Fixed-area Plots Research Articles

Area-based lidar-assisted estimation of forest standing volume

Airborne laser scanning (lidar) technology is increasingly being applied in forest ecosystem surveys. This research note proposes a design-based approach for the lidar-assisted estimation of forest standing volume when ground surveys are performed by means of fixed-area plots. The lidar measurement of the height of the upper canopy (digital crown model) is performed for the whole study area, and the resulting pixel heights are adopted as auxiliary information to couple with the standing volume acquired on the ground by means of sample plots. The ratio estimator for the total volume of the forest is derived in a complete design-based framework together with an unbiased estimator of its sampling variance and the corresponding confidence interval. The proposed procedure has been tested in Bosco della Fontana, a lowland forest in Northern Italy, obtaining a 95% confidence interval for the total volume, which is approximately 2/3 smaller than that obtained by solely using information arising from field plots.

Comparing Double-Sampling Efficiency Using Various Estimators with Fixed-Area and Point Sampling

Abstract In this study, we compare the efficiency of double sampling using point sampling and fixed-area plots for sawtimber volume estimates in a mixed-hardwood, oak-dominated stand. Multiple sample sizes and combinations were evaluated to determine optimum double-sample ratios. Results indicatedthat double, point-sampling schemes are more efficient in terms of field time and sampling errors than double-sampling schemes incorporating fixed-area plots. Data suggested that the most efficient ratio of measured and nonmeasured points with double sampling varies on the basis of the nonmeasuredvariable used and desired SE percentage levels for the inventory.

Effects of competitor spacing in a new class of individual-tree indices of competition: semi-distance-independent indices computed for Bitterlich versus fixed-area plots

We illustrate effects of competitor spacing for a new class of individual-tree indices of competition that we call semi-distance-independent. This new class is similar to the class of distance-independent indices except that the index is computed independently at each subsampling plot surrounding a subject tree for which growth is to be modelled. We derive the effects of distance for this class as the expected value over independent samples containing a particular subject tree. In a previous paper, we illustrated distance effects implicit in eight indices of the distance-dependent class. Here, we present distance effects of four semi-distance-independent indices: density, sum of diameters, basal area, and tree-area ratio; each determined for small fixed-area plots of 0.04 ha and for Bitterlich samples of 6 m2·ha–1. We show that several members of this new class have distance effects very similar to the distance-dependent class and should, therefore, be equally effective in accounting for competitive effects in individual-tree increment models. The comparisons should inform selection of competition indices and sampling designs for growth modelling.

A comparison of five sampling techniques to estimate surface fuel loading in montane forests

Designing a fuel-sampling program that accurately and efficiently assesses fuel load at relevant spatial scales requires knowledge of each sample method’s strengths and weaknesses. We obtained loading values for six fuel components using five fuel load sampling techniques at five locations in western Montana, USA. The techniques included fixed-area plots, planar intersect, photoloads, a photoload macroplot, and a photo series. For each of the six fuels, we compared (1) the relative differences in load values among techniques and (2) the differences in load between each method and a reference sample. Totals from each method were rated for how much they deviated from totals for the reference in each fuel category. The planar-intersect method, which used 2.50 km of transects, was rated best overall for assessing the six fuels. Bootstrapping showed that at least 1.50 km of transect were needed to obtain estimates that approximate the reference sample. A newly developed photoload method, which compared fuel conditions on the forest floor with sets of pictures calibrated for load by fuel type, compared well with the reference and planar intersect. The commonly used photo series consistently produced higher mean load estimates than any other method for total fine woody debris (0.05–0.20 kg m–2) and logs (0.50–1.25 kg m–2).

An assessment of sample-based estimators of tree species richness in two wet tropical forest compartments in Panama and India

SuMMARY We assessed the performance of ten incidence-based estimators of tree species richness in simulated simple random sampling with fixed-area plots. Stem diameter-limited tree species and location data came from two species-rich wet tropical forest compartments in Panama and India. Lower limits of stem diameter were 1 cm and 30 cm, respectively. Estimators varied widely in their estimates of richness and their rankings changed frequently across sites and sample designs. A gammaPoisson estimator was overall best according to a performance score of three accuracy statistics and sample size. However, until corroborated by further studies, Chao’s 1981 non-parametric estimator is recommended for forest inventories with fixed-area plots.

Computer aided inventory analysis for sustainable management of non-timber forest product resources

Non-Timber Forest Products (NTFPs) consist of variety of useful products such as bark, fruits, seeds, leaves, gums and resins, tannins, dyes, roots etc. These resources are renewable if sustainably managed. Quantitative inventory of NTFP resources will in aid formulating sustainable NTFP management practices. Inventory of NTFP resources in the tropics is relatively new and has received little formal study. In this paper, we outline the methodology for inventorying NTFP resources highlighting various nested elements involved in the multi-resource NTFP inventory, possible sampling procedures, methods for the estimation of density and diversity measures and quantification of useful products. The details of the software ‘InventNTFP’ developed for carrying out inventory analysis are also presented. The software handles large number of species simultaneously and takes into consideration different aspects of inventory and reduces time greatly in carrying out analysis and generating results. It produces density and diversity measures and extrapolates availability of useful products from various species following non-destructive approach. Although the software is originally conceived for the analysis of NTFP inventory, it may be found useful in other areas involving fixed area plot based surveys for working out various density and diversity measures. Such application domain includes inventory of selected floral and faunal groups.

A new empirical approach for estimation in k-tree sampling

Point-to-tree distance techniques (also known as k-tree or fixed count sampling) are a practical field sampling method for forest inventories and ecological surveys, but statistical estimation remains a problem. Various approximation techniques have been proposed, partly on an empirical basis, partly under model assumptions about the underlying point processes. In this paper we present a brief history of k-tree sampling in forestry and introduce a new and simple empirical approximation method. This method builds on calculating the variable sample plot area from the distance not only to the k tree but averaging the distance to the k and to the ( k + 1) tree. Simulation studies were done on a tree map from the Miombo woodlands in northern Zambia and five artificially generated tree maps with different point patterns. In terms of bias, our techniques are superior to Prodan's approach; however, Eberhardt's estimator produces consistently the smallest bias for the studied clustered patterns, and is unbiased for random patterns. For the two uniform patterns studied, bias was smallest on average for our new estimators. A comparison with relascope sampling and fixed area sample plots sized to yield k trees per sample plot indicated that the k-tree sampling estimators studies produced higher standard errors for the estimation of basal area and number of stems than fixed area plots for the clustered and random maps. For the two uniform patterns, k-tree sampling with our mean-area estimator produces smallest errors, in particular for small values of k. It is concluded that it is mainly the simplicity of the method that makes point-to-tree distance sampling interesting for forest and ecological sampling, but its statistical performance remains somewhat uncertain.

Changes in tropical forest vegetation along an altitudinal gradient in the Udzungwa Mountains National Park, Tanzania

AbstractThis study describes changes in woody vegetation in the Mwanihana forest, Udzungwa Mountains National Park, Tanzania, over an altitude range of 470–1700 m. Two methods, fixed‐ and variable‐area plots, are compared to elucidate altitudinal variation in tropical forest structure, diversity and community composition. Six 25 m × 100 m fixed area plots recorded a total of 2143 woody stems of ≥3 cm d.b.h. from 204 species. The 78 variable‐area plots recorded the nearest twenty trees of ≥20 cm d.b.h. to an objectively chosen point, giving a total of 1560 stems in 9.1 ha from 156 species. A linear trend of increasing stem density with altitude was seen for variable‐area plots. Species diversity is highest at high elevations. There was no clear zonation of elevational vegetation types. Restricted range taxa occur at all altitudes sampled. The study also revealed some methodological considerations. Bias in sample size and plot area can be tested by employing two sampling methods. Of the two methods used, fixed area plots are preferred as variable area plots are impractical in tangled understorey. Plot size must be controlled for in order to make reliable observations of diversity. Sampling along a continuous or near‐continuous altitudinal gradient with sufficient replication is also important.

Residual Stand Damage from Cut-to-Length Thinning of a Mixed Conifer Stand in Northern Idaho

Abstract We collected residual stand-damage data from a mixed conifer stand in northern Idaho that had been commercially thinned with a cut-to-length harvesting system. The stand composition after harvesting was 76% grand fir (Abies grandis); 14% Douglas-fir (Pseudotsuga menziesii var. glauca); 5% western redcedar (Thuja plicata); and 5% lodgepole pine (Pinus contorta), western white pine (Pinus monticola), and ponderosa pine (Pinus ponderosa). For all crop trees, damage to the bole, roots, and crown was assessed using systematic sampling with a random start and fixed area plots. Wounding occurred on 37.4% of the remaining trees, but the severity of wounding varied significantly by species (P < 0.05). Eighty-four percent of wounding for all species combined was considered as small size (<194 cm2). The greatest average amount of damage to a bole occurred along the first 2 m up from the ground (67.2%) and also within 4 m of the forwarder centerline (67.7%). Gouges were present on 41% of all scars. Tree location to forwarder trail appears to have a significant effect on the number and height of scars on a tree (P < 0.05). We estimated that throughout a 20-year period, volume losses for grand fir because of decay would be 2.57% compared to 1.31% in an undamaged stand of similar composition. West. J. Appl. For. 21(3): 142–148.

Structural dynamics of riparian forests along a black cottonwood successional gradient

Black cottonwood Populus balsamifera L. subsp. trichocarpa (Brayshaw) dominated riparian forests are important habitats for organisms in both terrestrial and aquatic ecosystems in the Pacific Northwest. Using a chronosequence approach and fixed area plots, successional processes were quantified in 28 stands along 145 km of the Willamette River in Western Oregon. Cluster analysis yielded five successional stages; stand initiation, stem exclusion, early, mid, and late seral. Ordination, using non-metric multidimensional scaling, revealed stand variables were strongly associated with these groupings, e.g., stand age, cottonwood dbh, tree biomass, basal area, structural diversity. Tree densities decreased through time from a high of 96,200 to 410 trees/ha in late seral stands. Structural diversity increased through time as understory plant and late-successional tree species established in early seral stands, creating multiple vegetative layers. Aboveground standing tree biomass increased through time from <1 Mg/ha in stand initiation sites to as high as 594 Mg/ha in late seral stands. Grouping using early-successional tree species importance values revealed lower tree biomass in stands dominated by late-successional tree species when compared to late seral stands still dominated by cottonwood. Abundance of invasive plant species, e.g., reed canary grass Phalaris arundinacea L., appeared to negatively impact structural diversity and biomass accumulation by inhibiting understory re-initiation. A significant reduction in peak flow events in the last century may help explain the lack of riparian forest regeneration we noted, as germination and establishment of young cottonwoods was limited to low areas subject to annual flood inundation and scouring. This study reveals Willamette River black cottonwood-dominated forests as dynamic habitats consisting of a mosaic of diverse stands in varying successional stages. It is vital that we find ways to manage for regeneration of these forests as well as implementing control practices for invasive plant species.

Simulating spatial distributions of forest trees by using data from fixed area plots

Summary Cost is a critical factor for managing ecosystems. Common forest inventories are usually carried out at regular time intervals (e.g. every 10 years) and are the basis for management planning. This study investigated the potential of utilizing the data of common forest inventories for simulating the spatial distribution of forest tree locations. Fixed area plot sample data were taken from the University Forest of Pertouli, which is an uneven-aged Abies borisii regis Matf. forest. Using as criterion the index of dispersion, the tree spatial distribution was characterized as aggregated. The Neyman Type A distribution, a typical distribution of aggregated populations, was a good fi t to the data, while tree locations simulation was based on the Poisson cluster process. The simulations resulting from the application of fi xed area plot sampling do not incorporate information about the real distances between the trees, but they can describe adequately the spatial patterns of their locations in twodimensional space and seem to be useful tools for managers of forest ecosystems. For similar populations the detection of their aggregation seems not to be affected by a considerable decrease of the number (up to 36) and the size (up to 125 m 2 ) of sample fi xed-area plots. This method is cost effective and its use, potentially in combination with other methods, could be further investigated for its advantages.

Estimating forest canopy bulk density using six indirect methods

Canopy bulk density (CBD) is an important crown characteristic needed to predict crown fire spread, yet it is difficult to measure in the field. Presented here is a comprehensive research effort to evaluate six indirect sampling techniques for estimating CBD. As reference data, detailed crown fuel biomass measurements were taken on each tree within fixed-area plots located in five important conifers types in the western United States, using destructive sampling following a series of four sampling stages to measure the vertical and horizontal distribution of canopy biomass. The six ground-based indirect measurement techniques used these instruments: LI-COR LAI-2000, AccuPAR ceptometer, CID digital plant canopy imager, hemispherical photography, spherical densiometer, and point sampling. These techniques were compared with four aggregations of crown biomass to compute CBD: foliage only, foliage and small branchwood, foliage and all branchwood (no stems), and all canopy biomass components. Most techniques had the best performance when all canopy biomass components except stems were used. Performance dropped only slightly when the foliage and small branchwood canopy biomass aggregation (best approximates fuels available for crown fires) was employed. The LAI-2000, hemispherical photography, and CID plant canopy imager performed best. Regression equations that predict CBD from gap fraction are presented for all six techniques.

Population status of eucalypt trees on the River Murray floodplain, South Australia

AbstractFor populations to persist, recruitment must keep pace with mortality. In variable environments, opportunities for reproduction occur patchily in time and space, and favourable conditions must occur sufficiently often to allow growth to maturity (hence ‘recruitment’). The risks of local extinctions may be increased by anthropogenic factors. This scenario is illustrated by two tree species, river red gum (Eucalyptus camaldulensis) and black box (E. largiflorens), on the floodplain of the River Murray, South Australia. Fixed area plots were established at Banrock Station, where large, mature trees are common, although red gum outnumber black box by about four to one. Trunk diameter was measured as a surrogate for tree age. The smallest diameter (0–10 cm) black box are nearly as common as seedlings of that species, whereas the smallest diameter red gum (0–10 cm) are 10 times more abundant than seedlings. Small trees of both species occur in localized clumps, and the respective size‐class distributions exhibit series of peaks and falls, suggesting episodic recruitment and opportunistic survival. Population viability calculations suggest that more than 100% of existing saplings need to survive to maintain the local black box population (i.e. there are too few saplings). The red gum population apparently requires a smaller proportion of survivors, but the calculations may be biased by the clumped distribution of the saplings. Based on population structure and viability estimates, black box at this site appear to lack sufficient regeneration to compensate for adult mortality while red gum appear to have a much better balance. The methods established here may be useful for assessment of stands of these species in other areas. Copyright © 2005 John Wiley &amp; Sons, Ltd.

Forest community analysis and the point-centered quarter method

The point-centered-quarter (PCQ) method has been applied in community analysis since the publication of the method nearly 50 years ago. This and other distance methods offer increased sampling efficiency over fixed-area plots (FAP), but have long been known to produce biased density estimates when plant distribution deviates from random spatial patterns. Spatial indices have been developed to quantify the direction of this bias when plant distributions are aggregated or evenly distributed. Its continued use, especially in community analysis, requires additional scrutiny in measurements of community structure. We measured 14 forest stands of varying age, elevation and disturbance regime using FAP and PCQ methods. Density estimates were biased, with the point-centered quarter method lower than fixed-area plot estimates when stems were aggregated and higher when stems were evenly spaced. In general the PCQ method underestimated species richness. The efficiency of the PCQ method makes it popular for ordination studies, although comparison of community structure varied from 18% to 90% similarity between the measurements of species basal area in the same stands using the two different methods. The bias observed in calculations of stem density, species abundance and community similarity indicate that use of the PCQ method should be approached with caution when used in community level analysis.

Relationship between plot size and the variance of the density estimator in West African savannas

The relationship between the variance of the density estimator and the plot size in forest inventories based on fixed area plots was characterized in West African savannas. Nine sites ranging from dry to moist savanna were surveyed, and the variance of the density estimator was assessed for varying sample plot sizes. An approximate theoretical expression of the variance was derived, taking into account the uncertainty on plot limits. In six sites, a Matérn process was fitted to the spatial pattern of trees. The point process was used to generate spatial patterns, yielding simulated values of the variance. A power function was also fitted to observed and simulated data. The theoretical expression and simulations showed that the contribution of uncertain plot limits to the variability of the density estimator was negligible with respect to the contribution of the spatial pattern of trees. The theoretical expression matched the data for small areas, but was inaccurate for large areas. The power function provided better fits. Nevertheless, the theoretical expression did not require any statistical fit to data and established a clear link between the spatial pattern of trees and the variance of the density estimator, with interpretable parameters.

Area-Independent Sampling for Total Basal Area

Abstract An unbiased direct estimator of total basal area for a stand is presented; stand area need not be known. Sampling can use horizontal point samples or fixed-area plots distributed on a regular grid or randomly selected. The sample space extends beyond the stand boundary. The estimator for basal area makes no distinction between plots whose centers are interior to the stand boundary and plots whose centers are outside. Only trees within the stand boundary are tallied. For horizontal point sampling, the estimator is the product of the sample tree count, the basal area factor, and the area per grid point, with the latter being derived from grid spacing. This method avoids the complexities associated with most other estimators designed to minimize or eliminate edge bias. The proposed method may require more sample points to achieve the same level of precision as competing estimators. FOR. SCI. 50(4):512–517.

Analysis of sampling methods for coarse woody debris

Coarse woody debris (CWD) is an important structural component of forest ecosystems and plays a key role in ecosystem functioning. It includes whole fallen trees and branches, pieces of fragmented wood, stumps, and standing dead trees (snags). A number of sampling approaches have been used to quantify CWD but there has been little systematic analysis of alternative approaches in terms of sampling efficiency. This study analysed sampling methods for CWD in three Australian forest types: eucalypt-dominated woodlands in south central Queensland, eucalypt open forest (dry sclerophyll forest) in south coastal New South Wales and eucalypt tall open forest in southern Tasmania. All downed wood (greater than 10 cm in woodland and 15 cm in other forest types) was measured and mapped on several large plots (1 ha) at each site. Alternative sampling approaches were compared using a computer-based system. These included varying length and arrangement of lines for the line intersect method and varying sizes of fixed-area plots. Variability of CWD volume in forests is relatively high and achieving estimates with a precision (coefficient of variation) less than 50% of the mean requires a high sampling intensity. In analysis of the line intersect method, precision of volume estimates increased as total transect length increased. In denser forests, transects of at least 100 m are required to achieve a precision of less than 100% of the mean. A single long transect resulted in lower coefficient of variation than a number of short transects and is generally more cost effective. There was no improvement in precision by arranging the total transect length in an ‘L’ shape, a triangle or a square. In woodland vegetation (20–50% canopy cover) the amount of CWD is relatively low and variability is high. In these situations fixed-area plots are preferred. 20 m×20 m plots give estimates with a mean coefficient of variation less than 100% in the more dense forest structures but well over 100% in woodlands. Measuring all pieces on a 50 m×50 m plot in woodlands gave estimates with a precision of ±21–73% of the mean.

Forest inventory estimation with mapped plots

Procedures are developed for estimating means and variances with a mapped-plot design. The focus is on fixed-area plots, and simulations are used to validate the proposed estimators. The mapped-plot estimators for means and variances are compared with simple random sampling estimators that utilize only full plots. As expected, the mapped-plot estimates have smaller mean squared errors than the simple random sampling estimates. The theory for fixed-area plots is easy to apply, although additional work is required to map plots in the field. Corresponding theory for variable plots is developed but not tested with simulations. The difficulty of applying these methods to variable plots is greater, but not prohibitive.

The distribution and severity of beech bark disease in the Catskill Mountains, N.Y.

The distribution and severity of beech bark disease (BBD) in the Catskill Mountains of southeastern New York state, U.S.A., were measured between 1997 and 2000. Forest composition was measured using wedge prism surveys and fixed-area plot counts of canopy species. BBD severity on individual trees was ranked from 1 (no disease) to 5 (dead) based on bark health and canopy loss. These data were analyzed at multiple spatial scales to determine possible controls of disease distribution and severity. BBD was present on almost all Fagus grandifolia Ehrh. (American beech) individuals encountered &gt;10 cm diameter at breast height, but was most severe in larger diameter classes. Mortality due to BBD also increased with diameter class. A strong positive relationship was found between BBD severity and relative beech basal area at the tract (hundreds of hectares) and 100-m elevation band spatial scales, but not found at smaller scales. Successful long-range dispersal of the disease may therefore be dependent upon host density, while local conditions may control BBD severity within individual stands. The current status of BBD in the Catskills suggests these forests are entering the aftermath phase of the disease's progression and that BBD has become an endemic component of these forests.

Comparison of fixed-area plot designs for estimating stand characteristics and western spruce budworm damage in southwestern U.S.A. forests

Various sampling designs were evaluated using data on stand density, stocking, mortality, and top kill, as influenced by the western spruce budworm (Choristoneura occidentalis Freeman) in 17 stands in New Mexico and Colorado. Efficiency improved as plot size decreased from 0.04 to 0.02 ha for all variables and sampling designs, except for 0.01-ha plots, which required extremely large sample sizes and were subject to bias. Cluster designs were much more efficient than simple random sampling designs, allowing twice the reduction in sample size than was gained by relaxing the allowable error from 10 to 15%. Clusters of two plots were as precise as clusters of three plots. Of the four variables evaluated, density required the largest sample sizes, followed by stocking, percent mortality (for stands where mortality exceeded 10%), and top kill. Few plots were necessary to ascertain that mortality was less than 10%. On average, 10 pairs of 0.02-ha plots would estimate density, stocking, and mortality within a 10% allowable error. A field check of density and stocking variables is recommended, and additional samples are suggested in stands with large percent standard errors associated with those variables.