Survey Analysis Framework
1 ⍺-diversity
- full descriptions of metadata factors
- Observed species
2 β-diversity along depth profiles
- 2.1 Basin

Terms:

Landscape-scale = all Taylor Valley samples
Regional-scale = within basin and/or reach
- n = 2 basins (LHSS and WHC)
- n=4 reaches (upslope and downslope within each basin)
Local-scale = within each pit
- n = 4 pits per reach (max)
- n = 16 total survey pits
Depth profile = discrete 5 cm soil samples from soil surface to permafrost interface (up to 30 cm)
- n = 7 depth samples per pit (max)
- metadata factor descriptions
  - Depth
    - 7 levels: 0 cm, 5 cm, 10 cm, 15 cm, 20 cm, 25 cm, 30 cm
  - Depth2
    - Surface: 0 cm [0,5)
    - Intermediate: 5 - 25 cm (depending on pit)
    - Permafrost interface: 20 - 30 cm (depending on pit)
  - Depth3_combo0and5
    - Surface: 0 cm & 5 cm [0,10)
    - Intermediate: 10 - 25 cm (depending on pit)
    - Permafrost interface: 20 - 30 cm (depending on pit)
Regional species pool = γ-diversity of OTUs present within each basin and/or reach
- note: with 4 spatial scales to contend with (whole valley, basin, reach, and pit) the ‘landscape-regional-local scale’ & α, β γ-diversity terminology gets a little messy…
  - options to deal with this include dropping the whole valley scale, then landscape = basin, regional = reach, and local = pit
  - but I think it might be more interesting to keep landscape = whole valley and have regional = reach, this would let us do some analyses related to edaphic variable impacts on assembly / structure

Survey Analysis Framework

Big Picture:

Investigating landscape, regional, and local-scale diversity and dynamics of local-scale assembly & structuring along active layer depth profiles

How does the larger regional species pool influence the assembly of microbial communities along the active layer depth profiles?

Goals:

Describe patterns in β-diversity along depth profiles (including taxonomy)

1.1. by basin

1.2. by reach

1.3. individually
Explore γ-diversity

2.1. at landscape-scale (whole valley)

2.2. within each basin

2.3. within each reach
Investigate assembly mechanisms / species sorting / environmental filtering along depth profiles by incorporating edaphic variables

1 ⍺-diversity

full descriptions of metadata factors

Survey: Alpha Diversity Measures with Metadata Descriptives

Survey: Alpha Diversity Measures with Metadata

23 Variables 74 Observations

Experiment

n	missing	distinct	value
74	0	1	ActiveLayer_S

 Value      ActiveLayer_S
 Frequency             74
 Proportion             1

LocationAbbrev

n	missing	distinct
74	0	2

 Value       LHSS   WHC
 Frequency     40    34
 Proportion 0.541 0.459

Location

n	missing	distinct
74	0	2

 Value      Lake Hoare South Shore      Worm Herder Creek
 Frequency                      40                     34
 Proportion                  0.541                  0.459

FullDescription

n	missing	distinct
74	0	74

lowest :	LHSS_DownSlope_Pit1_00cm	LHSS_DownSlope_Pit1_05cm	LHSS_DownSlope_Pit1_10cm	LHSS_DownSlope_Pit1_15cm	LHSS_DownSlope_Pit1_30cm
highest:	WHC_UpSlope_Pit4_05cm	WHC_UpSlope_Pit4_10cm	WHC_UpSlope_Pit4_15cm	WHC_UpSlope_Pit4_20cm	WHC_UpSlope_Pit4_25cm

Basin

n	missing	distinct
74	0	2

 Value      Lake Bonney  Lake Hoare
 Frequency           34          40
 Proportion       0.459       0.541

Site

n	missing	distinct
74	0	2

 Value      DownSlope   UpSlope
 Frequency         35        39
 Proportion     0.473     0.527

BasinSite

n	missing	distinct
74	0	4

 Value      LHSS_DownSlope   LHSS_UpSlope  WHC_DownSlope    WHC_UpSlope
 Frequency              19             21             16             18
 Proportion          0.257          0.284          0.216          0.243

Pit

n	missing	distinct	Info	Mean	Gmd
74	0	4	0.937	2.459	1.265

 Value          1     2     3     4
 Frequency     19    20    17    18
 Proportion 0.257 0.270 0.230 0.243

For the frequency table, variable is rounded to the nearest 0

Depth

n	missing	distinct
74	0	7

 Value      00 cm 05 cm 10 cm 15 cm 20 cm 25 cm 30 cm
 Frequency      9    14    14    12    12     8     5
 Proportion 0.122 0.189 0.189 0.162 0.162 0.108 0.068

Depth2

n	missing	distinct
74	0	3

 Value                   Surface         Intermediate Permafrost Interface
 Frequency                     9                   53                   12
 Proportion                0.122                0.716                0.162

Depth3_combo0and5

n	missing	distinct
74	0	3

 Value                   Surface         Intermediate Permafrost Interface
 Frequency                    23                   39                   12
 Proportion                0.311                0.527                0.162

sample_sums.al.

n	missing	distinct	Info	Mean	Gmd	.05	.10	.25	.50	.75	.90	.95
74	0	74	1	6057	6018	1257	1432	2124	3808	7312	13841	21239

lowest : 1100 1189 1220 1239 1267 , highest: 20288 23006 28137 29007 31171

Depthcm

n	missing	distinct	Info	Mean	Gmd
74	0	7	0.975	13.24	10.16

 Value          0     5    10    15    20    25    30
 Frequency      9    14    14    12    12     8     5
 Proportion 0.122 0.189 0.189 0.162 0.162 0.108 0.068

For the frequency table, variable is rounded to the nearest 0

Observed

n	missing	distinct	Info	Mean	Gmd	.05	.10	.25	.50	.75	.90	.95
74	0	63	1	216.5	60.43	97.45	131.50	194.75	222.00	252.75	276.90	288.70

lowest : 58 66 76 89 102 , highest: 288 290 291 309 324

Chao1

n	missing	distinct	Info	Mean	Gmd	.05	.10	.25	.50	.75	.90	.95
74	0	74	1	334.3	113.2	136.6	207.2	291.3	329.3	410.4	458.3	476.3

lowest : 75.1 81.4 118 124.286 143.167 , highest: 474.708 479.233 492.016 505.241 531.875

se.chao1

n	missing	distinct	Info	Mean	Gmd	.05	.10	.25	.50	.75	.90	.95
74	0	74	1	31.8	13.29	14.72	18.73	23.73	30.01	37.15	47.00	50.72

lowest : 8.6785 10.4091 13.6612 14.3328 14.9258 , highest: 49.7729 52.4765 53.2518 70.3463 72.2256

ACE

n	missing	distinct	Info	Mean	Gmd	.05	.10	.25	.50	.75	.90	.95
74	0	74	1	344.5	111	146.1	226.6	305.6	343.3	423.4	458.4	488.5

lowest : 76.3742 88.469 126.824 145.93 146.231 , highest: 487.444 490.363 491.951 505.705 510.129

se.ACE

n	missing	distinct	Info	Mean	Gmd	.05	.10	.25	.50	.75	.90	.95
74	0	74	1	10.42	2.26	6.205	7.930	9.578	10.620	11.930	12.691	12.901

lowest : 4.47473 4.59813 5.67585 6.18504 6.21651 , highest: 12.8633 12.9723 13.592 13.7073 14.1845

Shannon

n	missing	distinct	Info	Mean	Gmd	.05	.10	.25	.50	.75	.90	.95
74	0	74	1	4.312	0.6901	2.406	3.678	4.252	4.439	4.706	4.961	5.071

lowest : 1.78214 2.0092 2.07157 2.08891 2.57598 , highest: 5.07047 5.07115 5.11229 5.13396 5.24583

Simpson

n	missing	distinct	Info	Mean	Gmd	.05	.10	.25	.50	.75	.90	.95
74	0	74	1	0.9475	0.05862	0.7259	0.9363	0.9587	0.9728	0.9822	0.9877	0.9887

 Value      0.590 0.645 0.680 0.695 0.740 0.755 0.845 0.935 0.945 0.950 0.955 0.960
 Frequency      1     1     1     1     1     1     1     2     2     5     4     5
 Proportion 0.014 0.014 0.014 0.014 0.014 0.014 0.014 0.027 0.027 0.068 0.054 0.068
                                               
 Value      0.965 0.970 0.975 0.980 0.985 0.990
 Frequency      6    13     7    10    10     3
 Proportion 0.081 0.176 0.095 0.135 0.135 0.041

For the frequency table, variable is rounded to the nearest 0.005

InvSimpson

n	missing	distinct	Info	Mean	Gmd	.05	.10	.25	.50	.75	.90	.95
74	0	74	1	42.05	28.85	3.67	15.71	24.20	36.80	56.08	81.22	88.88

lowest : 2.44842 2.82872 3.17169 3.29919 3.86957 , highest: 88.8079 89.0185 100.858 105.685 116.307

Fisher

n	missing	distinct	Info	Mean	Gmd	.05	.10	.25	.50	.75	.90	.95
74	0	63	1	83.85	33.24	25.88	39.13	68.86	84.02	102.96	119.31	127.78

lowest : 13.0575 15.4314 18.5577 22.8803 27.4907 , highest: 127.264 128.732 129.469 143.182 155.262

Variables with all observations missing:

Control

Observed species

Did the surface samples (0 and 5 cm) have higher richness than the intermediate and Permafrost Interface samples?

Answer: yes

wilcox.test(Observed ~ Depth3_combo0and5, data = Survey_alpha_meta,
            paired = F,p.adjust.method = "BH",alternative="less",
            subset = Depth3_combo0and5 %in% c("Permafrost Interface","Surface"))

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  Observed by Depth3_combo0and5
## W = 205, p-value = 0.9905
## alternative hypothesis: true location shift is less than 0

2 β-diversity along depth profiles

Active Layer survey: Exploring communities along depth profiles

Investigating landscape, regional, and local-scale diversity and dynamics of local-scale assembly / structuring along active layer depth profiles

Kelli Feeser

2023-11-06