Jessica Lin
17 November 2008
Geography 491
Lab #5: Plant Invasion and Land Use in the Galapagos Islands
(1) Questions:
For the agricultural zone
• What relationships exist between farm characteristics (in the attribute table) and the percent guava by farm?
Larger numbers of cattle, crops, and pasture are shown to decrease the percentage of guava by field, as evidenced by the negative trends in graphs 1, 2, and 3 below. The correlation between cattle ownership and the percentage of guava by farm is surprising, given the notion that cattle were originally thought to contribute to the dispersal of guava. This finding suggests that native vegetation may be more prone to guava invasion. Clearing farms for cattle, crops, and pasture may prove to be essential for reducing the likelihood of guava invasion. Since cattle trample the native vegetation and soil, they may help to keep the land clear, thus also explaining the correlation between land used for pasture and percentage of guava by farm.
Graph 4 shows the “Other” category, pertaining to activities such as tourism and fishing on the farms. Not surprisingly, commercial activity on farms such as fishing and tourism, correlate to greater percentages of guava on farms, at averages of 65% and 50%, respectively. Farms with neither of these activities have an average of 30% guava invasion. This finding supports the assumption that tourists contribute to the spread of guava. I might also make the assumption here that cattle crush the guava seeds under their significantly larger body masses or the defecation process helps to protect the soil from guava invasion, while the humans are more apt to spread the seeds with their lighter tread.
Graph 1: jslin_graph1_farms_cattle.jpg
Graph 2: jslin_graph2_farms_crops.jpg
Graph 3: jslin_graph3_farms_pasture.jpg
Graph 4: jslin_graph4_farms_other.jpg
• What, if any, proximity relationships exist between percent guava by field and (a) roads and (b) buildings?
See map: jslin_Lab5_Map1.jpg
Map 1 supports the conclusion that areas near roads and buildings are likely to have less guava invasion. It is not surprising, since the surrounding areas are most likely cleared to make room for construction. However, it seems as if the presence of buildings has a greater impact on smaller percentages of guava in the surrounding areas than the presence of roads (farm fields 49 and 42 contain large amounts of guava, but have direct access to roads).
• Assign each farm a category of “abandoned,” “partially abandoned,” or “active” based on percent guava present and the farm characteristics in the attribute table. Are fields or farms next to abandoned fields/farms more likely to contain a higher percentage of guava?
See map: jslin_Lab5_Map2.jpg
Map 2 shows the abandoned farms highlighted in blue. Farm 13 is bordered by two abandoned farms (16 and 14), and contains only 12.52% guava. Farms 1 and 14 both contain large percentages of guava (64.81%, and 81.15%, respectively). Farm 1 is not located near any of the abandoned fields and is considered active, while Farm 14 is considered an abandoned field. As we can see by examining each farm individually, there is no consistent evidence to support the assumption that abandoned farms, or farms near abandoned fields, contain a higher percentage of guava.
However, if we take into account the averages for farm activity, we can see results to support the previous assumption. Graph 5 illustrates a summary of averages between the average percentage of guava by farm and the average count of farm activity, rated by increases in activity on the farm. I added the percentages of crops and pasture on each farm to determine farm activity. Farm activity that totaled 50% and above were classified as “active”, farm activity that totaled under 40% were classified as “partially abandoned”, and farm activity that totaled under 20% were classified as “abandoned”. Graph 5 indicates that a greater amount of farm activity leads to a decrease in the percentage of guava.
Graph 5: jslin_graph5_farms_category.jpg
• Is there spatial clustering of guava in the agricultural zone or in individual fields, and does this spatial clustering provide clues to land use practices that could affect guava distribution?
See map: jslin_Lab5_Map3.jpg
The five farms with the lowest percent of guava invasion have a mean of 30 cattle, 24% of farm land allocated to crops, and 64% of farm land allocated to pasture. Compared to all farms, which have a mean of 18 cattle, 15.625% of farm land allocated to crops, and 43.75% of farm land allocated to pasture, we can see that the increase in all three of these factors aid in preventing the spread of guava.
For the Park zone
• Is there a relationship between guava in the Park and distance to the agricultural zone?
See map: jslin_Lab5_Map4.jpg
Higher numbers are seen in guava distribution for the areas closer to the park. These numbers decrease significantly as the boundaries move further away from the agricultural zone. This finding supports the assumption that clearing land and agricultural land use is directly related to guava invasion.
Graph 6: jslin_graph6_park_guava.jpg
• Is there a relationship between guava distribution in the Park and percent guava in fields
bordering the park?
See map: jslin_Lab5_Map5.jpg
Map 5 shows the allocation of guava pixels in the park to the nearest field in the agricultural zone. As we can see, the majority of guava pixels surrounding the agricultural zone within a 500 meter boundary are allocated directly from the farm fields, illustrating a major influence of agricultural land use on guava invasion. The level “0”, denoted by the color black, illustrates the overall spread of guava in the Park, indicating that there may be other factors not accounted for in this data set that are responsible for the dispersion of guava. Wind and soil composites could possibly play a important roles.
I would have made an assumption that fields cleared recently would be less likely to have a high percentage of guava invasion. However, the clearing dates do not appear to have any correlation with the count of guava by field.
The mean of the allocated park guava pixels is 19,079.31, which is significantly lower than the count for fields containing areas used for pasture (21,073.91) or crops (22,876.57). This finding adds further proof to the assumption that field proximity directly influences the spread of guava in the Park.
• How about between guava distribution in the park and characteristics of farms bordering the Park?
Graphs 7, 8, and 9 show the relationship of the park guava pixel count in bordering farms to factors such as cattle ownership, crops, and pasture. Increases in land used for crops and pasture correlates with a decrease in park guava pixel count as evidenced in Graphs 8 and 9. However, Graph 7 indicates a contradictory finding from previous results. In farms bordering the park, a larger number of cattle correlate to an increase in park guava pixel count. Previously, we had seen a decrease in percent guava for all farms with a larger number of cattle (see Graph 1: jslin_graph1_farms_cattle.jpg). It may be assumed that bordering farms also contain more natural vegetation, which may be more susceptible to guava invasion by means of cattle defecation, which does not seem to affect cleared farms, as shown by previous findings.
Graph 7: jslin_graph7_park_cattle.jpg
Graph 8: jslin_graph8_park_crops.jpg
Graph 9: jslin_graph9_park_pasture.jpg
(2) SUMMARY:
Factors that contribute heavily to guava invasion on farms and fields in the Galapagos islands are low agricultural activity, high commercial activity such as tourism and fishing, low numbers of cattle ownership, and low percentages of areas being used for pasture and growing crops.
These factors seem to be contradictory to the Park findings, which show that the percentage of guava found in the park directly correlates to the distance from the agricultural zone. While cattle are useful in keeping the percentage of guava down on fields, they appear to contribute to the spread of guava in Park areas surrounding bordering farms. However, the amount of land used for crops and pasture in bordering farms has the same effect on the spread of guava in the Park as on the farms. Larger amounts of the fields used for these purposes in all farms and in bordering farms correspond to lower percentages of guava.
These discrepancies between the spread of guava in farms and in the Park may be explained by other significant environmental and ecological factors not included in the data set.