Application of Most Similar Neighbor Inference for Estimating Marked Stand Characteristics Using Harvester and Inventory Generated Stem Databases

Malinen, Jukka; Maltamo, Matti; Harstela, Pertti

doi:10.1080/14942119.2001.10702444

Cited by 25 publications

(28 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According Malinen et al (2001) the mean tree variables are the most important search variables, while the signifi cance of the other variables is small. The design attributes used were obtained diameters in percentages of 0% (the smallest diameter), 20%, 40%, 60%, 80% and 100% (the largest diameter) of accumulated basal area, a and b of Näslund's height parameters (Näslund 1937) and volume of tree species.…”

Section: Comparison Of Resultsmentioning

confidence: 99%

“…In several studies where the k-nn method has been used in forestry the suitable number of reference stands has been set between 3 and 10. In this study the size of the neighbourhood in the k-nn MSN and local k-nn MSN method was set to 5 neighbours to give the best performance within the RMSE criteria (Malinen et al 2001). A smaller neighbourhood would reduce the bias, but the RMSE would indicate weaker local validity of estimates.…”

Section: Resultsmentioning

confidence: 99%

“…Although non-parametric methods have been found to be useful tools in many forestry situations (Haara et al 1997, Maltamo and Kangas 1998, Tommola et al 1999, Malinen et al 2001, there are some disadvantages in the use of these methods. The question of the optimal shape of the selected bandwidth or the number of neighbours to use is diffi cult because the answer usually depends on the shape and location of the data in the input space.…”

Section: Introductionmentioning

confidence: 99%

“…For a situation where values or demand change, the method is not appropriate. Malinen et al (2001) developed an application of k-nn MSN-inference (Moeur and Stage 1995) for estimating characteristics of a marked stand using a stem database. The stem database was collected while harvesting stands with modern harvesters, which collect diameters for every 10 centimetres from cut stems and save this detailed data into a specifi c fi le.…”

Section: Introductionmentioning

confidence: 99%

“…Local adaptation methods were compared to the k-nn MSNmethod (Malinen et al 2001) by using harvester collected stem data to predict stand characteristics. The fi rst locally adaptable non-parametric method presented uses the k-nn MSN-method and a sub-region of the data for estimating new local weighting for the predictor variables.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Locally adaptable non-parametric methods for estimating stand characteristics for wood procurement planning

Malinen¹

2003

Silva Fenn.

Self Cite

View full text Add to dashboard Cite

Malinen, J. 2003. Locally adaptable non-parametric methods for estimating stand characteristics for wood procurement planning. Silva Fennica 37(1): 109-120.The purpose of this study was to examine the use of the local adaptation of the non-parametric Most Similar Neighbour (MSN) method in estimating stand characteristics for wood procurement planning purposes. Local adaptation was performed in two different ways: 1) by selecting local data from a database with the MSN method and using that data as a database in the basic k-nearest neighbour (k-nn) MSN method, 2) by selecting a combination of neighbours from the neighbourhood where the average of the predictor variables was closest to the target stand predictor variables (Locally Adaptable Neighbourhood (LAN) MSN method). The study data used comprised 209 spruce dominated stands located in central Finland and was collected with harvesters. The accuracy of the methods was analysed by estimating the tree stock characteristics and the log length/diameter distribution produced by a bucking simulation. The local k-nn MSN method was not notably better than the k-nn MSN method, although it produced less biased estimates on the edges of the input space. The LAN MSN method was found to be a more accurate method than the k-nn methods.

show abstract

Section: Comparison Of Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Locally adaptable non-parametric methods for estimating stand characteristics for wood procurement planning

Malinen¹

2003

Silva Fenn.

Self Cite

View full text Add to dashboard Cite

show abstract

Modern Data Acquisition for Forest Inventories

Holopainen

Kalliovirta

Forest Inventory

View full text Add to dashboard Cite

Uncertainty in timber assortment estimates predicted from forest inventory data

et al. 2010

View full text Add to dashboard Cite

Uncertainty factors related to inventory methodologies and forest-planning simulation computings in the estimation of logging outturn assortment volumes and values were examined. The uncertainty factors investigated were (1) forest inventory errors, (2) errors in generated stem distribution, (3) effects of generated stem distribution errors on the simulation of thinnings and (iv) errors related to the prediction of stem form and simulation of bucking. Regarding inventory errors, standwise field inventory (SWFI) was compared with area-based airborne laser scanning (ALS) and aerial photography inventorying. Our research area, Evo, is located in southern Finland. In all, 31 logging sites (12 clear-cutting and 19 thinning sites) measured by logging machine in winter 2008 were used as field reference data. The results showed that the most significant source of error in the prediction of clear-cutting assortment outturns was inventory error. Errors related to stem-form prediction and simulated bucking as well as generation of stem distributions also cause uncertainty. The bias and root-mean-squared error (RMSE) of inventory errors varied between -11.4 and 21.6 m 3 /ha and 6.8 and 40.5 m 3 /ha, respectively, depending on the assortment and inventory methodology. The effect of forest inventory errors on the value of logging outturn in clear-cuttings was 29.1% (SWFI) and 24.7% (ALS). The respective RMSE values related to thinnings were 41.1 and 42%. The generation of stem distribution series using mean characteristics led to an RMSE of 1.3 to 2.7 m 3 /ha and a bias of -1.2 to 0.6 m 3 /ha in the volume of all assortments. Prediction of stem form and simulation of bucking led to a relative bias of -0.28 to 0.00 m 3 in predicted sawtimber volume. Errors related to pulpwood volumes were -0.4 m 3 to 0.21 m 3 .

show abstract

Application of Most Similar Neighbor Inference for Estimating Marked Stand Characteristics Using Harvester and Inventory Generated Stem Databases

Cited by 25 publications

References 5 publications

Locally adaptable non-parametric methods for estimating stand characteristics for wood procurement planning

Locally adaptable non-parametric methods for estimating stand characteristics for wood procurement planning

Modern Data Acquisition for Forest Inventories

Uncertainty in timber assortment estimates predicted from forest inventory data

Contact Info

Product

Resources

About