN) architectures. In this study, we tested and implemented an algorithm based on CNNs in an ArcGIS environment for automatic detection and mapping of damaged areas. The algorithm was trained and tested on a forest area in Bavaria, Germany. . It is a based on a modified U-Net architecture that was optimized for the pixelwise classification of multispectral aerial remote sensing data. The neural network was trained on labeled damaged areas from after-storm aerial orthophotos of a ca. 109 k m 2 forest area with RGB and NIR bands and 0.2-m spatial resolution. Around 10 7 pixels of labeled data were used in the process. Once the network is trained, predictions on further datasets can be computed within seconds, depending on the size of the input raster and the computational power used. The overall accuracy on our test dataset was 92 % . During visual validation, labeling errors were found in the reference data that somewhat biased the results because the algorithm in some instance performed better than the human labeling procedure, while missing areas affected by shadows. Our results are very good in terms of precision, and the methods introduced in this paper have several additional advantages compared to traditional methods: CNNs automatically detect high- and low-level features in the data, leading to high classification accuracies, while only one after-storm image is needed in comparison to two images for approaches based on change detection. Furthermore, flight parameters do not affect the results in the same way as for approaches that require DSMs and DTMs as the classification is only based on the image data themselves, and errors occurring in the computation of DSMs and DTMs do not affect the results with respect to the z component. The integration into the ArcGIS Platform allows a streamlined workflow for forest management, as the results can be accessed by mobile devices in the field to allow for high-accuracy ground-truthing and additional mapping that can be synchronized back into the database. Our results and the provided automatic workflow highlight the potential of deep learning on high-resolution imagery and GIS for fast and efficient post-disaster damage assessment as a first step of disaster management.

Download Full-text