Principal component analyis

I measured different plant traits and enviromental factors like radiaton or temperature for 42 different genotypes. I want to know which traits are related to each other and which traits are influenced to each other (e.g. radiation to trait). Thus a PCA came in my mind.

That is a part of my data frame (release.year means the release year of the genotype and each row is one measurement; there are multiple measurements per genotype):

 release.year PARmolmsmean TemperaturCmean     Fv.Fm    Fr1.Fm     NDVI2
1           1951     1160.480         14.5698 0.4252503 0.1992661 0.9069721
2           1951     1165.150         24.1412 0.4714152 0.3363943 0.9452823
3           1951     1478.050         18.8031 0.4845846 0.2643791 0.9388180
4           1951     1536.150         11.3179 0.3393415 0.1957040 0.9074890
5           1951     1566.880         25.4107 0.4181181 0.3234387 0.9385588
6           1951     1605.120         22.4142 0.4628020 0.2911468 0.9459246
7           1951     1653.270         20.8988 0.4710748 0.3004921 0.9350731
8           1951     1686.590         21.8360 0.5086655 0.3506261 0.9252237
9           1951     1741.540         14.9924 0.4425364 0.2233692 0.9208316
10          1951     1749.030         26.4696 0.4479734 0.3712510 0.9341932
11          1951     1916.180         28.8097 0.4193770 0.3272526 0.9442456
12          1951     2036.370         22.9217 0.4212041 0.3460669 0.9177171
13          1951     2058.530         28.3290 0.4435373 0.3465088 0.9285163
14          1951     2068.900         23.7608 0.4312131 0.3217924 0.9370028
15          1951      644.265         12.6475 0.4047917 0.1882771 0.9239126
16          1951      794.578         17.1793 0.5568734 0.2214873 0.9444661
17          1951      914.188         15.2715 0.5121282 0.2064338 0.9382595
18          1951      958.170         15.3889 0.4508183 0.2970802 0.9271799
19          1953     1429.090         15.0966 0.4379091 0.2478120 0.9374094
20          1953     1478.050         18.8031 0.5354153 0.2576067 0.9459906
21          1953     1576.210         24.7682 0.4743814 0.3157759 0.9372654
22          1953     1632.500         22.4675 0.3945049 0.3046481 0.9206227
23          1953     1683.670         12.5784 0.3904874 0.1848006 0.9317080
24          1953     1686.730         22.0717 0.4769295 0.3176542 0.9492195
25          1953     1770.230         15.1633 0.4263168 0.2296812 0.9389138
26          1953     1831.960         26.8531 0.4314113 0.3627886 0.9472129
27          1953     1857.880         21.6815 0.3569287 0.2964029 0.9423032
28          1953     1986.480         24.8260 0.3733148 0.3234840 0.9236503
29          1953     2058.530         28.3290 0.4024092 0.3511438 0.9283465
30          1953     2068.490         23.2445 0.4422827 0.3369628 0.9453007
31          1953     2099.660         24.0771 0.3738086 0.3148910 0.9433190
32          1953     2108.340         28.7854 0.3979736 0.3492899 0.9323239
33          1953      611.627         12.9907 0.5050644 0.1987066 0.9363292
34          1953      668.194         17.3683 0.5871649 0.2213303 0.9452587
35          1953      782.109         15.2782 0.4651738 0.2537262 0.9305000
36          1953      935.380         14.5458 0.3716472 0.2293471 0.9181275
37          1956     1159.140         12.5591 0.4584142 0.2130523 0.9209437
38          1956     1165.150         24.1412 0.4869044 0.3122720 0.9291310
39          1956     1302.300         22.3879 0.4255736 0.3140835 0.9240680
40          1956     1429.090         15.0966 0.4489692 0.2425797 0.9270232
41          1956     1491.960         19.5193 0.4793279 0.2783667 0.9412001
42          1956     1686.590         21.8360 0.5215572 0.3507388 0.9523238
43          1956     1728.250         24.0542 0.4537621 0.3304822 0.9238353
44          1956     1749.030         26.4696 0.5033194 0.3627435 0.9310118
45          1956     1770.230         15.1633 0.4727894 0.2429909 0.9355953
46          1956     1857.880         21.6815 0.4497379 0.2983577 0.9225823
...

sum.pca <- prcomp(SUM, center = TRUE,scale. = TRUE )

ggbiplot(sum.pca, alpha = 0)

But I realy don't know if that is useful to investigate correlations between the factors. E.g. Fv.Fm is a plant trait and temperature is an eviromental factor which doesn't discribe a plant trait, but it influences it. Thus the main question is: Is it statistical correct to calculate a PCA for that case?

I hope my explanation is comprehensible.

Thanks for your help.

The output of ggbiplot() gives you information of how the features in your data correlate together. Arrows which point in similar directions imply those features exhibit a similar trend. If your PCA is stored in sum.pca, you'll find the data for each component in sum.pca[["x"]] as PC1, PC2, PC3, etc. These can be used to create a scatter plot to visualize clustering of your data. You may also want to use a scree plot to see how much variation is captured by each dimension of your PCA. For a clearer picture of how features relate to each other, you may want to look at corrplot.