Data used for this analysis was gathered using the LecoS plugin for QGIS. It provides a number of statistical measures for Landscape Ecological analysis, including patch statistics and zonal statistics.

patch <- readxl::read_excel("./Patch Statistics.xlsx")
zonal <- readxl::read_excel("./Zonal Diversity Statistics.xlsx")

The Zonal Diversity Statistics are already in a long format while the Patch statistics are not. Let’s change that.

head(zonal,3)
## # A tibble: 3 × 3
##   Metric         Value  Year
##   <chr>          <dbl> <dbl>
## 1 LC_Mean         4.20  1977
## 2 LC_Sum  365998919     1977
## 3 LC_Min          1     1977
head(patch,3)
## # A tibble: 3 × 21
##    Year Class    Land …¹ Lands…² Edge …³ Edge …⁴ Numbe…⁵ Patch…⁶ Great…⁷ Small…⁸
##   <dbl> <chr>      <dbl>   <dbl>   <dbl>   <dbl>   <dbl>   <dbl>   <dbl>   <dbl>
## 1  1977 Mangrove  3.85e7  0.180   224506 2.58e-3       8 9.18e-8  1.64e7  370813
## 2  1977 Agricul…  3.32e6  0.0155   23262 2.67e-4       5 5.73e-8  1.78e6  136815
## 3  1977 Estuary   4.54e7  0.212   332280 3.81e-3      14 1.61e-7  4.39e7       1
## # … with 11 more variables: `Mean patch area` <dbl>, `Median patch area` <dbl>,
## #   `Largest Patch Index` <dbl>, `Fractal Dimension Index` <dbl>,
## #   `Mean patch shape ratio` <dbl>, `Overall Core area` <dbl>,
## #   `Like adjacencies` <dbl>, `Patch cohesion index` <dbl>,
## #   `Landscape division` <dbl>, `Effective Meshsize` <dbl>,
## #   `Splitting Index` <dbl>, and abbreviated variable names ¹​`Land cover`,
## #   ²​`Landscape Proportion`, ³​`Edge length`, ⁴​`Edge density`, …
patch <- patch %>% pivot_longer(cols = c("Land cover","Landscape Proportion",
                                         "Edge length","Edge density",
                                         "Number of Patches","Patch density",
                                         "Greatest patch area","Smallest patch area",
                                         "Mean patch area","Median patch area",
                                         "Largest Patch Index","Fractal Dimension Index",
                                         "Mean patch shape ratio","Overall Core area",
                                         "Like adjacencies","Patch cohesion index",
                                         "Landscape division","Effective Meshsize",
                                         "Splitting Index"),
                                names_to = 'metric',
                                values_to = "values")

Le’s begin our analysis by looking at change in Land Cover by class and year.

As can be seen in the figure, the land covered by agriculture increased exponentially between 1977 and 1990. During the same period land cover for mangroves and estuary decreased with area covered by mangrove decreasing the most. Between 1990 and 2004 both agriculture and mangrove areas experienced in increase in land cover leading to a decreasing estuary land cover. After 2004 the estuary continued to loose land cover while agriculture also commenced to decrease in land cover. The continued decrease in land cover by estuary signals that expansions in land cover by the other two classes present in the landscape in some measured occurred by expanding into estuary areas, particularly after 1990. Previous to 1990, areas covered by mangrove were also altered for agriculture purposes. For reasons unknown at the moment, after 1990 the need for more agriculture continued but expansion occurred toward estuary areas. The mangrove ecosystem also expanded after 1990, again in detriment of the estuary. Similar results can be shown by looking at changes in landscape proportions among the three classes between 1977 and 2011.

Landscapes are areas of interacting ecosystems formed through geomorphological processes and disturbances. Those resulting ecosystems can be classified into patches with specific assemblages within a matrix. In 1977 the landscape stretching from the Caráquez Bay into Ecuador showed little amount of patches, signaling a landscape with ecosystems stretching along an area without generating edges between them. As shown in the figure bellow, by 1990, all classes had experienced growth in the number of patches. Estuary experienced a dramatic growth in number of patches signaling that the land cover growth, particularly by agriculture, discussed above, was breaking estuary patches and creating edges. Edge length increased for all classes between 1977 and 1990 and decreased by 2004 with estuary decreasing the most. Mangrove experienced the largest increase in edge length after 2004 followed by agriculture and estuary. Nevertheless, the amount of patches for mangroves experienced relatively no change between 2004 and 2011. In fact, the only class to experience an increase in number of patches after 2004 was agriculture. This may signal that the different ecosystems in the landscape may have been becoming more cohesive, spanning larger areas as a single patch instead of being separated into multiple patches.

As shown by the Patch cohesion index, estuary remained relatively cohesive during the study period. The growth of agriculture, particularly between 1977 and 1990 signified an increase in cohesion that could be tied to a decrease in cohesion among mangrove patches. After 1990 the amount of patches for all classes decreased sharply leading to higher cohesion indices. Results from the Like Adjacencies metric show a disaggregation of mangrove patches between 1977 and 1990. After 1990 mangrove areas experienced a sustained increase in Like Adgacencies. Values for estuary and agriculture varied little between 1977 and 2004 but decreased sharply after. The dramatic decrease in the values of this metric for estuary signals that the changing dynamics of agriculture and mangroves after 2004 were impacting how connected estuary areas were.

The results of the statistical analysis carried out in QGIS using the LecoS plugin show that the growth in agriculture between 1977 and 1990 altered the ecosystem, generating larger amounts of patches among mangrove and estuary ecosystems. After 1990, the dynamics impacting this landscape changed and the mangrove ecosystem increaced in size while, along with the agriculture ecosystem, also growing more cohesive. The growth in size and cohesiveness of these two ecosystems meant that, although the number of patches for estuaries remained low, its presence throughout the landscape became sparser.