Lecture 5.0 vegetation_sampling
4 likes3,592 views
This document provides an introduction to vegetation sampling. It discusses key concepts like sampling, statistical and biological populations, statistical and physical samples. It also covers the purposes and importance of sampling, as well as different types of sampling designs including random, systematic, cluster and stratified sampling. The document concludes by outlining some advantages and disadvantages of these different sampling methods.
Lecture 5.0 vegetation_sampling
- 1. STT 2063 Forest Science and Management 5.0 Introduction to Vegetation Sampling 1
- 2. 5.1 Sampling Sampling is the process of selecting units (e.g., people, plants) from a population of interest so that by studying the sample we may fairly generalize our results back to the population from which they were chosen. Statistical population Population Biological population 2
- 3. Statistical population - The entire set of data of interest e.g. tree height, tree diameter, no. of tree Biological population - Aggregation of individual organism of a single species. 3
- 4. Statistical sample Sample Physical sample Statistical sample - A portion of a larger set of data i.e. (the stat. Pop.) - It’s part or subset of the population that is actually measured. Physical sample - A portion, or subset of a collection of one or more materials objects. 4
- 5. Examples of statistical and physical sample Example 1 Collect one–litre sample from pond PS Collected a portion of the entire volume of pond water. Measurement of PH, Temp. or phosphate SS Measurement made only from the collected portion i.e. the one-litre sample. 5
- 6. Example 2 Sample a vegetation of a forest PS Small portion of all vegetation from a forest. Volume of standing trees, basal area SS Result only from a small portion of the entire forest. - The sample is the group of (people, plant etc.) who you select to be in your study. 6
- 7. 5.2 Purpose of sampling? • To draw conclusions about populations from samples. • Enables us to determine a population`s characteristics by directly observing only a portion (or sample) of the population. 5.3 The importance of sampling • Impossible to observe/measure all • Time • Cost - cheaper to observe a part rather than the whole • Efficiency in resources (labour, logistics, infrastructure) 7
- 8. 5.4 Types of Sampling (i) Non-statistical sampling (ii) Statistical sampling Non-statistical sampling Refers to collection of data/information by subjective way/method. Characteristics: • Rely on skill, experience and thoroughness of researchers. • No constant or consistent sampling design. • Reliability can not be determine • Very low cost. 8
- 9. Statistical sampling Refers to data that are collected through scientific way. It involves selection of samples. Characteristics: • Accepted scientifically. • Selection of sample that represent the population. • Reliability can be measured e.g. 95% or 99% Advantages: • Quick, because sample are used. • Scope of study are wide • High reliability. • Data recorded are very systematic. • Collection of data can be repeated at the same reliability. • Similar method can be employ at different site for valid comparison. 9
- 10. 5.5 Vegetation sampling Vegetation populations are relatively easy to sample because: Vegetation doesn't move, hence as much time as is necessary can be taken to make measurements or observations. Vegetation is usually easily visible. Vegetation doesn't change much over short periods of time. 10
- 11. Few characteristics about vegetation do cause some concern when sampling, namely: There may be difficulties in defining exactly what constitutes a vegetation individual. For example, many plants consist of multiple stems emerging from a single root system interconnected by runners. Are these multiple individuals or one individual? Vegetation often grows in patches or groups. Is the patch or each plant in the patch the individual? Vegetation in one area can overlap, that is, exists in a number of scales (e.g. grass, shrubs, under canopy trees, canopy trees). 11
- 12. 5.6 Vegetation Sampling Approaches Most vegetation and forest sampling derives from one of the following approaches. • Use the individuals. If we are interested in average tree height, we simply locate some trees in the area of interest (usually chosen at random) and measure the height of all selected trees. • Use predefined areas. This is the basic idea behind quadrate and strip quadrate sampling units. Often it is easier to locate a randomly chosen study area than to locate/select a random individual. 12
- 13. • Use predefined lines. Lines may be used to identify which individuals are to be observed/measured (line intersect selection) or can themselves constitute the measurement (line intercept sampling). Line can also be used in conjunction with areas to locate study individuals (point quarter centre or strip counts). Line sampling methods are sometimes also referred to as plotless sampling. • Use distances between individuals. For determining certain characteristics relating to the distributions of individuals in space, individual-to- individual distances or point-to-individual distances may be used. 13
- 14. 5.7 Experimental Design Method how samples are: • Chosen to represent the population of study. • Determination of no. or % area to be sampled. • Determination of size, shape and method of data collection. It is the research plan either in the field or laboratory experiments. Done prior to data collection. It is a procedure for specific sampling and data analyses that are determine using statistical basis. 14
- 15. 5.8 Classification of sampling design Design Selective Objective Systematic Random Cluster Unstratified Stratified Simple Multistage Simple Stratified Two > Two Random Random stage stage Sampling Sampling 15
- 16. 5.8.1 Selective/Authoritative sampling Sample are selected in subjective manner. Depend very much on the skill, experience and objective of individual researcher. Employed certain strategy during measurement in the field. Method: •Samples are selected based on certain characteristics. •Sampling area must contain the individuals or species that are to be study. •Data collection are easy; avoid areas that are difficult such as steep slope, depression etc. 16
- 17. 5.8.2 Objective sampling Involve random selection and probability sampling does. We know the odds or probability that we have represented the population well. We are able to estimate confidence intervals for the statistic. In Selective samples, we may or may not represent the population well, and it will often be hard for us to know how well we've done so. In general, researchers prefer random sampling methods over selective ones, and consider them to be more accurate and rigorous. 17
- 18. 5.8.3 Random sampling • Selection of sample are random i.e. every parts of the population have the same chance to be selected as sample. • Avoid bias in choosing samples - Samples randomly selected usually by random numbers. 18
- 19. 19
- 20. Method - Area of study are divided into quadrates that have similar size. - Every quadrate are assigned with a number. - Selection of samples must be based on random numbers - Selection of samples are done either with replacement or without replacement. - Mostly without replacement - No. of samples are usually determined prior to random selection. 20
- 21. Simple random sampling & Stratified random sampling SIMPLE RANDOM SAMPLE STRATIFIED RANDOM SAMPLE - all plot have the same - Population are divided into probability to be chosen homogeneous group & samples are or selected homo group. - This will help to ensure that all types are represented in the overall sample. 21
- 22. Multistage Random Sampling • Sampling are done in stages. Usually involve subsample in the main sample. Main Sample Sub sample Double stage random sampling Random sampling > two stages 22
- 23. 5.8.4 Systematic sampling - Sampling with a system. - From the sampling frame, a starting point is chosen at random, and thereafter at regular intervals. - Assumption: a population of interest in distributed randomly over space. - Samples are selected systematically where distribution. - Samples are arranged I such a way that every part of the population have the same number of samples. - Easy to locate. - Distribution of samples are uniform. 23
- 24. For Forest Inventory systematic sampling is usually chosen because: (i) easily planned, (ii) faster in execution and mostly cheaper; (iii) it gives better estimates of the mean; (iv) it gives thus better precision compared to random sampling General characteristics: • Formation Transect line. • Study area is divided into quadrate or plot. • Distance between samples are the same. • Distribution of samples in a pattern. 24
- 26. 26
- 27. Diagrammatic plan for a 20 percent systematic strip cruise. Sample strips 1 chain wide are spaced at regular intervals of 5 chains. 27
- 28. 5.8.5 Cluster Sampling In cluster sampling the units sampled are chosen in clusters, close to each other. Examples are rattan in the same strip, or successive sample units along a transect line. The population is divided into clusters, and some of these are then chosen at random. Within each cluster units are then chosen by simple random sampling or some other method. Ideally the clusters chosen should be dissimilar so that the sample is as representative of the population as possible. 28
- 29. 29
- 30. Advantages and disadvantages of sampling method Random Systematic Cluster Stratified Advantages ideal for spreads the saving of saving of statistical sample more travelling time, travelling purposes evenly over and time, and the population consequent consequen easier to reduction in t reduction conduct than a cost in cost simple random sample spreads the sample more evenly over the population easier to conduct than a simple random sample 30
- 31. Advantages and disadvantages of sampling method Random Systematic Cluster Stratified Disadvantages hard to the system units close to units close achieve in may interact each other to each practice with some may be very other may requires an hidden pattern similar and so be very accurate list of in the less likely to similar and the whole population, represent the so less population e.g. every third whole likely to expensive to house along population represent conduct as the street larger the whole those sampled might always sampling error population may be be the middle than simple larger scattered over one of a random sampling a wide area terrace of sampling error than three simple random sampling 31
- 32. END OF LECTURE 32