Meeting X
Aanwezig: X ## Research question 1 How successful is the practice of sowing commercial seed mixtures on rubble-composed soils to create species-rich grasslands, and how is this success affected by soil conditions (pH and Olsen P), numbers of seed in the seed mixtures, and years since initial sowing?
Zie script > RQ1_Descriptive.R & RQ1_models.R
Stephanie: Enkel nog de toevoeging om eventueel ook het verschil tussen de groepen statistisch te testen. (kan met de functies emmeans en daarna cld uit het pakket emmeans, laat me weten als je daar hulp bij wil) Anders ga je geen echte uitspraken kunnen doen als “er zijn gemiddeld gezien meer spontane soorten dan ingezaaide soorten gespot in de proefvlakken”.
TEST DIFFERENCES
Difference between “observed and sown” and “observed and non-sown” #### 1) Model > mod <- lmerTest::lmer(value ~ category + (1|location), data = df_model) > > - Value = number of species/location > - Category = “observed and sown” and “observed and non-sown” > - (1|location) = Compare categories while accounting for the fact that measurements come from the same locations. > > emmeans(mod, pairwise ~ category) > is dit noidg? > ! singular fit warning happened
2) Wilcoxon signed rank test
wilcox.test(df_NS\(n_obs_sown, df_NS\)n_obs_nonsown,paired = TRUE)
MODELS
General: - For each characteristic, we look at three different groups: (i) observed, (ii) observed and non-sown, and (iii) observed and sown - The predictor variables are soil conditions (pH and Olsen P), numbers of seed in the seed mixtures, and years since initial sowing #### Description of predictor variables | Statistic | Number of seeds (seed mixture) | Age | Olsen P | pH | |—|—:|—:|—:|—:| | Mean | 36.04 | 8.02 | 50.39 | 7.65 | | Median | 34.5 | 7 | 49.66 | 7.81 | | Minimum | 10 | 3 | 9.88 | 6.53 | | Maximum | 63 | 17 | 114.82 | 8.20 |
Number of species
How are total, sown, and non-sown plant species richness influenced by soil conditions (pH and Olsen P), seed mixture richness, and years since sowing?
m_obs <- glmmTMB(n_obs ~ scale(sown_mixt) + scale(age) + scale(olsenp_2022) + scale(ph_2022), family = nbinom2, df)
m_obs_nonsown <- glmmTMB(n_nonsown ~ scale(sown_mixt) + scale(age) + scale(olsenp_2022) + scale(ph_2022), family = nbinom2, df)
m_obs_sown <- glmmTMB(n_sown ~ scale(sown_mixt) + scale(age) + scale(olsenp_2022) + scale(ph_2022), family = nbinom2, df)
Do green managers appear to adjust seed mixture richness according to soil characteristics (pH and Olsen P)?
m_sown <- glmmTMB(sown_mixt ~ scale(olsenp_2022) + scale(ph_2022), family = nbinom2, df)
Vegetation cover
Estimates of models are very high - What does this mean?
Grass Cover
Model for grass cover sown - rediduals >> KS TEST “deviation significant” >> a lot of zeros (not a problem with the “number of grass species”).
Annual Cover
Model for Annual cover (for all three labels) - rediduals >> KS TEST “deviation significant”
Overview
Research Question 2
What are the functional characteristics of commercial seed mixtures applied in the city of Ghent and of the plant community of urban grasslands established on rubble-composed soils through the sowing of these seed mixtures?
Lander: gewoon beschrijving van de variatie in CSR/Ellenberg in de verschillende pools (sown, etc.) tussen de locaties
Potential Nectar Value (PNV)
Mean
| dataset | Nectar (mean ± sd) | Nectar (min–max) |
|---|---|---|
| Observed | 29.98 ± 9.47 | 15.37–60.98 |
| Observed & Nonsown | 26.61 ± 12.38 | 1.98–72.51 |
| Observed & Sown | 37.17 ± 25.18 | 0.00–125.00 |
| Sown | 34.22 ± 10.06 | 16.14–67.52 |
Differences
| Category | Emmean | SE | df | Lower CL | Upper CL | Group |
|---|---|---|---|---|---|---|
| Observed | 30.0 | 2.26 | 188 | 24.3 | 35.7 | ab |
| Observed & Nonsown | 26.6 | 2.26 | 188 | 20.9 | 32.3 | a |
| Observed & Sown | 37.2 | 2.26 | 188 | 31.5 | 42.8 | b |
| Sown | 34.2 | 2.26 | 188 | 28.5 | 39.9 | ab |
CSR
Mean
| dataset | C (mean ± sd) | C (min–max) | S (mean ± sd) | S (min–max) | R (mean ± sd) | R (min–max) |
|---|---|---|---|---|---|---|
| Observed | 29.10 ± 4.85 | 20.32–45.41 | 19.92 ± 3.50 | 11.74–27.44 | 50.98 ± 4.85 | 33.70–60.03 |
| Observed and sown | 32.38 ± 15.40 | 0.00–71.82 | 20.54 ± 13.72 | 0.00–53.99 | 36.66 ± 15.42 | 0.00–71.94 |
| Observed and Non-sown | 26.77 ± 6.20 | 15.92–48.22 | 19.43 ± 3.58 | 10.28–26.14 | 53.80 ± 5.76 | 35.12–64.15 |
| Sown | 32.86 ± 7.35 | 20.78–52.24 | 22.82 ± 7.21 | 2.98–34.24 | 44.32 ± 3.40 | 36.28–55.05 |
Differences
| trait | group | emmean | SE | lower.CL | upper.CL | .group |
|---|---|---|---|---|---|---|
| C | Observed | 29.21 | 0.63 | 27.63 | 30.79 | a |
| C | Observed_Nonsown | 27.24 | 0.73 | 25.42 | 29.05 | a |
| C | Observed_Sown | 35.34 | 1.29 | 32.14 | 38.55 | b |
| C | Sown | 32.51 | 0.51 | 31.23 | 33.79 | b |
| S | Observed | 19.89 | 0.65 | 18.28 | 21.50 | a |
| S | Observed_Nonsown | 19.41 | 0.74 | 17.55 | 21.26 | a |
| S | Observed_Sown | 21.39 | 1.31 | 18.11 | 24.66 | ab |
| S | Sown | 23.42 | 0.52 | 22.11 | 24.72 | b |
| R | Observed | 50.90 | 0.62 | 49.35 | 52.46 | b |
| R | Observed_Nonsown | 53.36 | 0.72 | 51.57 | 55.14 | b |
| R | Observed_Sown | 43.27 | 1.27 | 40.11 | 46.42 | a |
| R | Sown | 44.07 | 0.50 | 42.82 | 45.33 | a |
EIVE
Differences
| trait | group | emmean | SE | lower.CL | upper.CL | .group |
|---|---|---|---|---|---|---|
| EIVE_N | Observed | 5.25 | 0.04 | 5.14 | 5.35 | a |
| EIVE_N | Observed_Nonsown | 5.47 | 0.05 | 5.35 | 5.59 | b |
| EIVE_N | Observed_Sown | 4.55 | 0.09 | 4.33 | 4.77 | c |
| EIVE_N | Sown | 4.63 | 0.04 | 4.54 | 4.72 | c |
| EIVE_M | Observed | 4.29 | 0.02 | 4.23 | 4.34 | a |
| EIVE_M | Observed_Nonsown | 4.38 | 0.03 | 4.32 | 4.45 | b |
| EIVE_M | Observed_Sown | 3.99 | 0.05 | 3.88 | 4.10 | c |
| EIVE_M | Sown | 3.99 | 0.02 | 3.94 | 4.03 | c |
| EIVE_R | Observed | 6.02 | 0.03 | 5.95 | 6.09 | a |
| EIVE_R | Observed_Nonsown | 6.03 | 0.03 | 5.95 | 6.11 | a |
| EIVE_R | Observed_Sown | 5.97 | 0.06 | 5.83 | 6.11 | a |
| EIVE_R | Sown | 6.08 | 0.02 | 6.03 | 6.14 | a |
| EIVE_T | Observed | 4.28 | 0.01 | 4.25 | 4.32 | ab |
| EIVE_T | Observed_Nonsown | 4.32 | 0.02 | 4.28 | 4.36 | b |
| EIVE_T | Observed_Sown | 4.16 | 0.03 | 4.10 | 4.23 | c |
| EIVE_T | Sown | 4.26 | 0.01 | 4.23 | 4.29 | a |
APPENDIX
Total number of species
Potential Nectar Value
CWM
| dataset | Nectar (mean ± sd) | Nectar (min–max) |
|---|---|---|
| Observed | 31.21 ± 23.55 | 2.39–88.07 |
| Observed & Nonsown | 21.35 ± 22.32 | 0.41–111.38 |
| Observed & Sown | 45.78 ± 38.35 | 0.00–125.00 |
CSR
With “CWM”
\[ \text{CWM}_{CSR} = \sum_{i=1}^{S} (P_i \times T_i) \]
With \(S\) the absolute measure
With \(P_{\text{i}}\) the relative abundance of species i
With \(T_{\text{i}}\) the trait value (CSR) of species i
CWM
| dataset | C (mean ± sd) | C (min–max) | S (mean ± sd) | S (min–max) | R (mean ± sd) | R (min–max) |
|---|---|---|---|---|---|---|
| Observed | 28.59 ± 10.00 | 13.38–71.90 | 24.74 ± 10.42 | 4.22–57.67 | 46.68 ± 10.09 | 16.27–66.20 |
| Observed and sown | 31.10 ± 18.53 | 0.00–71.82 | 21.26 ± 19.02 | 0.00–64.38 | 37.22 ± 17.73 | 0.00–85.95 |
| Observed and Non-sown | 26.37 ± 11.98 | 10.26–77.92 | 24.59 ± 10.36 | 3.75–48.64 | 49.04 ± 11.75 | 10.53–68.63 |
CWM (contribution)
With “CWM (contribution)”
\[ X_{\text{total}} = X_{\text{sown}} + X_{\text{non-sown}} \]
| dataset | C (mean ± sd) | C (min–max) | S (mean ± sd) | S (min–max) | R (mean ± sd) | R (min–max) |
|---|---|---|---|---|---|---|
| Observed | 28.59 ± 10.00 | 13.38–71.90 | 24.74 ± 10.42 | 4.22–57.67 | 46.68 ± 10.09 | 16.27–66.20 |
| Observed and sown | 11.06 ± 10.44 | 0.07–34.72 | 8.09 ± 10.51 | 0.00–55.86 | 13.72 ± 11.45 | 0.08–39.72 |
| Observed and Non-sown | 18.68 ± 11.36 | 2.77–67.14 | 17.49 ± 10.67 | 1.72–48.43 | 34.39 ± 16.09 | 5.63–64.30 |
EIVE
With “CWM (contribution)”
\[ X_{\text{total}} = X_{\text{sown}} + X_{\text{non-sown}} \]
With “CWM”
\[ \text{CWM}_{CSR} = \sum_{i=1}^{S} (P_i \times T_i) \]
With \(S\) the absolute measure
With \(P_{\text{i}}\) the relative abundance of species i
With \(T_{\text{i}}\) the trait value (CSR) of species i