Modelling dominant height growth from national forest inventory individual tree data with short time series and large age errors

We developed dominant height growth models for Norway spruce (Picea abies (L.) Karst.) and Scots pine (Pinus sylvestris L.) in Norway using national forest inventory (NFI) data. The data were collected for a different purpose which potentially causes problems for dominant height growth modelling due to short time series and large age errors. We used the generalized algebraic difference approach and fitted 15 different models using nested regression techniques. Despite the potential problems of NFI data the models fitted to these data were unbiased for most of the age and site index range covered by the NFI data when tested against independent data from long-term experiments (LTE). Biased predictions for young stands and better site indices that are better represented in the LTE data, led us to fit models to a combined data set for unbiased predictions across the total data range. The models fitted to the combined data that were unbiased with little residual variation when tested against an independent data set based on stem analysis of 73 sample trees from southeastern Norway. No indications of regional differences in dominant height growth across Norway were detected. We tested whether the better growing conditions during the short time series (22 years) of the NFI data had affected our dominant height growth models relative to long-term growing conditions, but found only minor bias. The combination with LTE data that have been collected during a longer period (91 years) reduced this potential bias. The dominant height growth models presented here can be used as potential height growth models in individual tree-based forest growth models or as site index models.     

Potential productivity of forested areas based on a biophysical model. A case study of a mountainous region in northern Spain

? Today’s forest managers face a number of important challenges involving an increasing need for precise estimates of forest structure and biomass, potential productivity or forest growth. The objective is to develop a model for potential productivity in a mountainous region of Spain. The model combines climatic, topographic and lithological data using a variant of a traditional biophysical model: the Paterson index.

? In a first approach, the climatic productivity is assessed by modelling the required parameters using different geostatistical techniques and software supported by GIS. A second approach includes the correction of the former productivity classes considering the different lithological facies. The potential forest productivity model involves the integration of both models.

? Finally, data from the National Forest Inventory (NFI) are used to compare the real and potential yield data within different regions of the studied area.

? The results of these analyses demonstrate the usefulness of the model, particularly in mountainous regions, where no significant differences are found between the data from the NFI and the model, but they also show the discrepancies between the estimates and real data when the latter are considered for different tree species, diameter classes or management.

     

Predicting the need for tending of conifer seedling stands in southern Finland

Assessing the need for and timing of tending in seedling stands is based on the stem numbers and heights of crop trees and competing broadleaves, as well as the expected forthcoming development of stand. The assessment is partly an outcome of field worker’s opinion and experience. The tending need of Norway spruce- and Scots pine-dominated seedling stands was modeled using the National Forest Inventory (NFI) data from southern Finland. The models predict the probabilities that the NFI field team leader’s proposal falls in the following four categories: tending is late, during the first or second 5-year period or no need for tending. The predictors such as stem numbers, tree heights, site fertility, regeneration method and accomplished tending logically explained the tending need. The overall accuracy of the models was only fair: 54% (kappa 0.27) for spruce and 55% (kappa 0.33) for pine. However, about 95% of the stands needing immediate tending were classified as stands needing immediate or first 5-year period tending. The surveyor-specific random effects were statistically significant, and the surveyors were likely to propose tending similarly in spruce and pine stands. The models can be utilized in forest planning systems and practical forest inventory.     

Estimating forest data for analyses of forest production and utilization possibilities at local level by means of multi-source National Forest Inventory

The sample plot data of National Forest Inventories (NFI) are widely used in the analysis of forest production and utilization possibilities to support national and regional forest policy. However, there is an increasing interest for similar impact and scenario analyses for strategic planning at the local level. As the fairly sparse network of field plots only provides calculations for large areas, satellite image data have been applied to produce forest information for smaller areas. The aim of this study was to test the feasibility of generating forest data for a Finnish forest analysis tool, the MELA system, by means of the Landsat satellite imagery and the NFI sample plot data. The study was part of the preparation of a local forestry programme, where a strategic scenario analysis for the forest area of two villages (ca 8000 ha) was carried out. Management units that approximate forest stands were delineated by image segmentation. Stand volume and other parameters for each forest segment were estimated from weighted means of the NFI sample plots, where the individual sample plot weights were estimated by the k nearest neighbour (kNN) method. Two different spectral features were tested: single pixel values and average pixel values within a segment. The estimated forest data were compared with the forest data based on independent stand-level field assessments in two subareas, a national park and an area of forest managed for timber production.In the national park, the estimated mean volume of the growing stock from both spectral feature sets (about 160 m³ ha⁻¹) was clearly lower than that obtained from stand-level field assessment (186 m³ ha⁻¹). Using average pixel values within a segment resulted in a higher proportion of pine and a lower proportion of spruce volume than using single pixel values. It also resulted in an estimated felling potential nearly 10% higher over the first 10-year period in the scenario analysis of the area dedicated to timber production. However, the maximum long-term sustainable removal was at the same level (about 30,000 m³ year⁻¹) for both feature sets over the simulated 30-year period. The resulting annual felling area in the first 10-year period was 12% lower when the segment averages were applied, but the difference subsequently levelled off. The kNN approach in estimating initial forest data for scenario analyses at the local level was found promising.     

Individual-tree diameter growth model for sugar maple trees in uneven-aged northern hardwood stands under selection system

An individual-tree diameter model was developed for sugar maple (Acer saccharum Marsh.) in northern hardwood stands managed under selection system. We fitted long-term remeasurement data to a linear mixed model to account for the temporal autocorrelation of the remeasurements. The model was evaluated using independent data from two physiographic regions and representing a range of tree diameter classes, residual basal areas and years since cut. We compared our model to several individual-tree models based on data from stands with varied management histories. Several competition indices were also tested for an improvement in model fitting and prediction. Our model had lower bias and prediction error when compared to two previous models, as it better accounted for the increased diameter growth that occurred in trees from appropriately managed stands. The addition of a tree-specific competition index failed to improve model fit and predictive ability over stand-level basal area.     

Comparison between TanDEM-X- and ALS-based estimation of aboveground biomass and tree height in boreal forests

Interferometric Synthetic Aperture Radar (InSAR) data from TerraSAR-X add-on for Digital Elevation Measurement (TanDEM-X) were used to estimate aboveground biomass (AGB) and tree height with linear regression models. These were compared to models based on airborne laser scanning (ALS) data at two Swedish boreal forest test sites, Krycklan (64°N19°E) and Remningstorp (58°N13°E). The predictions were validated using field data at the stand-level (0.5–26.1 ha) and at the plot-level (10 m radius). Additionally, the ALS metrics percentile 99 (p99) and vegetation ratio, commonly used to estimate AGB and tree height, were estimated in order to investigate the feasibility of replacing ALS data with TanDEM-X InSAR data. Both AGB and tree height could be estimated with about the same accuracy at the stand-level from both TanDEM-X- and ALS-based data. The AGB was estimated with 17.2% and 14.6% root mean square error (RMSE) and the tree height with 7.6% and 4.1% RMSE from TanDEM-X data at the stand-level at the two test sites Krycklan and Remningstorp. The Pearson correlation coefficients between the TanDEM-X height and the ALS height p99 were r?=?.98 and r?=?.95 at the two test sites. The TanDEM-X height contains information related to both tree height and forest density, which was validated from several estimation models.     

Mean species cover: a harmonized indicator of shrub cover for forest inventories

Because shrub cover is related to many forest ecosystem functions, it is one of the most relevant variables for describing these communities. Nevertheless, a harmonized indicator of shrub cover for large-scale reporting is lacking. The aims of the study were threefold: to define a shrub indicator that can be used by European countries for harmonized shrub cover estimation using data from their respective national forest inventories (NFIs); to quantify the effects of using different NFI field cover scales; and to establish bridges to facilitate harmonized estimation. Data for shrub species cover from the Third Spanish NFI together with scales for cover assessment from 16 European NFIs were used. The indicator, mean species cover (MSC), was defined for each species and each European forest category. Estimates of MSC calculated using species covers recorded for field plots, with 1% interval widths (MSCobs), were compared with the MSC values that would be obtained for the same data with the different European cover scales (MSCpred). Residuals calculated as differences between MSCobs and MSCpred were analyzed, and a linear mixed model was used as bridging function to adjust predictions and thus further harmonize estimates. Scales with only two or three intervals produced the greatest residuals, while all the other analyzed scales had residuals less than 5%. Most scales, except those most similar to Braun-Blanquet, displayed a tendency to be unreliable for larger covers. The proposed mean species cover indicator provides comparable estimates for shrub communities at large scales. The linear models improved the harmonization of MSC for the scales having two and three intervals.     

Propagating the errors of initial forest variables through stand- and tree-level growth simulators

Developments in the field of remote sensing have led to various cost-efficient forest inventory methods at different levels of detail. Remote-sensing techniques such as airborne laser scanning (ALS) and digital photogrammetry are becoming feasible alternatives for providing data for forest planning. Forest-planning systems are used to determine the future harvests and silvicultural operations. Input data errors affect the forest growth projections and these effects are dependent on the magnitude of the error. Our objective in this study was to determine how the errors typical to different inventory methods affect forest growth projections at individual stand level during a planning period of 30 years. Another objective was to examine how the errors in input data behave when different types of growth simulators are used. The inventory methods we compared in this study were stand-wise field inventory and single-tree ALS. To study the differences between growth models, we compared two forest simulators consisting of either distance-independent tree-level models or stand-level models. The data in this study covered a 2,000-ha forest area in southern Finland, including 240 sample plots with individually measured trees. The analysis was conducted with Monte Carlo simulations. The results show that the tree-level simulator is less sensitive to errors in the input data and that by using single-tree ALS data, more precise growth projections can be obtained than using stand-wise field inventory data.     

Combining ALS and NFI training data for forest management planning: a case study in Kuortane,Western Finland

Forest inventories based on airborne laser scanning (ALS) have already become common practice in the Nordic countries. One possibility for improving their cost effectiveness is to use existing field data sets as training data. One alternative in Finland would be the use of National Forest Inventory (NFI) sample plots, which are truncated angle count (relascope) plots. This possibility is tested here by using a training data set based on measurements similar to the Finnish NFI. Tree species-specific stand attributes were predicted by the non-parametric k most similar neighbour (k-MSN) approach, utilising both ALS and aerial photograph data. The stand attributes considered were volume, basal area, stem number, mean age of the tree stock, diameter and height of the basal area median tree, determined separately for Scots pine, Norway spruce and deciduous trees. The results obtained were compared with those obtained when using training data based on observations from fixed area plots with the same centre point location as the NFI plots. The results indicated that the accuracy of the estimates of stand attributes derived by using NFI training data was close to that of the fixed area plot training data but that the NFI sampling scheme and the georeferencing of the plots can cause problems in practical applications.     

河北省主要树种单木和林分生长率模型研建

利用第六次至第九次全国森林资源清查河北省2001,2006,2011,2016年4个年度的固定样地调查数据,采用非线性回归估计方法,建立了18个树种组的单木胸径生长率和材积生长率模型,以及12个树种组的林分材积生长率模型。结果表明,单木生长率模型的平均预估误差(MPE)基本都在3%以内,而平均百分标准误差(MPSE)、胸径生长率模型大都在10%以内,材积生长率模型大都在20%左右;林分生长率模型的平均预估误差(MPE)基本都在5%以内,平均百分标准误差(MPSE)大都在25%以内。所建模型可为河北省开展森林资源年度更新提供技术支撑。     

Integration of forest inventory and satellite imagery:: a Canadian status assessment and research issues

Canada's ability to sustainably manage approximately 10% of the global forest cover is a critical environmental and economic issue. The capacity to meet such demands and to deliver on national and international commitments regarding forest management is enabled through collaboration between federal, provincial, and territorial agencies. A principal collaborator is the National Forest Inventory (NFI); a systematic photo-plot based monitoring system designed specifically for reporting purposes and as an important input for scientific models. Satellite imagery is illustrated here as a support data set to ensure the quality of the NFI, for auditing the photo-plot contents, and to detect spatial biases. The Canadian Forest Service, in collaboration with the Canadian Space Agency and other federal and provincial agencies, is producing a national land cover database of the forested area of Canada (Earth Observation for Sustainable Development of Forests (EOSD)) using Landsat-7 ETM+ data for circa 2000 conditions. The integration between the plot-based NFI with classified EOSD data is presented for central British Columbia, an area comprising 6 Landsat scenes and 324 2 km × 2 km photo-plots. Traditional accuracy assessment measures based on the analysis of coincidence matrices are reported as levels of agreement for hierarchically aggregated land cover categories (overall agreements of 91%, 79%, 64% and 26% for 3, 4, 6 and 20 classes respectively) to demonstrate coincidence between the different data products. Local agreement between NFI and EOSD is demonstrated as a means of photo-plot auditing while spatial biases are detected through investigations of geographic pattern in the coincidence values. The illustrated approaches may be expanded or applied to different mapped attributes (e.g., biomass) that are of utility to those attempting to characterize large areas in a consistent and rigorous fashion.     

Large-scale dynamics of a heterogeneous forest resource are driven jointly by geographically varying growth conditions, tree species composition and stand structure

? Context

Forest resource projections are required as part of an appropriate framework for sustainable forest management. Suitable large-scale projection models are usually based on national forest inventory (NFI) data. However, sound projections are difficult to make for heterogeneous resources as they vary greatly with respect to the factors that are assumed to drive forest dynamics on a large spatial scale, e.g. geographically varying growth conditions (here represented by NFI regions), tree species composition (here broadleaf-dominated, conifer-dominated and broadleaf-conifer mixed stands) and stand structure (here high forest, coppice forest and high-coppice forest mixture).

? Question and objective

Our question was how does the variance of forest dynamics parameters (i.e. growth, felling and mortality, and recruitment processes) and that of 20-year forest resource projections partition between these factors (NFI region, tree species composition and stand structure), including their interactions. Our objective was to capitalise on the suitability of an existing multi-strata, diameter class matrix model for the purposes of making projections for the highly heterogeneous French forest resource.

? Methods

The model was newly calibrated for the entire territory of metropolitan France based on most recent NFI data, i.e. for years 2006?C2008. The forest resource was divided into strata by crossing the factors NFI region, tree species composition and stand structure. The variance partitioning of the parameters and projections was assessed based on a model sensitivity analysis.

? Results

Growth, felling and mortality varied mainly with NFI region and species composition. Recruitment varied mainly with NFI region and stand structure. All three factors caused variations in resource projections, but with unequal intensities. Factor impacts included first order and interaction effects.

? Conclusions

We found, by considering both first order and interaction effects, that NFI region, species composition and stand structure are ecologically relevant factors that jointly drive the dynamics of a heterogeneous forest resource. Their impacts, in our study, varied depending on the forest dynamics process under consideration. Recruitment would appear to have a particularly great impact on resource changes over time.     

Assessment of bias due to random measurement errors in stem volume growth estimation by the Swedish National Forest Inventory

Abstract

We evaluated the performance of two methods for estimating stem volume increment at individual tree level with respect to bias due to random measurement errors. Here, growth is either predicted as the difference between two consecutive volume estimates where single-tree volume functions are applied to data from repeated measurements or by a regression model that is applied to data from a single survey and includes radial increment. In national forest inventories (NFIs), the first method is typically used for permanent plots, the second for temporary plots. The Swedish NFI combines estimates from both plot types to assess growth at national and regional scales and it is, therefore, important that the two methods provide similar results. The accuracy of these estimates is affected by random measurement errors in the independent variables, which may lead to systematic errors in predicted variables due to model non-linearity. Using Taylor series expansion and empirical data from the Swedish NFI we compared the expected bias in stem volume growth estimates for different diameter classes of Scots pine (Pinus sylvestris L.) and Norway spruce (Picea abies (L.) Karst.). Our results indicate that both methods are fairly insensitive to random measurement errors of the size that occur in the Swedish NFI. The empirical comparison between the two methods showed greater differences for large diameter trees of both pine and spruce. A likely explanation is that the regressions are uncertain because few large trees were available for developing the models.     

Spruce diameter growth in young mixed stands of Norway spruce (Picea abies (L.) Karst.) and birch (Betula pendula Roth B. pubescens Ehrh.)

A stand-based model for predicting basal-area mean diameter growth for Norway spruce (Picea abies (L.) Karst.) in young mixed stands of spruce and birch (Betula pendula Roth, B. pubescens Ehrh.) was developed and compared with two existing growth models developed for older stands. The main data were from experiments with four different pre-commercial thinning regimes. A multiplicative model with four independent variables was found suitable. The independent variables were total number of trees per hectare of all the species, site index, dominant height of spruce, and a measure of competition between birch and spruce, i.e. dominant height of spruce divided by the dominant height of birch multiplied by the proportion of spruce of total number of trees. The R² value was 0.59 and the coefficient of variation was 12%. A test with an independent data set from the National Forest Inventory (NFI) indicated that the function developed in this study is suitable for young stands at medium to highly productive areas. Large deviations between observed and predicted growth for the two existing functions were revealed in highly productive stands. The tests based on data from the NFI also indicated that the existing function developed for spruce in older mixed stands is suitable for practical purposes for young stands.     

Comparison of treewise and standwise forest simulators by means of quantile regression

Predicting forest development under varying treatment schedules forms the basis of forest management planning. The actual growth predictions are made with a forest simulator which includes growth equations and additional models for predicting a number of varying tree, forest and site properties. Forest growth simulators typically include either tree-level or stand-level growth models, but these two approaches have not been thoroughly compared. We set out here to compare these two approaches with the SIMO simulator framework in a small data set from southern Finland based on 60 sample plots in 30 stands, the development of which was known for 20 years. The stands chosen were very dense, so that the simulators could be tested under extreme conditions. The results show that the stand-level model is more accurate in almost all cases and its computational burden is much lower. It could therefore be advisable to use tree-level models for short-term predictions, which would ensure detailed information on forest structure for planning the near-future operations. Stand-level models would be more advisable in longer term predictions, especially when accurate volume estimates are considered more important than the forest structure. The errors observed in these simulators were analysed further by quantile regression, which allows empirical estimates of confidence intervals to be obtained for the simulator.     

Estimating annual GPP,NPP and stem growth in Finland using summary models

In this study we introduce and test a new simple approach for estimating annual stand-level gross primary production (GPP), net primary production (NPP) and stem biomass growth based on carbon acquisition and allocation, by combining existing summary models. The focus is on the variation of GPP and NPP across different parts of Finland caused by climate.     

Using multi-source data to map and model the predisposition of forests to wind disturbance

Natural disturbances such as wind are known to cause threats to ecosystem services as well as sustainable forest ecosystem management. The objective of this research was to better understand and quantify drivers of predisposition to wind disturbance, and to model and map the probability of wind-induced forest disturbances (P_DIS) in order to support forest management planning. To accomplish this, we used open-access airborne light detection and ranging (LiDAR) data as well as multi-source National Forest Inventory (NFI) data to model P_DIS in southern Finland. A strong winter storm occurred in the study area in December 2011. High spatial resolution aerial images, acquired after the disturbance event, were used as reference data. Potential drivers associated with P_DIS were examined using a multivariate logistic regression model. The model based on LiDAR provided good agreement with detected areas susceptible to wind disturbance (73%); however, when LiDAR was combined with multi-source NFI data, the results were more promising: prediction accuracy increased to 81%. The strongest predictors in the model were mean canopy height, mean elevation, and stem volume of the main tree species (Norway spruce and Scots pine). Our results indicate that open-access LiDAR data can be used to model and map the probability of predisposition to wind disturbance, providing spatially detailed, valuable information for planning and mitigation purposes.     

A variance-covariance structure to take into account repeated measurements and heteroscedasticity in growth modeling

This study proposes a within-subject variance-covariance (VC) structure to take into account repeated measurements and heteroscedasticity in a context of growth modeling. The VC structure integrates a variance function and a continuous autoregressive covariance structure. It was tested on a nonlinear growth model parameterized with data from permanent sample plots. Using a stand-level approach, basal area growth was independently modeled for red spruce (Picea rubens Sarg.) and balsam fir [Abies balsamea (L.) Mill.] in mixed stands. For both species, the implementation of the VC structure significantly improved the maximum likelihood of the model. In both cases, it efficiently accounted for heteroscedasticity and autocorrelation, since the normalized residuals no longer exhibited departures from the assumptions of independent error terms with homogeneous variances. Moreover, compared with traditional nonlinear least squares (NLS) models, models parameterized with this VC structure may generate more accurate predictions when prior information is available. This case study demonstrates that the implementation of a VC structure may provide parameter estimates that are consistent with asymptotically unbiased variances in a context of nonlinear growth modeling using a stand-level approach. Since the variances are no longer biased, the hypothesis tests performed on the estimates are valid when the number of observations is large.     

Developing an airborne laser scanning dominant height model from a countrywide scanning survey and national forest inventory data

Abstract

An airborne laser scanning (ALS) dominant height model was developed based on data from a national scanning survey with the aim of developing a digital terrain model (DTM) for Denmark. Data obtained in the ongoing Danish national forest inventory (NFI) were used as reference data. The data comprised a total of 2072 measurements of dominant height on NFI sample plots inventoried in 2006–2007 and their corresponding ALS data. The dominant height model included four variables derived from the ALS point cloud distribution. The variables were related to canopy height, canopy density and species composition on individual plots. The RMSE of the final model was 2.25 m and the model explained 93.9% of the variation (R ²). The model was successful in predicting dominant height across a wide range of forest tree species, stand heights, stand densities, canopy cover and growing conditions. The study demonstrated how low-density ALS data obtained in a survey not specifically aimed at forest applications may be used for obtaining biophysical forest properties such as dominant height, thereby reducing the overall forest inventory costs.