Lemur catta tutorial

=== Downloading the map of Madagascar

```{r}
country_map <- ne_countries(country = "Madagascar",
                            scale = "medium",
                            returnclass = "sf")

image::data-use::Taxon_key.png[align=center,width=900,height=125]

Looks like we have a good matched accepted name here. The `usageKey` is the key that identify the usage of a name in the https://www.gbif.org/dataset/d7dddbf4-2cf0-4f39-9b2a-bb099caae36c[GBIF Taxonomy Backbone]. Once we decide which 'usage' is what we are looking for, We will use the key as the `taxonKey` for specifying the taxon for its occurrence data.

The following code block store the usageKey of _Lemur catta_ in a variable `usageKey`.

```{r} usageKey <- 2436412 # Alternatively, since we know there is only one match, the following line will store the key in the variable, too. # usageKey <- name_backbone(name = "Lemur catta", rank = "species") %>% pull(usageKey) ```

=== Use the taxonKey to generate the occurrence download of the species of interest
The following code block sends a request for download to the GBIF Occurrence API. A successful request will return a download key and initiate the preparing of the requested data. Note that as we've stored the key in the `usageKey` variable, for the required `taxonKey` for occ_download(), we simply supply the variable. Also, we need to supply our GBIF.org login credentials to request a download.

```{r}
gd_key <- occ_download(
    pred("taxonKey", usageKey),
#   pred("hasCoordinate", TRUE),
    format = "SIMPLE_CSV",
    user = "USERNAME",
    pwd = "PASSWORD",
    email = "EMAIL_ADDRESS"
)

gd <- occ_download_get(gd_key, overwrite = TRUE) %>%
    occ_download_import()

head(gd, 10) # view first ten rows ```
image::data-use::download.png[align=center,width=900,height=250]

### Citing GBIF data Always cite the downloaded dataset in your report. The following code helps generate the citation.

```{r} print(gbif_citation(occ_download_meta(gbif_download_key))$download) ```

== Data Visualization
_Lemur catta_ is native to Madagascar; but just to make sure, let's verify the data by plotting occurrences on a map.

```{r, message = FALSE, error = FALSE}
ggplot() +
  geom_sf(data = world_map) +
  geom_point(data = gd,
             aes(x = decimalLongitude,
                 y = decimalLatitude),
             shape = "+",
             color = "red") +
  theme_bw()

```{r} table(gdf$countryCode) ```
image::data-use::countries.png[align=center,width=600,height=75]

It appears that many records have coordinates outside Madagascar. Let's also have a look at the nature of these records.

```{r} gdf %>% distinct(basisOfRecord) gdf %>% distinct(establishmentMeans) ``` There are 5 distinct values for https://dwc.tdwg.org/terms/#dwc:basisOfRecord[DwC:basisOfRecord]. We also looked at https://dwc.tdwg.org/list/#dwc_establishmentMeans[dwc:establishmentMeans], where only `NATIVE` is noted for some records.

== Data cleaning

After some exploring of our data, we know that there are potential quality issues in our download. Apparently points outside Madagascar are suspicious, and we have just looked at basisOfRecord and establishmentMeans columns for cues of needed data actions.

**Data cleaning** typically involves procedures to remove unwanted records based on some criteria, or to correct values to achieve overall operable consistency. In the following section we will try to filter data, show the difference, and plot it on the map.

== Step 1: basisOfRecord

We would like to evaluate observations and collections, so `FOSSIL_SPECIMEN` and `MATERIAL_SAMPLE` is not our concern here, let's try to exclude them from our download.

```{r}
clean_step1 <- gdf %>%
  as_tibble() %>%
  filter(!basisOfRecord %in% c("FOSSIL_SPECIMEN", "MATERIAL_SAMPLE"))

print(paste0(nrow(gdf)-nrow(clean_step1), " records deleted; ",
             nrow(clean_step1), " records remaining."))

ggplot() +
  geom_sf(data = world_map) +
  geom_point(data = gdf,
             aes(x = decimalLongitude,
                 y = decimalLatitude),
             shape = "+",
             color = "black") +
  geom_point(data = clean_step1,
             aes(x = decimalLongitude,
                 y = decimalLatitude),
             shape = "+",
             color = "red") +
  theme_bw()

ggplot() +
  geom_sf(data = world_map) +
  geom_point(data = gdf,
             aes(x = decimalLongitude,
                 y = decimalLatitude),
             shape = "+",
             color = "black") +
  geom_point(data = clean_step2,
             aes(x = decimalLongitude,
                 y = decimalLatitude),
             shape = "+",
             color = "red") +
  theme_bw()

ggplot() +
  geom_sf(data = country_map) +
  geom_point(data = gdf,
             aes(x = decimalLongitude,
                 y = decimalLatitude),
             shape = "+",
             color = "black") +
  geom_point(data = clean_step2,
             aes(x = decimalLongitude,
                 y = decimalLatitude),
             shape = "+",
             color = "red") +
  coord_sf(xlim = st_bbox(country_map)[c(1,3)],
           ylim = st_bbox(country_map)[c(2,4)]) +
  theme_bw()

ggplot() +
  geom_sf(data = country_map) +
  geom_point(data = gdf,
             aes(x = decimalLongitude,
                 y = decimalLatitude),
             shape = "+",
             color = "black") +
  geom_point(data = clean_step3,
             aes(x = decimalLongitude,
                 y = decimalLatitude),
             shape = "+",
             color = "red") +
  coord_sf(xlim = st_bbox(country_map)[c(1,3)],
           ylim = st_bbox(country_map)[c(2,4)]) +
  theme_bw()