ABSTRACT
BACKGROUND AND AIMS: Leaf area (A) is a crucial indicator of the photosynthetic capacity of plants. The Montgomery equation (ME), which hypothesizes that A is proportional to the product of leaf length (L) and width (W), is a valid tool for nondestructively measuring A for many broad-leaved plants. At present, the methods used to compute L and W for ME can be broadly divided into two kinds: using computer recognition, and measuring manually. However, the potential difference in the prediction accuracy using either method has not been thoroughly examined in prior studies. METHODS: In the present study, we measured 540 Alangium chinense leaves, 489 Liquidambar formosana leaves, and 215 Liriodendron × sinoamericanum leaves, utilizing computer recognition and manual measurement methods to determine L and W. ME was used to fit the data determined by the two methods, and the goodness of fits were compared. The prediction errors of A were analyzed by examining the correlations with two leaf symmetry indices (areal ratio of the left side to the right side, and standardized index for bilateral asymmetry), as well as the leaf shape complexity index (the leaf dissection index). KEY RESULTS: The results indicate that there is a neglectable difference in the estimation of A between both methods. This further validates that ME is an effective method for estimating A in broad-leaved tree species, including those with lobes. Additionally, leaf shape complexity significantly influenced the estimation of A. CONCLUSIONS: These results show that the use of computer recognition and manual measurement in the field are both effective and feasible, although the influence of leaf shape complexity should be considered when applying ME to estimate A in the future.
