How well do Predators Adjust to Climate-Mediated Shifts in Prey Distribution? A Study on Australian Water Pythons

The spatial relocation of many species’ vital resources has been attributed to climate change. These resources are often other terrestrial prey species. Therefore, predatory species’ resilience and survival in the face of climate change depends on their ability to shift their activities away from unsuitable territories to the new ranges their prey inhabits. Ujvari et al. (2011) monitored two spatially separate groups of Northern Australian water pythons (Liasis fuscus) that were tracking the movements of their primary prey, the dusky rat (Rattus colletti), during the wet season. One population of pythons inhabited an area (Beatrice Hill) that contained a large population of rats. The second group of pythons inhabited Fogg Dam, an area that had recently experienced a crash in rat numbers due to a severe flood. The authors measured the population size, the survival rate, and the residual body mass (RBM) of both python groups and found that food (rat) availability was correlated with python RBM. Therefore, the pythons’ lower RBM at Fogg Dam was attributed to limited prey availability. As a result, the population’s survival rate and size were greatly reduced through starvation, despite the presence of a close, large population of rats at Beatrice Hill. These results suggest that python migrations were signaled by habitat features that indicate prey availability—a life strategy that could inhibit the adaption of pythons to climate induced prey range shifts.—Megan Smith

Ujvari, B., Shine, R., Madsen, T., 2011. How well do predators adjust to climate-mediated shifts in prey distribution? A study on Australian water pythons. Ecology 92: 3, 777 – 783.

          Ujvari et al. conducted their study in the Adelaide River floodplain, which lies in the Northern Territory of Australia within the “wet-dry” tropics (131°18’48.19”E, 12°34’14.81”S). The Adelaide River floodplain is a flat, treeless area containing sparsely distributed sedges and grasses. Its mean daily maximum air temperatures are greater than 30°C every month, and the region experiences a short wet season from December to March and a long dry season from April to November. A substantial amount of rain falls during the wet season. Seventy five percent of the 1440 millimeters of mean annual rainfall falls during these few months. The increase in water levels within the Adelaide River floodplain is usually gradual. However, in March 2007, 244 mm of rain fell within 24 hours, causing a massive flood down the river. Since the flood coincided with a high tide, the floodwater was unable to exit into the sea, resulting in an “inland tsunami” with a wave height of 1 m. The wave reached the authors’ study sites, and within a few hours, the entire river floodplain was completely underwater.
          The study species were non-venomous water pythons (Liasis fuscus), which grow up to 3 meters in length. The snakes primarily feed on the dusky rat (Rattus colletti), a native rat species. Pythons often migrate for distances up to 10 kilometers during the wet season to track the movements of rat populations avoiding rising water levels. Data on water pythons was collected between 1991 and 2009, between the months of August and December on a 1.3 km long dam wall at Fogg Dam Conservation Reserve. The snakes were identified and captured at night with a spotlight, and were marked by clipping the ventral scale. The snout-vent length, mass, and female reproductive status were recorded for each python captured. Prey records were assessed through fecal and palpation analyses. The snakes were released 24 hours after being captured.
          Dusky rat demographic data were collected during 5-day trapping periods every year in August. Since the floodplain at Fogg Dam was completely inundated, the researchers moved 5 km northeast to an area previously used for rice production (Rice Field). After assessing past data, the researchers found that rat numbers at Fogg Dam and Rice Field were highly correlated. Rice Field was also located within the pythons’ wet season migratory region. Therefore, the authors resumed collecting rat demographic data at Rice Field.
          The authors conducted additional snake trappings and rat demographic recordings at a second site, Beatrice Hill, after locating a large population of dusky rats inhabiting the area. The abundance of rats at Beatrice Hill contrasted to the virtual disappearance of rat populations observed at Rice Field. The rat population size at Rice Field plummeted after the large flood in 2007. Beatrice Hill was 8 km south of Rice Field and was situated near high ground.
          Ujvari et al. analyzed the data with standard statistical analyses methods. To quantify the among-year variation in python body mass, the authors calculated residual body mass (RBM) from a general linear regression comparing python mass and snout-vent length. The survival rates and population numbers of the Fogg Dam water pythons were assessed using the Jolly-Seber model.
          The authors found that the total numbers of rats captured at Rice Field varied among years, with rat populations dropping significantly after the “tsunami” in 2007. Only 7 rats were captured in 2007, none in 2008, and 4 in 2009. However, over 150 rats were captured at Beatrice Hill in 2009, which was higher than any rat numbers recorded during a trapping session throughout the 19-year study at Rice Field. A figure displaying the number of dusky rats captured at the Rice Field site from 1991 to 2009 and at Beatrice Hill in 2009 was constructed.
          In accordance with the rat data collected at Rice Field and Beatrice Hill, Ujvari et al. captured a higher proportion of freshly fed snakes at Beatrice Hill (47 of 72) than at Rice Field (16 of 124). The pythons captured at Beatrice Hill had fed entirely on dusky rats. However, the pythons at Fogg Dam were emaciated and had a broader diet that included rats, snakes, and lizards.
          The authors also found that python RBM varied among years and fluctuated with rat numbers. This suggests that the temporal variation in python RBM was driven by food (rat) availability. The mean RBM of pythons captured at Beatrice Hill in 2009 was significantly greater than the RBM of pythons captured at Fogg Dam throughout the researchers’ 19-year study. In contrast, the RBM of pythons in 2008 and 2009 at Fogg Hill were the lowest RBM numbers ever recorded during the study. The low prey availability and low RBM for pythons at Fogg Dam during 2008 and 2009 reduced the annual python survival rate by 42%, which was significantly lower than the survival rate of pythons (79%) between 1987 and 2003. Ujvari et al. also did not capture any reproductive females at Fogg Dam during 2008 and 2009, resulting in no recruitment of yearling pythons in 2009. A figure displaying the RBM of water pythons from 1991 to 2009 was constructed. A graph demonstrating the linear relationship between number of rats and python RBM was also constructed.
Python population size was estimated to be 2183 + 197 snakes in 2007, 1434 + 175 snakes in 2008, and 536 + 106 snakes in 2009. The population size in 2009 was the lowest number ever recorded during the study. Fogg Dam and Beatrice Hill also differed in their proportion of recaptured snakes. Of the 124 pythons captured at Fogg Dam in 2009, 33 were recaptures while all 72 pythons captured at Beatrice Hill were unmarked. This also indicates that the population size of Fogg Hill pythons was significantly lower than the population size of Beatrice Hill pythons. 
          These results demonstrate broad patterns of floodplain inundation, and rodent and snake responses to that inundation throughout the long-term study. The authors found that wet season rainfall increased the river’s water levels, causing flooding throughout the Adelaide River floodplain. Therefore, only the levee banks along the river channel remained above the water. Rats congregated on these high points to avoid drowning, and in response, the pythons migrated to the levee banks to feed on the rodents. However, the “tsunami” in 2007 submerged the entire floodplain, including the levee banks. Although the authors cannot prove the flood was a result of climate change, it nevertheless acted as an extreme weather event that could increase in frequency with global warming. The authors’ data suggest that this flood drowned the majority of the rats at Rice Field. Rats inhabiting regions along the floodplain edges survived by moving up to high ground, such as the high ground next to Beatrice Hill.
          Ujvari et al. claimed that the spatial divergence in rat abundance caused the observed differences in python feeding rates and RBMs between the two sites. Therefore, feeding conditions for water pythons were better at Beatrice Hill than at Fogg Dam after the tsunami decimated the rat population at Rice Field. As a result, the Fogg Dam pythons ate less often and consumed other prey like smaller snakes and lizards, which caused their emaciation. The authors were unable to find a direct relationship between the reduction in python numbers at Fogg Dam and starvation rates. Yet, the lack of snake recruitment, lack of rats, the highly emaciated pythons, the strong relationship between python RBM and survival, and the increased snake mortality rates between 2008 and 2009 suggest that increased mortality caused the decrease in python population numbers in 2009. However, despite the lack of rats in the area, the pythons remained at Fogg Dam instead of searching out more the more abundant rat population at Beatrice Hill.
           Ujvari et al. found that water pythons experienced a strong dry season philopatry. The pythons migrated great distances during the wet season in search of rats that moved to higher land to avoid flooding. Yet, the snakes always returned to their original territory when the dry season began since the water levels would usually subside, allowing large numbers of rats to return to the area. So, instead of searching out rat populations in different regions, the snakes returned to Fogg Hill because the beginning of the dry season had previously signaled the return of rats to their territory.
Overall, many predatory animals migrate to new locations by tracking prey abundance, and should therefore experience few difficulties moving to their prey’s new ranges. However, other species, such as the water python, track the movement of their prey through habitat features that signal prey availability. Thus, if changes in climate alter these habitat attributes, predatory species may be unable to shift their activities to new areas. Therefore, climate change may pose a greater threat for mobile species than previously asserted. 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s