Context

Summary

On this webpage we will look on how sand mining in Stradbroke Island has affected native ants, and we will (statitiscally) explore if rehabilation of mine sties has had any effect on the ants’ abundance and diversity

Ant in the sand “Sand mining”

Figure 1. Ants vs Sand Minining, Who will win? Only time and stats will tell if Sand Mine Rehabilitation is Effective for increasing native ant diversity and abundance

Sand Mining in Stradbroke

Stradbroke Island, also known as Minjerribah, is located near the Moreton Bay, just 30 km away southeast of Brisbane CBD. Sand Mining on Stradbroke began in the 1950’s. Concerns about environmental impacts began a decade later, with cases of dredge mining. This type of mining levels high sand dunes and stripping native vegetation. Due to the high economic importance of minerals found in Stradbroke’s sands, like Silica, Rutile, Zircon, Ilimenite, sand mining kept going.

Silbelco Australia operated 3 mine sites, including the Enterprise Mine, which operated close to Ramsar-Listed wetlands but had a small buffer zone. After several environmental inicidents and legal disputes, Tte Enterpise Mine closed in 2019. Rehabilitation of the area is still underway.

Figure 2. Location of Stradbroke Ilsand and Beachside Scenery of Stradbroke Figure 2. Location of Stradbroke Ilsand and Beachside Scenery of Stradbroke

Figure 2. Location of Stradbroke Island and Beachside Scenery of Stradbroke

Analysis

To measure the efficacy of mine rehabilitation efforts, students analyze data collected from the sites.

Data Collection

Students of the Masters of Conservation Biology program traveled to Stradbroke Island to survey ant fauna and their response towards mine rehabilitation.

Each site is classified according to the year in which the rehabilitation happened, There are four distinct possibilities for the year of rehabilitation. At each of these possibilities, three representative sites are randomly chosen by the students, and paired to a nearby control site that has never been sandmined. That makes six sites at each “year of rehabilitation”.

Ants are collected using pitfall traps (See Figure 3). Each study site contains 10 traps, seperated by a meter from each other. Ants are collected after one night of trapping.

Figure 3. Basic architecture of insect pitfalltrap (c) Bedfordshire Natural History Society

Species Abundance and Diversity

Fisrt lets look on how each species is represented in control and rehabilitation sites.

library("viridis")
## Warning: package 'viridis' was built under R version 4.1.1
## Loading required package: viridisLite
## Warning: package 'viridisLite' was built under R version 4.1.1
par(mar=c(11,5,2,2))
Xlocations <- barplot(Abundance_at_Control_Sites, beside=T, names.arg=c(names(Species_Names)), ylab="Abundance", xlab="", main="Graph1: Abundance of Ants - Control Sites", ylim=c(0,1500), las=2, cex.names=0.8, col="magma" (5))    
text("<-Nylanderia obscura", y=300, x=33)
text("<-Aphaenogaster longiceps", y=1400, x=12)

Xlocations <- barplot(Abundance_at_Mine_Sites, beside=T, names.arg=c(names(Species_Names)), ylab="Abundance", xlab="", main="Graph 2: Abundance of Ants - Mine Sites", ylim=c(0,1500), las=2, cex.names=0.8, col="magma"(5))  
text("<-Nylanderia obscura", y=1400, x=33)
text("<-Aphaenogaster longiceps", y=300, x=12)

In the ANOVA below we can observe that…..

In both of these graphs we can see that Aphaenogaster longiceps and Nylanderia obscura are amongst the most common abundant species across all sites. Aphaenogaster longiceps is more common in control sites while Nylanderia obscura is more common in the Mine Sites. The “overabundance” of Aphaenogaster longiceps could be explained by ant interspecific competition (Majer, 1985)

Ants<- read.csv("Ants.txt")  
attach(Ants) # make 'Ants' the default data set 
## The following objects are masked from Ants_2015_2019:
## 
##     Rehab_Year, Rep, Rep_Year, Samples, Site, Treatment
fit1 <- lm(Abundance ~ Diversity)  # fit the regression line of best fit through the data 


fit2 <- lm(log(Abundance+1) ~ Diversity) # try log-transforming 'Abundance' 

plot( Diversity,log(Abundance+1), pch= 3 ) # plot log-abundance against "Diversity"

Here we can observe that Ant Species Diversity is correlated with increased Abundance. However looking at the previous graphs, diversity is not uniformly represented, with several ants “monopolozing” the diversity.

Aphaenogaster longiceps and Nylanderia obscura

For our two most common species we can see that the majority of observations lean towards zero, hinting that despite their supposed abundance, the records for species is quite small.

ANOVA

In the ANOVA table below we an observe that there is a highly significant interaction between the treatment and year of rehabilitation.

Ants<- read.csv("Ants.txt")  
attach(Ants) # make 'Ants' the default data set 
## The following objects are masked from Ants (pos = 3):
## 
##     Abundance, Diversity, Rehab_Year, Rep, Rep_Year, Samples, Site,
##     Treatment
## The following objects are masked from Ants_2015_2019:
## 
##     Rehab_Year, Rep, Rep_Year, Samples, Site, Treatment
full_model <- lm(log(Abundance+1) ~ Treatment * Rep_Year * Rehab_Year )
anova(full_model)
## Analysis of Variance Table
## 
## Response: log(Abundance + 1)
##                                Df Sum Sq Mean Sq F value    Pr(>F)    
## Treatment                       1   1.46   1.460  1.3883  0.239054    
## Rep_Year                        2  27.23  13.614 12.9456 2.955e-06 ***
## Rehab_Year                      3  16.63   5.545  5.2724  0.001327 ** 
## Treatment:Rep_Year              2  22.30  11.150 10.6024 2.870e-05 ***
## Treatment:Rehab_Year            3  95.60  31.867 30.3026 < 2.2e-16 ***
## Rep_Year:Rehab_Year             6  14.64   2.441  2.3209  0.031521 *  
## Treatment:Rep_Year:Rehab_Year   6  14.12   2.353  2.2375  0.037897 *  
## Residuals                     767 806.60   1.052                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Conclusions

Being a draft, this report definetly needs improvement, hwoever some quick conclusions can be made. Firstly, Ant diversity improves in control sites, indicating that rehabilitation efforts are not as useful. Secondly, time influences diversity. This indicates that ant ecology and/or behaviour influences diversity and abundance on sites (Majer, 1985)

References

MAJER, J.D. (1985), Recolonization by ants of rehabilitated mineral sand mines on North Stradbroke Island, Queensland, with particular reference to seed removal. Australian Journal of Ecology, 10: 31-48. https://doi.org/10.1111/j.1442-9993.1985.tb00861.x