Small area estimations of proportion of forest and timber volume combining Lidar data and stereo aerial images with terrestrial data

Abstract

Methods for small area estimations were compared for estimating the proportion of forest and growing stock volume of temperate mixed forests within a district of a member state (canton) in Switzerland. The estimators combine terrestrial data with remotely sensed auxiliary data. By using different model types, different sources of auxiliary data and different methods of processing the auxiliary data, the increase in estimation precision was tested. Using the canopy height derived from remote sensing data, the growing stock volume and the proportion of forest were estimated. The regression models used for the small area estimation provided a coefficient of determination of up to 68% for the timber volume. The proportion of plots correctly classified into forest and non-forest plots ranged between 0.9 and 0.98. Models calibrated over forest area only resulted in a maximal coefficient of determination of 37%. Even though these coefficients indicate a moderate model quality, the use of remote sensing data clearly improved the estimation precision of both the proportion of forest and the growing stock volume. Generally, Lidar data led to slightly higher estimates compared to data from aerial photography. It was possible to reduce the variance of the estimated proportion of forest to nearly one tenth compared with the variance based on the terrestrial measurements alone. Similarly, the variance of the growing stock volume could be reduced to one fourth as compared with the variance based solely on the terrestrial measurements.     

Modeling to assess growth respiration of stems inPinus densiflora trees

Multiple regression analyses were applied to the respiration data obtained by an excision method to distinguish between maintenance and growth respiration in stems ofPinus densiflora. Among several types of regression models, a few models showed marked stability of coefficient of growth related respiration that are independent of degrees of correlation between predictors and any combinations of predictors. These models predicted growth respiration as 0.45 g CO₂ g (dry weight)⁻¹. At 15°C, sapwood maintenance respiration rate was estimated to 0.72 mg CO₂ g⁻¹ day⁻¹. These estimates were not different from the results obtained with standing trees.     

Using model scenarios to predict and evaluate forest-management impacts on soil base saturation at landscape level

Silviculture, forest conversion and technical tools of ecosystem management, such as forest liming, display their effects at the landscape level. Therefore their planning and control should take place at the same scale. The primary objective of this work was to assess soil chemical properties and their changes in relationship to ecosystem management, especially forest conversion and forest liming. We calculated scenario models, based on regression analysis, which allow such an examination in the context of understanding landscape processes which can be expected to operate in the sampling region. Stepwise multiple linear regression analysis was used to predict soil chemical attributes (base saturation, pH, C/N content and stock, exchangeable stocks of Ca and Mg) as indicators of site stability or off-site effects of forest ecosystems such as effects on clean drinking water from forested watersheds. Because of space limitations, in this paper only the modelling results of base saturation are presented. Base saturation was used as an integrative example for other soil chemical properties. The transformation of measurements to the regional scale, i.e., the regionalization, was calculated for the forested parts of two test regions in the Black Forest on the basis of measured chemical properties of 90–150 soil profiles per test region. The models have a spatial resolution of 50×50 m, which is a spatial scale relevant for forestry practice and forest management. Topographic variables (e.g., elevation, aspect, slope gradient, and slope length), the stratigraphic classification of the geologic substrate, stand characteristics from forest inventory data, and finally technical information about forest liming were the auxiliary variables (secondary site properties) that provided indirect information about base saturation and were available for the whole forested area of the test regions. Base saturation could be predicted with an accuracy of ~50–70% (in terms of the multiple R²) by using these properties as predictor variables in multiple linear regression analyses. The explained proportion of variance was unexpectedly high considering the high geomorphological heterogeneity of the two test regions. Based upon the regionalization models, it was possible to establish scenarios showing the landscape-related effects on base saturation that may be achieved by forest conversion towards a higher proportion of forests with broad-leaved mixed stands and by forest liming. These scenarios allow the interactions between several influencing factors and management strategies and the impacts on the target variable to be synoptically judged. Thus the presented regionalization models achieve the role of decision support tools for the planning of forest management at the landscape level. They allow an assessment of the environmental effects of forest management strategies in terms of site sustainability or preservation of water resources in forested catchments.     

Genotype-Environment Interactions and Stability Analysis for Herbage and Seed Yields of Forage Peas Under Rainfed Conditions

Twenty-five genotypes of forage peas (Pisum sativum L.) were evaluated under rainfed conditions for two years at four locations in Syria and one in Lebanon. Since local climatic variation is significant, each location in each year was treated as a separate environment, to give ten environments. Genotype-environment interactions (GXE) were analysed using linear regression techniques. There was considerable variation in herbage and seed yields within both genotypes and environments. Genotype-environment interactions were present for both herbage and seed yields; a large proportion of these interactions was accounted for by the linear regression. Although the non-linear component was also significant, its magnitude was smaller than that of the linear component. The highest and most stable herbage yields were obtained from accessions 46/61, 135/175, 3211/323, 99/103 and 88/335. The highest and most stable seed yields were produced by accessions 4/166, 2428/240, 424/242 and 88/335. Accession 88/335 from Germany, F.R. was the only genotype to combine high herbage and seed yields with wide adaptation and stability and could thus be considered the most widely adapted genotype. Other stable genotypes were identified as suitable for poor environments. Yield of the locally collected genotype, though often high, was found to be unstable. The importance of genotype-environment interactions in future breeding strategies is discussed.     

基于SVM的CFRW导电性建模预测研究

针对碳纤维增强木质复合材料（CFRW）导电性具有非线性的特点,采用基于支持向量回归机（SVM）的机器学习方法建立碳纤维增强木质复合材料导电性预测模型。结果表明,所建模型的预测精度高、泛化能力强.     

Water adsorption process of bamboo heated at low temperature

Water adsorption capacities were evaluated for moso bamboo samples that were heated at 200°C for various times and conditioned in a closed container at 97% relative humidity at 20°C. Logistic regression analysis was used for curve fitting to the adsorption data and its parameters were analyzed. These parameters were compared with those derived previously from the Dubinin and Radushkevich theory. The properties of the heat-treated samples changed after 5 h of heating. With less than 5 h of heating, hydroxyl groups provided the main adsorption sites but their numbers decreased on heating. After 5 h, gasifi cation of the bamboo increased and capillaries formed. Part of this article was presented at the 57th Annual Meeting of the Japan Wood Research Society, Hiroshima, August 2007     

Germination of <Emphasis Type="Italic">Plasmodiophora brassicae</Emphasis> resting spores stimulated by a non-host plant

Plant-induced germination of Plasmodiophora brassicae resting spores was studied in a laboratory experiment. Spore reaction was analysed in nutrient solution with exudates from growing roots of different plant species – one host plant (Brassica rapa var. pekinensis) and four non-host plants (Lolium perenne, Allium porrum, Secale cereale and Trifolium pratense) – and in controls with distilled water and nutrient solution. It was found that root exudates from L. perenne stimulated spore germination more than exudates from the other plants, including those from the host plant. The effect could not be explained by differences in the nutritional composition of the solutions due to differential uptake of the plant species, or by differences in root activity, measured as exudation of soluble sugars. This is the first time such a separation of factors has been done in analysing the influence of plants on P. brassicae germination. Although stimulation of P. brassicae resting spore germination is not restricted to the presence of host plants, it seems to vary depending on the plant species.     

Epidemiology of a Neotropical rust (Puccinia psidii): periodical analysis of the temporal progress in a perennial host (Syzygium jambos)

The epidemiology of the Neotropical rust caused by Puccinia psidii in the Indian Myrtaceae Syzygium jambos (rose apple) was investigated in central Brazil. Disease severities recorded during a 12-month period were fitted by a Fourier curve with three cyclic components or harmonics. The first, second and third harmonics accounted for 49·6, 25·9, and 1·5% of total disease measured, respectively. A highly significant linear correlation was observed between the first harmonic and the accumulated number of days having at least 6 h wetness, or a minimum of 90% relative humidity, combined with temperatures that ranged from 18 to 20°C. No significant correlation was observed between the second and third harmonics and the weather variables evaluated. Rose apple trees showed a flush of growth with new susceptible host tissue during the evaluation period, with two major outbreaks of rust of variable intensity. Host growth was fitted by a Fourier curve with two significant harmonics. The first and second harmonics accounted for 37·5 and 22·1% of total host growth measured, respectively. A highly significant negative linear correlation was observed between the first harmonic and daily maximum and minimum temperatures, rainfall, and duration of the light period. Inoculum availability was not a limiting factor for disease progress since urediniospores were present during most of the period studied. A major peak in numbers of rust spores followed the main peak of disease severity. Thus, rust epidemics on rose apple in central Brazil were shown to depend on the duration of leaf wetness in the dark, and also on night-time temperatures during that same wetness period. This study is the first example of a periodical analysis of an epidemic in a perennial crop.