Introduction

There are numerous factors that define the vessel traffic construct of a maritime port. For the Port of Corpus Christi, Texas, these factors include the port being the third-largest oil export port, and being a premier fishing destination just to name a few. What other factors may effect the traffic through the port? Utilizing the Marine Cadastre database, we will consider if the size of a vessel’s draft is a predictor of the voyage duration of a vessel and if major events can be identified from teh vessel traffic.

Data

A tidied data set containing vessel type, draft, and duration information was utilized to conduct this study.

Load Packages

library(tidyverse)
library(here)
library(ggbeeswarm)
library(beepr)
library(RColorBrewer)

Read data

plot_vessels <- read_csv(here("data", "cleanAIS_new.csv"))
## Rows: 56591 Columns: 18
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (7): vessel_group, start, start_month, start_year, end, end_month, end_year
## dbl (9): draft, duration, start_day, end_day, duration_hrs, duration_days, l...
## lgl (2): longer_duration, longer_group
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
glimpse(plot_vessels)
## Rows: 56,591
## Columns: 18
## $ vessel_group     <chr> "Tanker", "Tanker", "Tanker", "Tanker", "Tanker", "Ta…
## $ draft            <dbl> 11.4, 13.7, 8.9, 9.9, 12.9, 12.9, 12.9, 12.9, 12.9, 1…
## $ duration         <dbl> 5233, 2588, 3175, 3615, 5744, 2430, 3871, 45, 51, 166…
## $ start            <chr> "2022-01-04 9:33:49", "2022-01-08 3:20:28", "2022-01-…
## $ start_day        <dbl> 2022, 2022, 2022, 2022, 2022, 2022, 2022, 2022, 2022,…
## $ start_month      <chr> "01", "01", "01", "01", "01", "01", "01", "01", "01",…
## $ start_year       <chr> "04", "08", "20", "23", "09", "13", "14", "17", "17",…
## $ end              <chr> "2022-01-08 0:47:40", "2022-01-09 22:29:08", "2022-01…
## $ end_day          <dbl> 2022, 2022, 2022, 2022, 2022, 2022, 2022, 2022, 2022,…
## $ end_month        <chr> "01", "01", "01", "01", "01", "01", "01", "01", "01",…
## $ end_year         <chr> "08", "09", "23", "25", "13", "14", "17", "17", "17",…
## $ duration_hrs     <dbl> 87.2166667, 43.1333333, 52.9166667, 60.2500000, 95.73…
## $ duration_days    <dbl> 3.63402778, 1.79722222, 2.20486111, 2.51041667, 3.988…
## $ logduration      <dbl> 8.562740, 7.858641, 8.063063, 8.192847, 8.655911, 7.7…
## $ logduration_days <dbl> 1.2903416, 0.5862423, 0.7906645, 0.9204487, 1.3835127…
## $ durationDiff     <dbl> 5224.43726, 2580.14136, 3166.93694, 3606.80715, 5735.…
## $ longer_duration  <lgl> TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, FALSE, FALS…
## $ longer_group     <lgl> TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, TRUE, FALSE, FALS…

Does Draft Indicate Voyage Duration

The relationship between vessel draft and voyage duration did not have a strong correlation and only showed a slight increase in voyage length with an increase in draft (Figure 1).

plot_vessels %>%
  na.omit() %>%
  filter(vessel_group %in% c("Cargo",
                             "Fishing",
                             "Other",
                             "Passenger",
                             "Pleasure Craft/Sailing",
                             "Tanker")) %>%
  ggplot(aes(x = draft, 
             y = logduration)) +
  geom_jitter(aes(color = vessel_group)) +
  geom_smooth() +
  scale_color_brewer(name = "Vessel Type", 
                     palette = "Dark2") +
  labs(title = "Correlation of Vessel Drafts and Voyage Duration", 
       caption = "Data from: https://hub.marinecadastre.gov/pages/vesseltraffic",
       x = "Draft (ft)",
       y = "Log(Duration)")
Figure 1 - Vessel draft was found to not correlate with voyage duration. Vessel drafts for all transitting vessels were plotted against the logorthmicly-scaled voyage duration and showed only a slight increase with draft.

Figure 1 - Vessel draft was found to not correlate with voyage duration. Vessel drafts for all transitting vessels were plotted against the logorthmicly-scaled voyage duration and showed only a slight increase with draft.

Can Vessel Traffic Patterns show Major Events

The next consideration was whether vessel transit patterns wer affected by major events. To begin, we considered the number of transits of vessels by month (Figure 2).

plot_vessels %>%
  filter(vessel_group %in% c("Cargo",
                             "Fishing",
                             "Other",
                             "Passenger",
                             "Pleasure Craft/Sailing",
                             "Tanker")) %>%
  ggplot(aes(x = vessel_group, fill = vessel_group)) +
  geom_bar() +
  facet_wrap(~start_month) +
  theme(axis.text.x = element_blank()) +
  labs(title = "Comparison of Vessel Transits by \n Type each Month", 
       caption = "Data from: https://hub.marinecadastre.gov/pages/vesseltraffic",
       x = "Vessel Type",
       y = "Total Transits",
       fill = "Vessel Type")
Figure 2 - Vessel Transits Counts by Vessel Type in each month. A noticible increase in fishing vessel transits can be identified in the months of July through September.

Figure 2 - Vessel Transits Counts by Vessel Type in each month. A noticible increase in fishing vessel transits can be identified in the months of July through September.

It became immediately apparent that there was a noticeable increase in fishing vessel transits from July through September. Was there an event that could be identified that might have occurred in this time frame? A focused look at fishing vesssel transits by month was necessary (Figure 3).

plot_vessels %>%
  na.omit() %>%
  filter(vessel_group == "Fishing") %>%
  ggplot(aes(x = start_month)) +
  geom_bar() +
  coord_polar()+
  labs(title = "Total Fishing Vessel Transits by Month", 
       caption = "Data from: https://hub.marinecadastre.gov/pages/vesseltraffic",
       x = "Month",
       y = "Total Transits")
Figure 3 - A Polar Plot of Fishing Vessel Transits by Month. An increasing trend beginning in June tjat peaks in July and decreases through October can be identified.

Figure 3 - A Polar Plot of Fishing Vessel Transits by Month. An increasing trend beginning in June tjat peaks in July and decreases through October can be identified.

From the resulting polar plot of the transit counts, an increase in transits can be found in June that peaks in July decreasing through October becomes clear. Looking through local fishing guides for events that occur around this timeframe, a prominent offshore shrimp fishing season can be found lasting from July 16 - Nov 30. With fish stocks being most likely at there highest at teh start of the season and quotas being hit throughout, the transit peak and subsiquent decrease fit into this time frame.

Conclusion

Vessel Traffic can provide a picture of the unique climate and construct of a port. Through this short study of the Port of Corpus Christi, it was concluded that vessel draft could not be used as a reliable predictor of vessel voyage duration. It was found that transit quantities could be impacted by external events with a significant impact to fishing vessels coming from the shrimping season. Further studies could be made into the impact of oil prices on tanker vessel transits or the average voyage duration of recreational vessels.