Category: Post 6: Data Collection

Blog 6: Data Collection

I collected samples in Muir Creek, an old growth coastal rainforest west of Victoria. The terrain is very rough but there is a path through the site. I used the path as a transect line and sampled in four 30x30m quadrants space 100m apart along the line using a GPS. In each quadrant I located all stumps and logs hosting vascular growth and compared to the growth on the ground nearby. I quickly realized that just counting plants around stumps was not a fair comparison density wise. I measured the diameter of each stump and log, looked for the densest group of plants on the ground and superimposed the surface area of the stump on that spot. That way I could determine if plant growth on the ground was more or less dense than on the stump. I found a total of 21 samples. Only problems were minor. Every visit it was raining quite heavily in the rainforest, which made note taking difficult. I brought a camera to supplement the notes. I discovered previously that this very helpful as long as the picture file numbers are carefully recorded in the notes with each object. Otherwise I would have a couple hundred pictures of stumps that all look the same which would be completely useless.

Data collection thus far has been very enjoyable. I am investigating the species distribution of six species, Carex praegracilis, Andropogon gerardii, Gymnocarpium dryopteris, Elymus repens, Cyperus odoratus, and Sonchus arvensis, along the environmental gradient present at Milliken District Park, and am expecting to see a greater distribution of species along the gradient, as there is more exposure to sunlight.

In order to collect data, transect lines and square quadrats will be used to outline the study sites. I used survey poles and a line level in order to measure the distance and elevation.I’ve decided to collect data on percentage coverage, abundance, and presence/absence. Percentage coverage was measured using a 1m² quadrat, the abundance of the species using a 0.5m² quadrat, and absence/presence using a 0.25m² quadrant. The measurements were determined based on the amount of plants, as well as the size of the plants present in the area selected. Each of the three quadrats was placed randomly five times at each site, and data was collected. Abundance was measured using the ACFOR scale (Abundant, Common, Frequent, Occasional, or Rare). A species is considered abundant if it was present 10 or more times within the quadrat. It is considered common if it was present 7-9 times within the quadrat, it is considered frequent if it was present 5-6 times within the quadrat, it is considered occasional if it was present 3-4 times within the quadrat and it is considered rare if it is present 2 or fewer times within the quadrat. When looking at the variable of presence/absence, an ‘X’ represents the species that were present, and no ‘X’ represents that the species was absent in that region. I initially decided to repeat this procedure on three different days, but decided that 5 data sets for each study site on one collection date would be sufficient to obtain the information needed.

The data collected on percentage coverage, abundance, and presence/absence all seem to follow a similar trend. To some extent, my hypothesis and prediction has been supported and proven upon data collection and analysis, however, other factors that impact these plant species directly such as the climate and condition in which the species are found, have been seen to play a more essential role in the distribution.

Although I did not face any major difficulties implementing my sampling design, some minor issues that I was able to overcome include challenges with the weather and physically choosing the correct area to collect my sample. However, after these issues were overcame, I was able to successfully gather results.

The data collection was performed at three city parks: Riverside Park, McDonald Park and McArthur Island Park. The counting would start at 10 a.m. and would last 1 hour where 5 counts were done 20 minutes apart. This procedure was conducted for all of the sites. Both Riverside Park and McArthur Island Park contained two flowerbeds, but the whole territory of McDonald Park contained only one. This meant that haphazard sampling will be used and sites are subjectively chosen as no other opportunity of pollinators observation was present. Therefore, no problems appeared in implementing sampling design. On the duration of five non-continuous days’ counts were done with rotation of order of sites visited. The time of 10 a.m. was chosen by a recommendation of local beekeeper who stated that at 10 a.m. to 1 p.m. I could observe adult individuals and then from 2 p.m. to 5 p.m. I could observe the training flights of the young. Rotation of times for visit was implemented in order to avoid errors occurring because some species could appear at the site at later times of the day. Species observed were counted and described in the field journal and later identified using COMMON POLLINATORS OF BRITISH COLUMBIA Visual Identification Guide that could be found at http://borderfreebees.com/wp-content/uploads/2017/02/common-pollinaotrs-of-bc-v40.pdf

The only problem I found is that sometimes description I provide in the field is not specific enough to be identified as a single species but fits a few, and therefore will be excluded from the further analysis unless representing unique species. Few additional patterns were noticed during data collection. Wasps appeared to be dependent on the outside temperature as encounters were observed only at the hottest timeslots. No wasps were observed in a suburb McDonald Park even though it has almost no shade by big trees. The gardener who was servicing the park told me that it is very common for local wasps to build their nests under the rooftops of the houses which could explain their absence in this park.

Through extensive research of information online and gardening books from my parents I found out the exact amount of water each tree should be receiving and what counts as “excessive” and “the right amount”. In my study, the plum tree is the moderately watered fruit tree, in which it gets 8 liters of water per week. Ideally, a plum tree should be watered twice a week, which means each watering day the tree gets 4 liters.

The pear tree on the other hand gets less water, if it were the moderately watered tree. The right amount of water is 4 liters per week and watering twice a week, so 2 liters of water per watering. In this study, the pear tree is the “excessively” watered tree so I am watering it the same amount as the pear tree, which is 4 liters twice a week.

The cherry tree does not receive any watering, although would normally be maintained at 4 liters twice a week.

I choose two nights per tree to water them, but never both trees on the same night. Each tree expands an area of 1 square meter, so my quadrats are split into 9 sections, and fruit is counted individually in each section. So far I have done 2 replicates, as the fruit is just starting to produce and grow. I plan on collecting data until August to get the maximum yield of fruit, so 10 replicates altogether. In the end I will do a final count of fruit on each tree to complete my study and most likely just use that number for my conclusion. The overall number throughout the whole study is mainly just for my curiosity.

So far my sampling design has been working out for me. If I were to redo this experiment I would net the trees so birds couldn’t pick at the fruit and potentially disrupt a true fruit yield. Patterns I have noticed come from more the leaves of the tree than what the fruit looks like, although it might yet be too early to tell how this is going to affect the fruit itself.  The cherry tree and the pear tree both have unhealthy looking leaves that curl up, with some turning reddish/brown and wilting. The plum tree has healthy looking dark green leaves. In comparison to last year, the cherries growing on the tree don’t seem to be as abundant this year, but that can be due to other variables I haven’t accounted for in this study.

Cherry: no watering- drought

Plum: normal watering- control

Pear tree: excessive watering- too much water stress

My data collection has come with a few hardships to over come because of the change of seasons, we are currently going into our dry season. The change has brought about colder weather and more overcast days, we have also had an unusually high amount of rain for this time of year. This late rain has kept the natural water pans full and productive. The grass in the southern areas of the Timbavati is still green whilst in the north where I am is a far drier. From my observations it seems like the breeding herds of Elephants have stayed in the South to utilize the greener food.

To try and maximize my data I can collect I have chosen 10 replicates or water holes to monitor this has allowed me to monitor both the Northern and Southern areas of our traversing area. I have set the cameras up generally durning the mid-day as it is when I have time off in between game drives.  They then spend 5 days and night at the pre-determined water hole. Each camera is set to capture a photo at 15 minute intervals, each camera also takes a picture if the motion sensor is triggered. Generally an animal will trigger a photo at about 15 meters from the camera so I try to pick an area closest to the water that gives the biggest field of view to maximize photos taken. One thing to take into account is the curious hyenas who have on one occasion tasted my one camera trap ( it is still in one piece, barely) and on the other occasions come for a quick sniff.  On 11  different occasions I have had only Elephant bulls coming down to drink while in the month my cameras have been up I have not counted one breeding herd of Elephants. I am however starting to see a pattern as to when the Elephants, bulls in particular come down to drink. It seems as the heat of the day does have influence when the Bulls come down to drink rather than the time of day. On most occasions the Elephants have drank from 17 -20 Degrees Celsius  which seems to be the preferred temperature for the Bulls to drink irrelevant to the time of day. This ancillary pattern shows that the bulls as unfortunately that is all I have been able to capture on traps do have a preference as to what temperature they drink at. This could be assumed that the heat of the day does have something to do with their drinking activities

Most water holes that I have monitored have however been very quiet in terms of Elephant activity, Impala and Warthog, have been the far busiest at the water points.  This is not of any surprise as the Impala and the Warthog are water dependant species. I have also captured, lion, leopard, Hyena, buffalo, waterbuck, doves,  Egyptian geese, and mongoose. The Elephant bulls have favoured one dam much more than any other and in hind sight it may have been more beneficial to leave the camera traps at my first two water holes that I chose to study.

I sampled 10 replicates on each side of the staircase. My sampling strategy was straight forward and easy to implement. One difficulty I had was that I intended to randomly measure three projections at each sampling site; however, at nine of 20 sampling sites, no projections met my inclusion criteria (at least 30cm in length). This means that I have no data for these sites. This is relevant for my hypothesis that there are more projections on one side of the stairs compared to the other. Unfortunately, this also means that I am not able to calculate an average length of projections for these sampling sites, decreasing the amount of data I have to analyze. This means that I am less likely to find a difference between conditions as differences are difficult to detect in small samples, unless the difference is large. I believe that my hypotheses are sound, although there are other variables that could explain any potential differences in groups (e.g., amount of sunlight received, ground moisture and run-off) which I will have to address in my study limitations. Patterns in these variables may be of interest; however, I am unable to measure them.