Body Condition Analysis of California Tiger Salamander

California Tiger Salamander (Ambystoma californiense).

Background and Data

In this report, I set out to compare weight-length relationships of California tiger salamanders (Ambystoma californiense, CTS) across 2012 - 2024 trapping years and 9 different trapping sites. This analysis follows the methodology described by Derek H. Ogle as he describes in his blog post Weight-Length Relationship for 3+ Groups on his website Introductory Fisheries Analysis with R. Weight-Length relationships are often used in fisheries management to compare body condition across time, among habitats, among bodies of water, and among length categories (Ogle & Winfield, 2009). For this analysis, weights (grams) and total lengths (mm) were recorded from 1,714 CTS captured at nine sites within the Concord Naval Weapons Station in Concord, CA, between 2010 and 2024. The original data was collected in field notebooks and later transcribed into Excel spreadsheets (associated meta data is provided in the README.md file within the data folder). The combined trapping data is stored in the dataframe cts. Below are some photos of the trapping site, pitfall traps, CTS data collection, and a CTS relocation into the nearby breeding pond.

Data collection on CTS sex, weight, snout vent length, total length, and location.

Pitfall trap and ERTEC fencing at trapping site.

Me releasing a CTS into the breeding pond.

Me holding a CTS.

Methods

Set up & Data Wrangling

To begin the analysis, I cleaned and combined the data from the 2010–2021 and 2022–2024 trapping datasets. The data was filtered to include only California tiger salamanders (Ambystoma californiense) and cleaned up to ensure consistency with site names and age class.

#load libraries
library(tidyverse)
library(here)
library(readxl)
library(janitor)
library(lubridate)
library(dplyr)
library(emmeans)  ## for emtrends()
library(FSA)
library(ggplot2)
library(viridis)
library(showtext)
library(kableExtra)

#load data
trap_data_2022 <- read_excel(here("posts/2025-03-14-pop-ecology-cts/data/clean_cts_data/Data_CTS-CRLF TrappingData_2023-2024.xlsx")) %>% 
  clean_names()

trap_data_2010 <- read_excel(here("posts/2025-03-14-pop-ecology-cts/data/clean_cts_data/Copy of CTS_2010-2021.xlsx")) %>% 
  clean_names()

#clean up data 
trap_2010 <- trap_data_2010 %>% 
  dplyr::select(date, site, species_6, species_7, class, weight, svl, tl) %>% 
  rename(species = species_6,
         scientific_name = species_7,
         age_class = class) %>% 
  mutate(weight = as.numeric(weight))

trap_2022 <- trap_data_2022 %>% 
  dplyr::select(date, site, species, scientific_name, age_class, weight_grams, sv_mm, tl_mm) %>%
  rename(svl = sv_mm, 
         tl = tl_mm,
         weight = weight_grams) %>% 
  mutate(svl = as.numeric(svl),
         weight = as.numeric(weight)) %>% 
  drop_na()

#combine data
trap_2010_2025 <- bind_rows(trap_2010, trap_2022)


#filter for CTS and standardize values 
cts <- trap_2010_2025 %>% 
  filter(species == "California tiger salamander") %>% 
  drop_na() %>% 
  mutate(
    # Clean age_class
    age_class = case_when(
      age_class == "adult" ~ "Adult",
      age_class == "Unk" ~ "Adult",
      age_class == "Juv" ~ "Juvenile",
      TRUE ~ age_class
    ),
    
    # Clean site names
    site = recode(site, 
                  "24A" = "Pistol Range",
                  
                  "Outside Eagles Nest" = "Eagles Nest",
                  
                  "100A" = "IA-100 Area A",
                  
                  "100B" = "IA-100 Area B",
                  
                  "100E" = "IA-100 Area East",
                  "IA-100 AreaE" = "IA-100 Area East",
                  "IA100E" = "IA-100 Area East",
                  "IA-100E" = "IA-100 Area East",
                  "IA-100 East" = "IA-100 Area East",
                  
                  "100N" = "IA-100 Area North",
                  "IA-100 AreaN" = "IA-100 Area North",
                  "IA100N" = "IA-100 Area North",
                  "IA-100N" = "IA-100 Area North",
                  "IA-100 North" = "IA-100 Area North",
                  
                  "100W" = "IA-100 Area West",
                  "IA-100 AreaW" = "IA-100 Area West",
                  "IA100W" = "IA-100 Area West",
                  "IA-100W" = "IA-100 Area West",
                  "IA-100 West" = "IA-100 Area West"),
                  
    
    # Extract year from date
    year = year(date)
  )

Weight-Length Relationship Analysis

In this analysis, I set out to determine if the mean log-transformed length of CTS varied across trapping years and trapping sites. Specifically, I hypothesize that the weight-length relationship of CTS would differ across years due to environmental factors or population dynamics. Additionally, I hypothesize that the weight-length relationship of CTS would also differ across trapping sites. To analyze the relationship between weight and total length, I first explored the data through scatterplots, visually inspecting the patterns and checking for any trends across the years and sites.

cts %>% 
  ggplot(aes(x = tl, y = weight)) + 
  geom_point(alpha = 0.5) + 
  facet_wrap(~year) +
  labs(title = "Weight-Length Relationship of CTS Across Years",
       x = "Total Length (mm)",
       y = "Weight (g)")

When comparing across years, we can see that the weight-length relationship of CTS varies over time, with some years showing a stronger relationship between total length and weight than others. This suggests that the body condition of CTS may be influenced by environmental factors or population dynamics that differ between years.

cts %>% 
  ggplot(aes(x = tl, y = weight, )) + 
  geom_point(alpha = 0.5) + 
  facet_wrap(~site) +
  labs(title = "Weight-Length Relationship of CTS Across Sites",
       x = "Total Length (mm)",
       y = "Weight (g)")

When comparing across sites, we can see that the weight-length relationship of CTS varies between different trapping locations, with some sites showing a stronger relationship between total length and weight than others. This suggests that habitat characteristics or local environmental conditions may influence the body condition of CTS at different sites in addition to comapring across years.

Statistical Modeling

I used linear regression models to assess the relationship between log-transformed weight and total length, including year as a categorical variable to evaluate differences over time. I log-transformed both the weight and total length variables to meet the assumptions of linear regression and stabilize variance. This allowed for easier interpretation of the relationships and comparisons across years and sites. The linear regression model was fitted using the formula:

log(Weight) = beta[0] + beta[1] * log(Total Length) + beta[2] * Year + beta[3] * (log(Total Length) * Year)

I then run a linear regression to assess the relationship between log-transformed weight and total length, including site as a categorical variable to evaluate differences across the 9 trapping sites. The model was fitted using the formula:

log(Weight) = beta[0] + beta[1] * log(Total Length) + beta[2] * Site + beta[3] * (log(Total Length) * Site)

Model Diagnostics

After fitting the model for both year and again for sites, I assessed the assumptions by examining residual plots and histograms of the residuals. These diagnostics indicated that the log-transformed data met the assumptions of normality and homoscedasticity, which supported the validity of the model.

#log transform data
cts <- cts %>%
  mutate(
    logW = log(weight),      # Log-transform weight
    logL = log(tl)           # Log-transform tl (total length)
  ) %>% 
  mutate(year = as.factor(year))

#create subset of the data for different years
cts <- cts |>
  filter(year %in% c(2012,2014, 2016, 2018, 2020, 2022, 2024)) |>
  droplevels()

# lm of log data across years
fit1 <- lm(logW~logL*year,data=cts)

par(mfrow=c(1,2))
# Scatter of residuals
plot(resid(fit1)~fitted(fit1),xlab="Fitted Values",ylab="Residuals")

# Histogram of residuals
hist(resid(fit1), xlab = "Residuals")

Normality Across Years. Residual plot (Left) and histogram of residuals (Right) from fitting a dummy variable (site) regression to the log-transformed weights and lengths. The residuals appear to be normally distributed, indicating that the model assumptions are met.

#summary(fit1)

# lm of log data across sites
fit2 <- lm(logW~logL*site,data=cts)

par(mfrow=c(1,2))
# Scatter of residuals
plot(resid(fit2)~fitted(fit2),xlab="Fitted Values",ylab="Residuals")

# Histogram of residuals
hist(resid(fit2), xlab = "Residuals")

Normality Across Sites. Residual plot (Left) and histogram of residuals (Right) from fitting a dummy variable regression to the log-transformed weights and lengths. The residuals appear to be normally distributed, indicating that the model assumptions are met.

#summary(fit2)

Comparing Slopes Across Years and Sites

To further explore the differences in the weight-length relationship over time and sites, I used the emmeans package to estimate means and compare the slopes across years and sites. This analysis provided pairwise comparisons of the regression slopes for each year and site, allowing me to identify statistically significant differences in the weight-length relationships.

Statistical Significance

I assessed statistical significance using p-values obtained from the Tukey test for pairwise comparisons of slopes across years and sites. Significant differences were identified where the p-value was less than 0.05. These results were then summarized in a table with highlighted significant contrasts for easier interpretation.

This methodology allows for a robust comparison of body condition across years and sites, while providing insights into temporal and spatial trends in the California tiger salamander population.

Results

Weight-Length Relationship Across Years

ANOVA Results

The ANOVA provides statistically significant results (p-value < 0.05) when comparing across trapping years. Therefore, at least 2 of the log-transformed mean lengths are statistically different across years.

car::Anova(fit1)

Anova Table (Type II tests)

Response: logW
           Sum Sq  Df  F value    Pr(>F)    
logL      181.620   1 4313.166 < 2.2e-16 ***
year        6.576   6   26.027 < 2.2e-16 ***
logL:year   4.467   6   17.679 < 2.2e-16 ***
Residuals  37.645 894                       
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Differences Among Slopes

The log(Total Length) across different years show several significant differences, particularly between 2012 and 2016 (p = 0.000), 2012 and 2018 (p = 0.000), 2012 and 2024 (p = 0.000), 2014 and 2016 (p = 0.000), and 2016 and 2022 (p = 0.000). These results suggest that the total length of the individuals significantly differs between these years. In contrast, some year comparisons, such as 2012 vs. 2022 (p = 1.000) and 2018 vs. 2020 (p = 0.988), show no significant differences, indicating that in these comparisons, the total length measurements remain relatively consistent over time.

# Run the emtrends for years
cs <- emtrends(fit1,
               specs=pairwise~year,
               var="logL")

css <- summary(cs,infer=TRUE)

# Extract contrasts across years
contrast_table <- data.frame(
  Contrast = css$contrasts$contrast,
  Estimate = css$contrasts$estimate,
  Std.Error = css$contrasts$SE,
  t.value = css$contrasts$t.ratio,
  p.value = css$contrasts$p.value
)

# Highlight significant contrasts (p-value < 0.05)
contrast_table$highlight <- ifelse(contrast_table$p.value < 0.05, "significant", "not_significant")

kable(contrast_table, 
      caption = "Comparison of log(Total Length) Across Years", 
      digits = 3) %>%
  kable_styling() %>%  
  row_spec(0, background = "white") %>%  
  row_spec(which(contrast_table$highlight == "significant"), 
           color = "black", 
           background = "lightyellow") %>%  
  row_spec(which(contrast_table$highlight == "not_significant"), 
           color = "black", 
           background = "white") %>%  
  # Explicitly reset row colors to white for all other rows to avoid alternating rows
  kable_classic(full_width = F, html_font = "arial", position = "left")

Comparison of log(Total Length) Across Years

Contrast	Estimate	Std.Error	t.value	p.value	highlight
year2012 - year2014	-0.329	0.173	-1.897	0.483	not_significant
year2012 - year2016	-1.223	0.152	-8.052	0.000	significant
year2012 - year2018	-0.887	0.155	-5.713	0.000	significant
year2012 - year2020	-0.758	0.198	-3.839	0.003	significant
year2012 - year2022	0.058	0.235	0.249	1.000	not_significant
year2012 - year2024	-0.791	0.147	-5.370	0.000	significant
year2014 - year2016	-0.894	0.135	-6.611	0.000	significant
year2014 - year2018	-0.558	0.139	-4.015	0.001	significant
year2014 - year2020	-0.430	0.185	-2.322	0.235	not_significant
year2014 - year2022	0.387	0.224	1.725	0.599	not_significant
year2014 - year2024	-0.462	0.130	-3.554	0.007	significant
year2016 - year2018	0.336	0.111	3.024	0.041	significant
year2016 - year2020	0.464	0.165	2.811	0.075	not_significant
year2016 - year2022	1.281	0.208	6.153	0.000	significant
year2016 - year2024	0.432	0.100	4.330	0.000	significant
year2018 - year2020	0.128	0.168	0.762	0.988	not_significant
year2018 - year2022	0.945	0.211	4.486	0.000	significant
year2018 - year2024	0.096	0.105	0.914	0.970	not_significant
year2020 - year2022	0.817	0.244	3.353	0.015	significant
year2020 - year2024	-0.033	0.161	-0.202	1.000	not_significant
year2022 - year2024	-0.849	0.205	-4.145	0.001	significant

Comparison of log(Total Length) Across Years. Significant differences (p-value < 0.05) are highlighted in yellow, indicating that the total length of the individuals significantly differs between these years.

Summary Plot

This summary plot of the regression results highlights the relationship between log-transformed weight and total length for each year. The plot shows differences in the weight-length relationship across years, with some years showing steeper slopes than others. These differences suggest that the body condition of California tiger salamanders varies over time, which may be influenced by environmental factors (precipitation, humidity, etc.) or population dynamics.

# Load fonts
font_add_google("Lato", "lato")
showtext_auto()

# Create plot 
plot <- ggplot(cts, aes(x = logL, y = logW, color = as.factor(year))) +
  geom_point(alpha = 0.6, size = 2) +  
  geom_smooth(method = "lm", formula = y ~ x, se = FALSE, linewidth = 1.2) +  
  
  scale_color_viridis_d(option = "magma", name = "Year") +  
  theme_minimal(base_family = "lato") +  
  
  labs(
    title = "CTS Weight-Length Relationship Across Years",
    subtitle = "Weights and total lengths vary across survey years (2012 to 2024)",
    x = "Log(Total Length) (mm)", 
    y = "Log(Weight) (g)",
    caption = "Data Source: CTS Trapping Data from CCCI | Analysis by Madi Calbert"
  ) +
  
  theme(
    plot.title = element_text(size = 14, face = "bold", hjust = 0),
    plot.subtitle = element_text(size = 12, hjust = 0, 
                                 margin = margin(b = 10)),
    axis.title = element_text(size = 10),
    axis.text = element_text(size = 10),
    legend.position = "right",
    legend.text = element_text(size = 10),
    legend.title = element_text(size = 10),
    plot.caption = element_text(size = 8, hjust = 0, face = "italic", 
                                margin = margin(0.75, 0, 0, 0, "cm")),
    plot.background = element_rect(fill = "#D3D3D3", color = NA),
    panel.grid = element_blank(),  # Remove all gridlines
    axis.ticks = element_line(color = "black"),
    axis.line = element_line(color = "black"))
plot

Weight-length Relationship of CTS Across Years. The regression lines represent the relationship between log-transformed weight and total length for each year. The plot highlights the differences in the weight-length relationship across years, with some years showing steeper slopes than others. These differences suggest that the body condition of California tiger salamanders varies over time, which may be influenced by environmental factors or population dynamics.

#ggsave("outputs/cts_year.png", plot, width = 8, height = 6, dpi = 400)

Weight-Length Relationship Across Sites

ANOVA Results

Likewise to across years, when comparing across sites there are statistically significant results (p-value < 0.05), therefore at least 2 of the log-transformed total length means are different across sites.

car::Anova(fit2)

Anova Table (Type II tests)

Response: logW
           Sum Sq  Df   F value    Pr(>F)    
logL      252.511   1 5303.6867 < 2.2e-16 ***
site        2.724   8    7.1506 3.334e-09 ***
logL:site   3.590   8    9.4266 1.469e-12 ***
Residuals  42.373 890                        
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Differences Among Slopes

The only statistically significant difference (p-value < 0.05) is between the Eagles Nest site and the other sites. This suggests that the total length of the individuals at the Eagles Nest site significantly differs from the other sites. In contrast, the other site comparisons show no significant differences, indicating that the total length measurements remain relatively consistent across the other sites.

# Run the emtrends
cs <- emtrends(fit2,
               specs=pairwise~site,
               var="logL")

css <- summary(cs,infer=TRUE)

# Extract contrasts across sites
contrast_table <- data.frame(
  Contrast = css$contrasts$contrast,
  Estimate = css$contrasts$estimate,
  Std.Error = css$contrasts$SE,
  t.value = css$contrasts$t.ratio,
  p.value = css$contrasts$p.value
)

# Highlight significant contrasts (p-value < 0.05)
contrast_table$highlight <- ifelse(contrast_table$p.value < 0.05, "significant", "not_significant")

kable(contrast_table) %>%
  kable_styling() %>%  
  row_spec(0, background = "white") %>%  
  row_spec(which(contrast_table$highlight == "significant"), 
           color = "black", 
           background = "lightyellow") %>%  
  row_spec(which(contrast_table$highlight == "not_significant"), 
           color = "black", 
           background = "white") %>%  
  # Explicitly reset row colors to white for all other rows to avoid alternating rows
  kable_classic(full_width = F, html_font = "arial", position = "left")

Contrast	Estimate	Std.Error	t.value	p.value	highlight
Bermed Area - Eagles Nest	1.2636532	0.2370430	5.3309029	0.0000044	significant
Bermed Area - (IA-100 Area A)	0.4362837	0.2448144	1.7821002	0.6941643	not_significant
Bermed Area - (IA-100 Area B)	0.6840371	0.2375973	2.8789768	0.0951899	not_significant
Bermed Area - (IA-100 Area East)	0.4205597	0.2772832	1.5167156	0.8477568	not_significant
Bermed Area - (IA-100 Area North)	0.3016716	0.3232627	0.9332087	0.9910400	not_significant
Bermed Area - (IA-100 Area West)	0.5447018	0.2567392	2.1216150	0.4590710	not_significant
Bermed Area - Pistol Range	-0.3183328	1.0937994	-0.2910340	0.9999985	not_significant
Bermed Area - RPA	0.3949073	0.2212644	1.7847761	0.6923985	not_significant
Eagles Nest - (IA-100 Area A)	-0.8273695	0.1509924	-5.4795426	0.0000020	significant
Eagles Nest - (IA-100 Area B)	-0.5796161	0.1389860	-4.1703187	0.0011087	significant
Eagles Nest - (IA-100 Area East)	-0.8430935	0.1993755	-4.2286715	0.0008669	significant
Eagles Nest - (IA-100 Area North)	-0.9619816	0.2595446	-3.7064209	0.0068827	significant
Eagles Nest - (IA-100 Area West)	-0.7189514	0.1696457	-4.2379575	0.0008334	significant
Eagles Nest - Pistol Range	-1.5819860	1.0766902	-1.4693047	0.8695639	not_significant
Eagles Nest - RPA	-0.8687459	0.1087316	-7.9898205	0.0000000	significant
(IA-100 Area A) - (IA-100 Area B)	0.2477534	0.1518612	1.6314467	0.7873334	not_significant
(IA-100 Area A) - (IA-100 Area East)	-0.0157240	0.2085552	-0.0753951	1.0000000	not_significant
(IA-100 Area A) - (IA-100 Area North)	-0.1346122	0.2666610	-0.5048063	0.9998931	not_significant
(IA-100 Area A) - (IA-100 Area West)	0.1084181	0.1803451	0.6011699	0.9996029	not_significant
(IA-100 Area A) - Pistol Range	-0.7546165	1.0784278	-0.6997377	0.9987925	not_significant
(IA-100 Area A) - RPA	-0.0413764	0.1247688	-0.3316246	0.9999958	not_significant
(IA-100 Area B) - (IA-100 Area East)	-0.2634774	0.2000342	-1.3171618	0.9262642	not_significant
(IA-100 Area B) - (IA-100 Area North)	-0.3823655	0.2600510	-1.4703484	0.8691047	not_significant
(IA-100 Area B) - (IA-100 Area West)	-0.1393353	0.1704194	-0.8176024	0.9963577	not_significant
(IA-100 Area B) - Pistol Range	-1.0023699	1.0768124	-0.9308678	0.9911885	not_significant
(IA-100 Area B) - RPA	-0.2891298	0.1099348	-2.6300119	0.1756173	not_significant
(IA-100 Area East) - (IA-100 Area North)	-0.1188881	0.2967490	-0.4006352	0.9999818	not_significant
(IA-100 Area East) - (IA-100 Area West)	0.1241421	0.2224326	0.5581113	0.9997719	not_significant
(IA-100 Area East) - Pistol Range	-0.7388925	1.0862589	-0.6802177	0.9990162	not_significant
(IA-100 Area East) - RPA	-0.0256524	0.1803306	-0.1422518	1.0000000	not_significant
(IA-100 Area North) - (IA-100 Area West)	0.2430302	0.2776492	0.8753141	0.9941772	not_significant
(IA-100 Area North) - Pistol Range	-0.6200044	1.0988954	-0.5642069	0.9997526	not_significant
(IA-100 Area North) - RPA	0.0932358	0.2452182	0.3802156	0.9999879	not_significant
(IA-100 Area West) - Pistol Range	-0.8630346	1.0811972	-0.7982212	0.9969188	not_significant
(IA-100 Area West) - RPA	-0.1497945	0.1467931	-1.0204463	0.9839372	not_significant
Pistol Range - RPA	0.7132401	1.0733267	0.6645135	0.9991700	not_significant

Comparison of log(Total Length) Across Sites. Significant differences (p-value < 0.05) are highlighted in yellow, indicating that the total length of the individuals significantly differs between these sites.

Summary Plot

This summary plot of the regression results highlights the relationship between log-transformed weight and total length for each trapping site. The plot shows differences in the weight-length relationship across sites, with only one site showing a slightly steeper slopes than eight others. These similarities suggest that the body condition of California tiger salamanders are consistent across sites.

# Create plot 
plot_site <- ggplot(cts, aes(x = logL, y = logW, color = as.factor(site))) +
  geom_point(alpha = 0.6, size = 2) +  
  geom_smooth(method = "lm", formula = y ~ x, se = FALSE, linewidth = 1.2) +  
  
  scale_color_viridis_d(option = "magma", name = "Site") +  
  theme_minimal(base_family = "lato") +  
  
  coord_cartesian(clip = "off") +  # Allow points to extend beyond the plot area
  
  labs(
    title = "CTS Weight-Length Relationship Across Sites",
    subtitle = "Weights and total lengths are consistent across most survey sites",
    x = "Log(Total Length) (mm)", 
    y = "Log(Weight) (g)",
    caption = "Data Source: CTS Trapping Data from CCCI | Analysis by Madi Calbert"
  ) +
  
  theme(
    plot.title = element_text(size = 14, face = "bold", hjust = 0),
    plot.subtitle = element_text(size = 12, hjust = 0, 
                                 margin = margin(b = 10)),
    axis.title = element_text(size = 10),
    axis.text = element_text(size = 10),
    legend.position = "right",
    legend.text = element_text(size = 10),
    legend.title = element_text(size = 10),
    plot.caption = element_text(size = 8, face = "italic", hjust = 0, 
                                margin = margin(0.75, 0, 0, 0, "cm")),
    plot.background = element_rect(fill = "#D3D3D3", color = NA),
    panel.grid = element_blank(),  # Remove all gridlines
    axis.ticks = element_line(color = "black"),
    axis.line = element_line(color = "black")) 

plot_site

Weight-length Relationship of CTS Across Sites. The regression lines represent the relationship between log-transformed weight and total length for each site. The plot highlights similar slopes in the weight-length relationship across sites, with only the Eagles Nest site showing steeper slopes than others. These similarities suggest that the body condition of California tiger salamanders across trapping locations are the same.

#ggsave("outputs/cts_sites.png", plot_site,width = 8, height = 6, dpi = 400)

Discussion

The results of this study indicate that the weight-length relationship of California tiger salamanders (Ambystoma californiense) varies significantly across certain years, but only varies across one of the trapping sites. The statistical analyses confirmed that differences in log-transformed mean total lengths were significant across multiple years, suggesting that body condition fluctuates over time. This iimplies that temporal environmental conditions and habitat characteristics influence salamander growth patterns. However differences across trapping sites were only significant between Eagle’s Nest and the other sites. Meanwhile the other eight sites had a similar weight-length relationship. This suggests that the body condition of California tiger salamanders is consistent across most trapping locations.

The observed differences in weight-length relationships across years likely reflect environmental variations and population dynamics affecting the growth and condition of CTS. Significant pairwise differences between certain years (e.g., 2012 vs. 2016, 2012 vs. 2024) suggest that environmental factors such as precipitation levels, temperature fluctuations, and resource availability may play a role in salamander growth. Since CTS depend on vernal pools and seasonal wetlands for breeding and development, interannual variability in rainfall could impact juvenile recruitment and adult body condition.

The lack of significant differences in weight-length relationships across most trapping sites indicates that the body condition of CTS is relatively consistent across different sites within the Concord Naval Weapons Station. This spatial consistency suggests that the environmental conditions and habitat characteristics at these sites are similar enough to maintain a stable weight-length relationship. However, the significant difference observed at the Eagle’s Nest site may be due to unique habitat features, resource availability, or predator-prey dynamics that influence salamander growth and condition. The Eagle’s Nest site is the farthest away from the other trapping sites, which could indicate that this is a separate sub-population of CTS with different habitat conditions and availability of resources, both of which could greatly influence body condition.

Further research is needed to explore the underlying factors driving these variations in weight-length relationships. Further studies incorporating environmental variables such as precipitation patterns, temperature data, and habitat quality assessments could provide deeper insights into the drivers of body condition changes. Additionally, investigating genetic diversity and population connectivity across sites may help determine whether observed differences are due to environmental influences or population structure.

Works Cited

Ogle, D. H., & Winfield, I. J. (2009). Ruffe Length–Weight Relationships with a Proposed Standard Weight Equation. North American Journal of Fisheries Management, 29(4), 850–858. https://doi.org/10.1577/M08-176.1