Ecological Statistics

  • Published on
    19-Dec-2014

  • View
    4.241

  • Download
    3

Embed Size (px)

DESCRIPTION

What we look at ecological statistics, descriptive and inductive statistics, biodiversity index, examples from Sharavathi river basin

Transcript

  • 1. Gururaja KV IISc, Bangalore gururajakv@gmail.com

2. Systematic stratified random sampling Night survey, from 2003 2006, seasonal, search for all Identify and record species, numbers etc. Secondary data on Vegetation studies, RS and GIS Opportunistic observations also included for overall diversity in the region Shannons index (H = - pi ln pi),(H Simpsons index (D = 1/ pi2) (D 3. Varieties of species (Species richness) Abundance of species (#indviduals/man hour (#indviduals/man search) Derived data (Endemics, Endemic abundance) Secondary source (IUCN red list) Percent Evergreens, rainfall, percent landuses, landuses, forest fragmentation data, 4. SamplingLongitude (E) Latitude(N) Altitude (m) sites175.089613.773584 275.0804 13.8532586 374.8839 13.9269580 474.7268 13.965 563 575.1055 13.9735602 674.8428 13.9786598 775.1084 14.0209559 875.1245 14.0418557 975.07314.0831680 1075.0414.058 596 11 75.10913.737969612 5. Euphlyctis cyanophlyctis11 11 Euphlyctis hexadactylus 1 Hoplobatrachus tigerinus1 12 11 Indirana beddomii 1 Indirana semipalmatus151 2 Fejervarya granosa1 Fejervarya caperata 41 523 3 4 Minervarya sahyadris 1 Micrixalus fuscus6 Micrixalus saxicola 2 Nyctibatrachus aliciae2 3 Nyctibatrachus major 15 Philautus sp1 12 Philautus amboli2 637 64 3 7 8 Philautus sp2 2 31 1 1 Philautus tuberohumerus 3 2 21 Polypedates maculatus 1List is incompletespecies wise..as well as site wise 6. Sub-basinOAESE MDP A OFRFETE IFPFEF Nandihol11.14 3.3138.12 5.5211.33 30.56 1650.30 27.00 43.00 9.5819.37 7.15 Haridravathihol 13.86 2.2828.18 5.5818.19 31.92 1409.00 9.0016.00 5.0213.15 5.36 Mavinahol 11.60 4.3741.62 7.889.8124.68 2950.40 12.10 14.60 9.9621.26 7.83 Sharavathihol 12.49 19.16 22.95 14.66 10.32 20.42 5041.30 48.45 84.48 8.0918.96 6.27 Hilkunjihol 5.5743.27 22.46 11.56 4.1812.96 5041.30 43.20 83.60 21.02 32.98 8.52 Hurlihol7.5632.78 27.91 10.74 1.9718.17 4073.40 50.00 89.90 16.47 30.05 9.74 Nagodihol 7.4652.14 16.58 13.66 1.089.075689.00 86.70 43.10 24.91 28.98 10.38 Yennehol10.10 37.89 19.76 15.88 1.3714.86 6531.00 39.40 80.40 14.69 25.20 10.12And more 7. Back to the questions asked Statistical applicationfind out the relationship (among variables and between amphibian diversity and these variables) 8. Ecology deals with The study of spatial and temporal patterns of distribution and abundance of organisms, including causes and consequences (Scheiner and(Scheiner Willig, Willig, 2007) A statistic is a estimate of the value of a parameter. A set of procedures and rules for reducing large masses of data into manageable proportions allowing us to draw conclusions from those data Limitations of statistics Snap shot, data Unlimited Ecology Dynamicsfeedbackcomplexnon linear 9. Part I Framing a good question? Setting the objectives? Sampling design, Replication, Randomnization Part II Parameteric vs Non parametric Goodness of fit Measuring central tendency Hypothesis testing Measuring relationships Part III Measuring biodiversity Ordination, Multivariate analysis Spatial analysis 10. Measurement assignment of a number to something Data information on species, variable, parameter Sample Small portion of a population Population all possible units of a given area Variable Varies with the influence of other. Parameter A measurable unit (eg. Temperature, pH) 11. Ordinal rank order, (1st,2nd,3rd,etc.) Nominal categorized or labeled data (red, green, blue, male, female) Ratio (Interval) indicates order as well as magnitude Bob Cosatner is greatly acknowledged for sharing his slides 12. Independent Variable controlled or manipulated by the researcher; causes a change in the dependent variable, generally represented on x-axis Dependent Variable the variable being measured, represented on y-axis Discreet Variable has a fixed value Continuous Variable - can assume any value 13. Mean (average) Median (middle) Mode (most frequent) 14. Variance standard deviation standard error 15. Sample variance is summation of square of deviance from each data point divided by n-1, In excel it is given by VAR(data) 16. Earlier part was descriptive statistics! Null (H0) vs alternate hypothesis(H1) Null Hypothesis - Statistical hypotheses usually assume no relationship between variables. Eg. there is no association between eye color and eyesight.No changeChange Change detected Reject (1) Accept No change Accept Reject (II) detected 17. 1400TreeTree1200 height1 height 2 1000(cm)(cm)800 11014 864600 2 684 636400 3 810 708200 4 990 7860 5 840 600 0 24 6 8 10 6 9781320 Tree height1 Tree height 2 71002 750 81110 594 9800750 18. Distribution free analysis Comparison between observed data vs expected data Generally for qualitative data and categorical data Chi square = sum (O-E)2/E 19. At df 8, 818.1, p0.05 1800 1600 1400 1200 1000 Y 800 600 400 200 Treeheight1 Treeheight2 0N9 91234 5 67 8 9Min684 594Max 11101320Sum 822870081300Mean 914.222 778.6671200Std. error 45.3719 73.91741100Variance 18527.4 491741000Stand. dev 136.116 221.752Y900Median 978 75080025 prcntil 805 61870075 prcntil1008 825600Treeheig Treeheig 22. F= var between groups/var within group Compares the variance of 2 or more groups With following Assumptions: Groups relatively equal. Standard deviations similar. (Homogeneity of variance) Data normally distributed. Sampling should be randomized. Independence of errors. 23. 42 species, 7 families 27 endemic to Western Ghats (64%) 12 vulnerable 17 near threatened Nandihol Nandihol least richness, abundance Yennehol Yennehol highest Su b-ba sin Rich n ess A bu nd. En dem ic En .A bu Non -en d. Non .A bu Sim pson Sh a nn on Na n di10364 12 6245 .4 5 6 1 .9 6 3 Ha r idr a v a th i14496 28 8218 .7 4 7 2 .3 5 6 Ma v in h ole14488 28 6207 .6 2 9 2 .2 9 8 Sh a r a v a th i14339 27 5 67 .3 1 6 2 .2 9 8 Hilku n ji 20481131 9171 0.1 2 82 .6 5 3 Na g odi 18591145 7147 .5 4 8 2 .4 3 6 Hu r li15389 26 6121 1 .1 002 .5 4 4 Yen n e22661335 9311 3 .2 9 9 2 .8 07 24. Correlation coefficient (r) at significance level (P