1   156 m³  1,4 °C
2   199 m³  0,3 °C
3   173 m³  2,3 °C
4   69 m³   9,6 °C
5   63 m³   12,2 °C
6   9 m³    16,9 °C
7   9 m³    20,8 °C
8   9 m³    18,5 °C
9   15 m³   14,3 °C
10  19 m³   11,4 °C
11  83 m³   5,2 °C
12  62 m³   4,2 °C

cons <- structure(c(156, 199, 173, 69, 63, 9, 9, 9, 15, 19, 83, 62), .Tsp = c(1,
  1.91666666666667, 12), class = "ts")
temp <- structure(c(1.4, 0.3, 2.3, 9.6, 12.2, 16.9, 20.8, 18.5, 14.3,
  11.4, 5.2, 4.2), .Tsp = c(1, 1.91666666666667, 12), class = "ts")

fit <- lm(c(cons) ~ poly(c(temp), 2))
summary(fit)
#Residuals:
#    Min      1Q  Median      3Q     Max
#-52.340  -7.920  -2.102  17.963  34.661
#Coefficients:
#                  Estimate Std. Error t value Pr(>|t|)
#(Intercept)         72.167      7.451   9.686 4.66e-06 ***
#poly(c(temp), 2)1 -201.911     25.810  -7.823 2.65e-05 ***
#poly(c(temp), 2)2   69.987     25.810   2.712   0.0239 *
#---
#Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#Residual standard error: 25.81 on 9 degrees of freedom
#Multiple R-squared:  0.8839,   Adjusted R-squared:  0.8582
#F-statistic: 34.28 on 2 and 9 DF,  p-value: 6.18e-05

plot(c(temp), c(cons), pch = 16)
lines(c(sort(temp)), predict(fit)[order(temp)], type = "b", col = "blue")

p <- predict(fit, newdata = data.frame(temp = c(5, 15)), se.fit = TRUE)
res <- cbind(p$fit - 1.96 * p$se, p$fit, p$fit + 1.96 * p$se)
colnames(res) <- c("lower limit", "pred", "upper limit")
res
#  lower limit     pred upper limit
#1   83.072304 102.5629   122.05356
#2   -5.487951  14.9201    35.32816