Current issue: 56(4)
Under compilation: 57(1)
Semiparametric models, ordinary regression models and mixed models were compared for modelling stem volume in National Forest Inventory data. MSE was lowest for the mixed model. Examination of spatial distribution of residuals showed that spatial correlation of residuals is lower for semiparametric and mixed models than for parametric models with fixed regressors. Mixed models and semiparametric models can both be used for describing the effect of geographic location on stem form.
In this study, model-based and design-based inference methods are used for estimating mean volume and its standard error for systematic cluster sampling. Results obtained with models are compared to results obtained with classical methods. The data are from the Finnish National Forest Inventory. The variation of volume in ten forestry board districts in Southern Finland is studied. The variation is divided into two components: trend and correlated random errors. The effect of the trend and the covariance structure on the obtained mean volume and standard error estimates is discussed. The larger the coefficient of determination of the trend model, the smaller the model-based estimates of standard error, when compared to classical estimates. On the other hand, the wider the range and level of autocorrelation between the sample plots, the larger the model-based estimates of standard error.
Old inventory data has widely been used as prior information in forest inventory using the method of sampling with partial replacement (SPR). In this method knowledge about forest growth has not been utilized. However, the accuracy of the inventory results can be improved if this knowledge is utilized. The usability of the inventory results can be improved if the prior information is updated by treewise growth models. In this paper a statistical basis is presented for a method in which such information can be used. The applicability of the method is also discussed. An example is given to demonstrate the method.
The PDF includes an abstract in Finnish.
In remote sensing-based forest inventories 3D point cloud data, such as acquired from airborne laser scanning, are well suited for estimating the volume of growing stock and stand height, but tree species recognition often requires additional optical imagery. A combination of 3D data and optical imagery can be acquired based on aerial imaging only, by using stereo photogrammetric 3D canopy modeling. The use of aerial imagery is well suited for large-area forest inventories, due to low costs, good area coverage and temporally rapid cycle of data acquisition. Stereo-photogrammetric canopy modeling can also be applied to previously acquired imagery, such as for aerial ortho-mosaic production, assuming that the imagery has sufficient stereo overlap. In this study we compared two stereo-photogrammetric canopy models combined with contemporary satellite imagery in forest inventory. One canopy model was based on standard archived imagery acquired primarily for ortho-mosaic production, and another was based on aerial imagery whose acquisition parameters were better oriented for stereo-photogrammetric canopy modeling, including higher imaging resolution and greater stereo-coverage. Aerial and satellite data were tested in the estimation of growing stock volume, volumes of main tree species, basal area and diameter and height. Despite the better quality of the latter canopy model, the difference of the accuracy of the forest estimates based on the two different data sets was relatively small for most variables (differences in RMSEs were 0–20%, depending on variable). However, the estimates based on stereo-photogrammetrically oriented aerial data retained better the original variation of the forest variables present in the study area.
Since the 1990’s, forest resource maps and small area estimates have been produced by combining national forest inventory (NFI) field plot data, optical satellite images and numerical map data using a non-parametric k-nearest neighbour method. In Finland, thematic maps of forest variables have been produced by the means of multi-source NFI (MS-NFI) for eight to ten times depending on the geographical area, but the resulting time series have not been systematically utilized. The objective of this study was to explore the possibilities of the time series for monitoring the key ecosystem condition indicators for forests. To this end, a contextual Mann-Kendall (CMK) test was applied to detect trends in time-series of two decades of thematic maps. The usefulness of the observed trends may depend both on the scale of the phenomena themselves and the uncertainties involved in the maps. Thus, several spatial scales were tested: the MS-NFI maps at 16 × 16 m2 pixel size and units of 240 × 240 m2, 1200 × 1200 m2 and 12 000 × 12 000 m2 aggregated from the MS-NFI map data. The CMK test detected areas of significant increasing trends of mean volume on both study sites and at various unit sizes except for the original thematic map pixel size. For other variables such as the mean volume of tree species groups, the proportion of broadleaved tree species and the stand age, significant trends were mostly found only for the largest unit size, 12 000 × 12 000 m2. The multiple testing corrections decreased the amount of significant p-values from the CMK test strongly. The study showed that significant trends can be detected enabling indicators of ecosystem services to be monitored from a time-series of satellite image-based thematic forest maps.
The tree stem volume models of Norway spruce, Scots pine and silver and downy birch currently used in Finland are based on data collected during 1968–1972. These models include four different formulations of a volume model, with three different combinations of independent variables: 1) diameter at height of 1.3 m above ground (dbh), 2) dbh and tree height (h) and 3) dbh, h and upper diameter at height of 6 m (d6). In recent National Forest Inventories of Finland, a difference in the mean volume prediction between the models with and without the upper diameter as predictor has been observed. To analyze the causes of this difference, terrestrial laser scanning (TLS) was used to acquire a large dataset in Finland during 2017–2018. Field-measured predictors and volumes predicted using spline functions fitted to the TLS data were used to re-calibrate the current volume models. The trunk form is different in these two datasets. The form height is larger in the new data for all diameter classes, which indicates that the tree trunks are more slender than they used to be. One probable reason for this change is the increase in stand densities, which is at least partly due to changed forest management. In models with both dbh and h as predictors, the volume is smaller a given h class in the data new data than in the old data, and vice versa for the diameter classes. The differences between the old and new models were largest with pine and smallest with birch.
The aim of this study was to estimate economic losses, which are caused by forest inventory errors of tree species proportions and site types. Our study data consisted of ground truth data and four sets of erroneous tree species proportions. They reflect the accuracy of tree species proportions in four remote sensing data sets, namely 1) airborne laser scanning (ALS) with 2D aerial image, 2) 2D aerial image, 3) 3D and 2D aerial image data together and 4) satellite data. Furthermore, our study data consisted of one simulated site type data set. We used the erroneous tree species proportions to optimise the timing of forest harvests and compared that to the true optimum obtained with ground truth data. According to the results, the mean losses of Net Present Value (NPV) because of erroneous tree species proportions at an interest rate of 3% varied from 124.4 € ha–1 to 167.7 € ha–1. The smallest losses were observed using tree species proportions predicted using ALS data and largest using satellite data. In those stands, respectively, in which tree species proportion errors actually caused economic losses, they were 468 € ha–1 on average with tree species proportions based on ALS data. In turn, site type errors caused only small losses. Based on this study, accurate tree species identification seems to be very important with respect to operational forest inventory.
Airborne laser scanning (ALS) has been the main method for acquiring data for forest management planning in Finland and Norway in the last decade. Recently, digital aerial photogrammetry (DAP) has provided an interesting alternative, as the accuracy of stand-based estimates has been quite close to that of ALS while the costs are markedly smaller. Thus, it is important to know if the better accuracy of ALS is worth the higher costs for forest owners. In many recent studies, the value of forest inventory information in the harvest scheduling has been examined, for instance through cost-plus-loss analysis. Cost-plus-loss means that the quality of the data is accounted for in monetary terms through calculating the losses due to errors in the data in the forest management planning context. These costs are added to the inventory costs. In the current study, we compared the losses of ALS and DAP at plot level. According to the results, the data produced using DAP are as good as data produced using ALS from a decision making point of view, even though ALS is slightly more accurate. ALS is better than DAP only if the data will be used for more than 15 years before acquiring new data, and even then the difference is quite small. Thus, the increased errors in DAP do not significantly affect the results from a decision making point of view, and ALS and DAP data can be equally well recommended to the forest owners for management planning. The decision of which data to acquire, can thus be made based on the availability of the data on first hand and the costs of acquiring it on the second hand.
Optical 2D remote sensing techniques such as aerial photographing and satellite imaging have been used in forest inventory for a long time. During the last 15 years, airborne laser scanning (ALS) has been adopted in many countries for the estimation of forest attributes at stand and sub-stand levels. Compared to optical remote sensing data sources, ALS data are particularly well-suited for the estimation of forest attributes related to the physical dimensions of trees due to its 3D information. Similar to ALS, it is possible to derive a 3D forest canopy model based on aerial imagery using digital aerial photogrammetry. In this study, we compared the accuracy and spatial characteristics of 2D satellite and aerial imagery as well as 3D ALS and photogrammetric remote sensing data in the estimation of forest inventory variables using k-NN imputation and 2469 National Forest Inventory (NFI) sample plots in a study area covering approximately 5800 km2. Both 2D data were very close to each other in terms of accuracy, as were both the 3D materials. On the other hand, the difference between the 2D and 3D materials was very clear. The 3D data produce a map where the hotspots of volume, for instance, are much clearer than with 2D remote sensing imagery. The spatial correlation in the map produced with 2D data shows a lower short-range correlation, but the correlations approach the same level after 200 meters. The difference may be of importance, for instance, when analyzing the efficiency of different sampling designs and when estimating harvesting potential.