Statistical power in forest monitoring

Foster, Jeffrey R.

doi:10.1016/s0378-1127(01)00591-6

Cited by 37 publications

(32 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An approach referred to look at each pairwise comparison by doing power analysis for a set of unpaired t-tests (Foster et al 2001). Further, we compared GS and BA means and evaluated differences from last measurement with targeted values prescribed with management plans for each forest type.…”

Section: Discussionmentioning

confidence: 99%

“…Many researchers emphasize that power analysis contributes the most in the planning research phase when desired effect size, significance level, test power and sample size should be chosen or estimated (Foster 2001, Di Stefano 2001. Foster (2001) demonstrated capacities of a priori power analysis discussion forest monitoring program in native uneven aged multispecies forests in the United States.…”

Section: Hybrid Power Analysis -Sample Size Estimationmentioning

confidence: 99%

“…Foster (2001) demonstrated capacities of a priori power analysis discussion forest monitoring program in native uneven aged multispecies forests in the United States. He stated the relative mean difference of forest productivity attribute (expressed in percentage) as desired effect size and calculated required sample size and estimated power (based on data collected in the first monitoring phase).…”

Section: Hybrid Power Analysis -Sample Size Estimationmentioning

confidence: 99%

“…Power analysis enables to calculate power in relation with experimental design, sample size, ES, significance level α and the variability of data (Foster 2001, Di Stefano, 2001). South and VanderSchaaf (2006) proposed a "hybrid" power analysis that could be useful to discuss reason for non-significant result: no effect or not enough replication to produce a small enough error term.…”

Section: Introductionmentioning

confidence: 99%

“…South and VanderSchaaf (2006) proposed a "hybrid" power analysis that could be useful to discuss reason for non-significant result: no effect or not enough replication to produce a small enough error term. Foster (2001) demonstrated application of power analysis related to forest monitoring program.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

The Post-Hoc Power Analysis of Forest Productivity Attributes in Experimental Study in Central Bosnia

Čabaravdić¹,

Ibrahimspahić²,

Starcevic³

2017

AgricultForest

View full text Add to dashboard Cite

SUMMARYForest productive attributes changes over time in native forests has been recognized as crucial challenge for management of uneven aged mixed forests in Bosnia and Herzegovina since middle of the last century. Experimental study has been carried out on set of experimental plots established in mixed stands on mountain Igman in central Bosnia. The most important forest productivity attributes changes based on repeated measures have been monitored over time. The aim of this research was to conduct the post-hoc power analysis for monitored forest attributes: basal area per ha (BA), growing stock per ha (GS) and current annual increment per ha (CAI v ). Here are used repeated measures conduced on the 10 experimental plots in two types of mixed stands: fir-spruce and fir-spruce-beech plots (five plots per each type) measured in five (BA and GS) and four (CAI v ) occasions in periods between 10-20 years. Analyses of variance (ANOVA) within and within-between repeated measures were applied and power analysis was performed. ANOVA within forest type over time showed highly significant differences for all attributes (α = 0.05, p < 0.001). Here, power analysis for comparison of stand attributes resulted in observed high power values ranged from 82% to 99% (very low risk of Type II errors). Then, ANOVA between two forest types over time showed different significances for forest attributes (α = 0.05, p BA = 0.25, p GS = 0.23 and p CAIv = 0.02). The risks of Type II errors were high for BA and GS (from 66% to 72%) while conclusions for CAI v could be accepted with very low risk (4%). So, the post-hoc power analysis of comparisons of stand attributes between forests types found low power for BA (28%) and GS (34%) and high power for CAI v (96%). These findings confirm importance of proper forest species composition planning in mixed stands related to highest wood productivity and other forest characteristics as biodiversity.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Hybrid Power Analysis -Sample Size Estimationmentioning

confidence: 99%

Section: Hybrid Power Analysis -Sample Size Estimationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The Post-Hoc Power Analysis of Forest Productivity Attributes in Experimental Study in Central Bosnia

Čabaravdić¹,

Ibrahimspahić²,

Starcevic³

2017

AgricultForest

View full text Add to dashboard Cite

show abstract

Detecting change in advance tree regeneration using forest inventory data: the implications of type II error

Westfall

McWilliams

2011

Environ Monit Assess

View full text Add to dashboard Cite

Achieving adequate and desirable forest regeneration is necessary for maintaining native tree species and forest composition. Advance tree seedling and sapling regeneration is the basis of the next stand and serves as an indicator of future composition. The Pennsylvania Regeneration Study was implemented statewide to monitor regeneration on a subset of Forest Inventory and Analysis plots measured by the U.S. Forest Service. As management techniques are implemented to improve advance regeneration, assessments of the change in the forest resource are needed. When the primary focus is on detecting change, hypothesis tests should have small type II (β) error rates. However, most analyses are based on minimizing type I (α) error rates and type II error rates can be quite large. When type II error rates are high, actual improvements in regeneration can remain undetected and the methods that brought these improvements may be deemed ineffective. The difficulty in detecting significant change in advance regeneration when small type I error rates are given priority is illustrated. For statewide assessments, power (1-β) to detect changes in proportion of area having adequate advance regeneration is relatively weak (≤0.5) when the change is smaller than 0.05. For evaluations conducted at smaller spatial scales, such as wildlife management units, the reduced sample size results in only marginal power even when relatively large changes (≥0.20) in area proportion occur. For fixed sample sizes, analysts can consider accepting larger type I error rates to increase the probability of detecting change (smaller type II error rates) when it occurs, such that management methods that positively affect regeneration can be identified.

show abstract

Using statistical power analysis as a tool when designing a monitoring program: experience from a large-scale Swedish landscape monitoring program

Christensen

Ringvall

2013

Environ Monit Assess

View full text Add to dashboard Cite

The National Inventory of Landscapes in Sweden (NILS) is a large-scale, sample-based monitoring program that combines aerial photointerpretation with field inventory to follow landscape-scale biophysical conditions and changes. A statistical power analysis was conducted before the NILS program began in 2003 with the aim to determine an appropriate sampling effort and compare some design alternatives. The chosen sampling effort was then evaluated in a second power analysis conducted just before the first 5-year re-inventory rotation started. The latter power analysis revealed which magnitude of actual change might be detected within the future for different central monitoring variables. This article reports results from these power analyses and discusses our experiences in using power analysis as a tool for designing large-scale monitoring programs. The results showed that even quite small changes in the more common variables, such as land cover types and more common plant species, can be detected on the national scale. However, on the regional scale, or for less common variables, changes will be more difficult to detect. The power analyses have revealed the size level of changes that will be possible to detect. The results have also generated incentives for further improvements of NILS, e.g., input to the modification and revision of the variable content, flow and hierarchy, and incentives for launching other complementary monitoring programs connected to NILS. They have also created a basis for a better and more user-oriented communication of results from NILS to different stakeholders.

show abstract

Statistical power in forest monitoring

Cited by 37 publications

References 9 publications

The Post-Hoc Power Analysis of Forest Productivity Attributes in Experimental Study in Central Bosnia

The Post-Hoc Power Analysis of Forest Productivity Attributes in Experimental Study in Central Bosnia

Detecting change in advance tree regeneration using forest inventory data: the implications of type II error

Using statistical power analysis as a tool when designing a monitoring program: experience from a large-scale Swedish landscape monitoring program

Contact Info

Product

Resources

About