Effect of game browsing on natural regeneration of European beech (Fagus sylvatica L.) forests in the Krušné hory Mts. (Czech Republic and Germany)

Tree damage by game browsing is one of the biggest threats to forest ecosystems at the time of climate change and large-scale forest disturbances. The aim of the paper was to determine the effect of browsing by ungulates on the diversity, abundance and species composition of natural regeneration in forest stands dominated by European beech (Fagus sylvatica L.). The research was conducted on 10 permanent research plots in the Krušné hory Mts. in the Czech Republic and Germany. The density of natural regeneration was in the range of 23,300–114,100 recruits ha−1. A higher proportion of silver birch (Betula pendula Roth.) and rowan (Sorbus aucuparia L.) was found in the regeneration compared to the mature stands. A total of 78% of recruits was damaged by browsing. The most frequently damaged tree species were sycamore (Acer pseudoplatanus L.; 98%) and black alder (Alnus glutinosa [L.] Gaertn.; 97%), while Norway spruce (Picea abies [L.] Karst; 31%) and sessile oak (Quercus petraea [Matt.] Liebl.; 50%) were the least affected. Seventy-nine percent of European beech recruits were damaged. The game significantly reduced the height of regeneration by up to 40%, especially by terminal browsing. Browsing also negatively affected the quality and abundance of regeneration. For successful dynamics of species-rich natural forest ecosystems, it is necessary to minimize tree damage by game browsing. These main measures include the reduction of ungulate population levels and the optimization of their age structure and sex ratio, an increase in the number of overwintering enclosures and food fields for game and a change in the political approach to game management with sufficient consideration of forestry interests.

Optimasi Model Artificial Neural Network untuk Klasifikasi Paket Jaringan

Klasifikasi jaringan internet dibutuhkan secara luas oleh berbagai pihak untuk penghematan, pengalokasian, pembatasan sumber daya internet. Berbagai teknik digunakan untuk hal tersebut, salah satunya adalah dengan pendekatan machine learning khususnya algoritma jaringan syaraf tiruan (artificial neural network) yang selanjutnya disingkat ANN. ANN bekerja dengan cara meniru cara kerja syaraf otak manusia. AAN pada penelitian ini digunakan untuk mengklasifikasikan paket jaringan. Dataset yang digunakan adalah terdiri dari 990.558 baris data, 4 kolom input(X) meliputi protocol, port, timestamp, packet length. Kolom output(Y) terdiri satu kolom 5 label (App, Sosmed, Game, Browsing, Streaming). Selanjutnya dataset ini dibagi menjadi 3 yaitu training, test, dan validation. Tujuan dari penelitian ini adalah, pertama, untuk mengetahui kemampuan algoritma ANN untuk pengklasifikasian paket berdasarkan per satu paket. Kedua, menemukan model ANN yang paling optimal untuk permasalahan diatas. Kesimpulan yang didapatkan adalah Algoritma neural network dapat digunakan pada klasifikasi paket jaringan, tetapi jika dataset yang digunakan memiliki karakteristik-karakteristik seperti jumlah variabel X kecil, data pada variabel X sangat lebar jaraknya seperti (port, packet size, time to previous packet, protocol) maka akurasi tinggi sulit untuk didapatkan. Kedua, jika permasalahan yang diselesaikan memiliki kemiripan dengan penelitian ini, maka arsitektur model ANN yang paling optimal adalah: jumlah neuron input layer adalah 7 kali jumlah variabel X, jumlah neuron hidden layer adalah 1/2,8 kali jumlah neuron input, jumlah hidden layer satu, nilai dropout 0,33, metode aktivasi tanh-tanh-softmax, metode optimasi adamax. Nilai accuracy stabil didapatkan pada iterasi (epoch) ke-600, nilai loss stabil didapatkan pada epoch ke-1000. Nilai accuracy yang didapatkan sebesar 0,8 dan nilai loss 0,32 pada iterasi ke seribu.

Teoretické východiská a praktické opatrenia na harmonizáciu záujmov lesného hospodárstva a poľovníctva na Slovensku / Theoretical bases and practical measures to harmonise the interests of forestry and game management in Slovakia

Strategic aims and priorities of forestry and game management in Slovakia are characterised. Using current knowledge in the field, common bases for both target sectors are introduced and possibilities for harmonisation of their interests are explained. Development of mutual relationship between forestry and game management is analysed and the causes of existing discrepancies are shown. The main reason of the discrepancies is unfavourable development of ungulate game population and high damage to forests stands caused by game browsing. The final part of the paper presents the recommendations for practical measures to be taken for reaching accordance between both sectors, eventually measures for achieving the primary goal to support all-society’s interests in this field. We base our suggestions on an assumption that both forestry and game management accept the proposed measures, or more precisely the measures will be implemented through mutual collaboration and integration.

Reaction of Silver fir (Abies alba Mill.) plantation to fertilization

Silver fir as a domestic species is an important tree in our forests. Unfortunately it has suffered from serious dieback in Central Europe in the last decades. Reintroduction of Silver fir is not an easy task as seedlings and saplings are sensitive to many environmental harmful factors and to game browsing damage especially in early stages after germination. The fertilization of Silver fir saplings could help shorten the sensitive period of the tree species. Silver fir research plots were founded in 1999 to observe an impact of different fertilizers on its height growth. Applications of conventional as well as modern fertilizers confirmed the positive influence on the height growth of the species. Apical dominance ratio was used to evaluate the growth status of saplings. 

Eine neue Methode zur Quantifizierung des Wildverbisses | A New Method to Assess the Effects of Game-Browsing on Forest Regeneration

In the context of the first and second Swiss National Forest Inventory (LFI), methods were developed to assess the effects of game-browsing on forest regeneration. Since the results of the LFI-1-method are closely related to the season the assessment was carried out, and the results obtained by the LFI-2-method overestimate the annual browsing rate, we propose a new procedure, which produces an unbiased estimation of the annual browsing rate from a single, non-destructive assessment.