How will changes in local climate affect hawksbill hatchling production in Brazil?

Local climatic conditions can influence sea turtle embryonic development and hatchling viability. Therefore, it is crucial to understand these influences as well as potential ramifications to population stability under future climate change. Here, we examined the influences of five climatic variables (air temperature, accumulated and average precipitation, humidity, solar radiation, and wind speed) at different temporal scales on hawksbill sea turtle (Eretmochelys imbricata) hatchling production at ten nesting beaches within two regions of Brazil (five nesting beaches in Rio Grande do Norte and five in Bahia). Air temperature and accumulated precipitation were the main climatic drivers of hawksbill hatching success across Brazil and in Rio Grande do Norte, while air temperature and average precipitation were the main climatic drivers of hatching success at Bahia. Solar radiation was the main climatic driver of emergence rate at both regions. Conservative and extreme climate scenarios show air temperatures are projected to increase, while precipitation projections vary between scenarios and regions throughout the 21st century. We predicted hatching success of undisturbed nests (no recorded depredation or storm-related impacts) will decrease in Brazil by 2100. This study shows the determining effects of different climate variables and their combinations on an important and critically endangered marine species.

likely that there is variability on how they can cope with different environmental conditions. This 68 indicates that local climate drivers of hatchling output need to be explored at a species and 69 nesting beach level. 70 To provide further insights into how sea turtle populations may be impacted by climate 71 change, we expand from previous studies and explore the influences of five different climatic 72 variables (air temperature, accumulated and average precipitation, humidity, solar radiation, and 73 wind speed) on hawksbill sea turtle, Eretmochelys imbricata, hatchling production from the 74 Southwest Atlantic Hawksbill Regional Management Unit on the coast of Brazil. This allowed us 75 to determine which variable(s) and combination of variables have the most influence on 76 hatchling production of this critically endangered species, to identify nesting regions that are 77 most susceptible to climate change, and to project future hatching success throughout the 21 st 78 century under extreme and conservative climate change scenarios.

80
Study site and species 81 We focused on the hawksbill sea turtle (Eretmochelys imbricata) population that nests 82 along the coast of Brazil. This population is part of the Southwest Atlantic Hawksbill Regional 83 Management Unit (RMU) [39]. In Brazil, hawksbills nest at two major nesting regions: southern 84 Rio Grande do Norte (RN) and northern Bahia (BA) [40], which represent the spatial extent of 85 the present study (Fig 1). We used data from five beaches in RN: Cacimbinhas,Chapadao,86 Madeiro, Minas, and Sibauma and five beaches in BA: Arembepe, Busca Vida, Imbassai, Praia 87 do Forte, and Santa Maria (Fig 1). Although the majority of hawksbill nesting occurs in BA, 88 42% of hawksbills are hybrids with loggerheads and 2% are hybrids with olive ridley sea turtles [41]. Despite having fewer hawksbill nests, RN has the highest density of nests per kilometer in 90 the South Atlantic Ocean, with some areas experiencing 48.5 nests per kilometer per season [42]. 91 The typical nesting season for hawksbills in RN is November -May, whereas in BA it is 92 October -April [42,43].  through the year 2100 due to predicted future stabilization of greenhouse gas concentrations.
127 Table 2. Available nest and climate data between regions.   [15,28,38,45]. Therefore, accumulated precipitation, average precipitation, and humidity 160 predictors were combined with average air temperature during incubation to explore their 161 combined effects on hatchling production. Precipitation projections were presented as mm/day.

162
Therefore, to project accumulated precipitation, these values were converted into mm/month.

164
Hatchling production (October -May) (Fig 3). RN is the warmest of the two regions being closer to the equator than 201 BA (Fig 3a). In RN, the warmest month is February ( May, there is no nesting in BA and nesting is finishing in RN, but hatchlings are still incubating 208 and emerging (Fig 3C). RN has more solar radiation than BA, likely due to its proximity to the 209 equator ( Fig 3D). The month with the highest average solar radiation in RN is October (1907.7

210
KJ/M 2 ± 198.7), while in BA it is January (1754.2 KJ/M 2 ± 158.2) ( Fig 3D). In RN, there is no 211 nesting in October, when solar radiation is highest (Fig 3d). On the other hand, January is when 212 solar radiation and nest proportions are at their highest in BA (Fig 3D). BA is more humid than 213 RN and May is the most humid month in both RN (79.2% ± 2.05) and BA (80.3% ± 5) ( Fig 3E).

214
In May, nesting has ended in BA and is ending in RN, but hatchlings are still incubating and 215 emerging ( Fig 3E). RN is windier than BA, with October being the windiest month in both RN

216
(5.29 m/s ± 0.5) and BA (1.82 m/s ± 0.34) (Fig 3F). There is no nesting in October in RN,while 217 October has the lowest proportion of nests in BA (Fig 3F).  Table). Emergence rate decreased with increasing solar radiation (Fig 4B).

233
Across RN, the model with the lowest AICc and high significance for hatching success 234 was average air temperature during incubation in combination with average precipitation during 235 the month nests were laid (p < 0.001 for both parameters; Fig 4C, S3 Table). This model 236 indicated that warmer and drier conditions decreased hatching success (Fig 4C). The model with 237 the lowest AICc and high significance for emergence rate across RN was average solar radiation 238 during incubation (p < 0.001) (Fig 4D, S3 Table). Here, higher solar radiation decreased 239 emergence rate (Fig 4D).

240
The model with the lowest AICc and high significance for hatching success across BA 241 was average air temperature during incubation in combination with accumulated precipitation 242 during the month nests are laid and 2 months prior (p < 0.001 for both parameters; Fig 4E, S3 243 Table). This model indicated that warmer and drier conditions decreased hatching success (Fig   244   4E). For emergence rate across BA, the model with the lowest AICc and high significance was average solar radiation during incubation (p < 0.001; Fig 4F, S3 Table). Here, higher solar 246 radiation decreased emergence rate (Fig 4F). at 29.3°C, but the warmest month under RCP8.5 is projected to be January at 32°C (Fig 5C).

259
Accumulated precipitation is projected to vary, but there is a general increase of 0.9 -22.3 mm 260 throughout the nesting season in BA (Fig 5D). May is projected to be the wettest month by 2100 261 under RCP4.5 at 185.7 mm and under RCP8.5 at 180.1 mm (Fig 5D).   Fig 6A). Hatching success in BA is projected to decrease from an average across 277 the study period of 76.3% to 65.1% under RCP4.5 and 69.6% under RCP8.5 (Fig 6B).  combined models are presented for each parameter in the order the model is written.