Applying Non-destructive Techniques to Evaluate the Nitrogen Status, Irrigation Index, and Seedling Vigor for Star Cluster (Pentas lanceolata)

本論文以常見之草本景觀植物繁星花(star cluster, Pentas lanceolata ''Butterfly Red'' )為試驗材料，應用非破壞性技術(non-destructive technique)，即時得知葉片氮素濃度(nitrogen concentration)、氮素施用飽和點時之葉片氮濃度(the saturation point of plant nitrogen irrigation)、葉片水勢(leaf water potential)、灌溉點(irrigation timing)及種苗活力(seedling vigor)，可建立即時生育品質評估系統。 氮素狀態方面，進行氮素施用飽和點預測試驗，以建立氮肥管理模式。將均質之受試植株，分為五種處理，施用不同濃度之氮素(0 mM、4 mM、8 mM、16 mM、24 mM) 30天，之後量測其葉綠素螢光(chlorophyll fluorescence)、葉綠素計(soil-plant analysis development)及常態化差異植被指數(normalized difference vegetation index)、葉片氮素濃度(nitrogen concentration)及外觀型態與生理參數等測值(morphological and physiological parameters)。結果得知，五種不同氮素處理之葉片氮素濃度分別為2.62%、 3.48%、 4.00%、 4.23%、 及4.69%，在0 到 8 mM 氮素時，狄克森品質指數(Dickson’s quality index)、地上部乾重(above ground dry weight)、全株乾種、開花率(flowering rate)、有效光合作用光化學量子產量(the effective photochemical quantum yield of PSII, ∆F/Fm’) 及光化學消散(photochemical quenching, qP)，趨勢向上且到達最高峰，超過8 mM之後則下降。葉片氮素濃度與葉綠素計、常態化差異植被指數、有效光合作用光化學量子產量及地上部乾重有較高之相關性(R2 = 0.60 to 0.85)，前三者可更精準且非破壞性量測葉片氮素濃度，取代地上部乾重之量測。建議氮素施用飽和點，以葉片氮素濃度、葉綠素計、常態化差異植被指數、有效光合作用光化學量子產量為例，其數值分別為4.00%、 50.68、 0.64及 0.137。 灌溉指標方面，利用非破壞性技術量測植株水分含量，預測澆水時間點，以建立精準之水分管理模式。將繁星花盆栽(介質為泥炭苔:真珠石=4:1 (v/v))進行連續乾旱 (0、3、5、7、12及16天)，讓土壤介質水分含量維持在10%至45%間，量測葉片水勢(leaf water potential)、介質水分含量(soil water content)、介質基質勢(soil matric potential)、葉綠素螢光(chlorophyll fluorescence)、光化學反射指數(photochemical reflectance index)、調整常態化差異植被指數(adjusted normalized difference vegetation index) 及1950 nm 處之反射值( the reflectance at 1950 nm)。結果顯示：繁星花暫時萎凋點時之葉片水勢為–3.87 MPa，此時上位葉與下位葉之葉綠素螢光值Fm’比值等於1.7。繁星花的非破壞性灌溉指標，建議於上位葉與下位葉之Fm’比值大於1.3時灌溉，此時下位葉葉片水勢 < –2.27 MPa、介質水分含量 < 21%、介質基質勢< –20 kPa、光化學反射指數< 0.0443、調整常態化差異植被指數< 0.0301及1950 nm 處之反射值> 8.904。故上位葉與下位葉之葉綠素螢光值Fm’可作為非破壞性水分預測指標。 種苗活力方面，其值與定植後的成活率及生育表現息息相關，常以狄克森品質指數(Dickson’s quality index) 及根生長潛勢(root growth potential)為依據，但其測量通常較費時繁瑣或需破壞樣品。本研究以繁星花為試驗材料，利用非破壞性量測技術，如葉綠素螢光(chlorophyll fluorescence)參數中之有效光合作用光化學量子產量(the effective photochemical quantum yield of PSII, ∆F/Fm’) 及光化學消散(photochemical quenching, qP)，以及葉面積指數(leaf area index)等數值，進行種苗活力預測試驗，以建立栽培管理模式。將繁星花放置於自然環境下120天，之後量測十三種外觀型態與生理參數等測值(morphological and physiological parameters)。根生長潛勢(root growth potential)為新生根數之多寡，可代表種苗活力，以該數值將繁星花分為五個族群，外觀型態與生理參數也依此分類，數字越大，代表種苗活力越佳。狄克森品質指數、葉面積指數、全株乾種、 有效光合作用光化學量子產量及光化學消散等參數，其趨勢皆與根生長潛勢一致。而狄克森品質指數與葉面積指數、有效光合作用光化學量子產量及兩者乘積具有高度之相關性(R2 =0.59 to 0.93)， 尤其是(葉面積指數乘以有效光合作用光化學量子產量)之數值，可取代根生長潛勢與狄克森品質指數，成為種苗活力之非破壞性量測指標。 上述成果可提供研究者及第一線從業人員快速且精準掌握植株現況，未來期望可將該模式應用至不同作物上，獲得即時植株生育品質評估系統，並探討不同植物間之反應，依其數據給予分類，提供種類選擇及栽培管理模式之建議，進而提升景觀品質。This project is using ornamental herbaceous star cluster (Pentas lanceolata) as the experimental plant materials. The chlorophyll fluorescence (ChlF, e.g., the effective photochemical quantum yield of PSII, ∆F/Fm’; photochemical quenching, qP), reflectance spectra and vegetation indices used to build up a realtime evaluation system for leaf nitrogen (N) concentration, N demand timing for fertilizer, leaf water potential (WP), irrigation timing and seedling vigor (SV). The system will provide suitable sugestions for plant material selection and cultivation management model and could improve landscaping qualities and functions. The objective of nitrogen (N) study was to predict the N demand timing for fertilizer application through ChlF, soil-plant analysis development (SPAD), and normalized difference vegetation index (NDVI). The tested plants were grown in potting soil by weekly irrigation with five concentrations (0, 4, 8, 16, and 24 mM) of N for 30 d. These five N application levels corresponding to the N concentration in leaves of tested plants were 2.62%, 3.48%, 4.00%, 4.23%, and 4.69%, respectively. The trends and rates of increase from 0 to 8 mM N treatments to its peak in Dickson’s quality index (DQI), above ground dry weight (DW), total DW, flowering rate, ∆F/Fm’, and qP were all similar to SPAD, NDVI, and the maximum photochemical quantum yield (Fv/Fm) indices. Consistent and strongly high correlations (R2 = 0.60 to 0.85) were observed among leaf N concentration (%) and SPAD, NDVI, ∆F/Fm’, and above-ground DW. With validation of these vegetation indices, leaf SPAD, NDVI, and ∆F/Fm’ are shown to be accurate and non-destructive predictors of leaf N concentration and can be used to accurately estimate N-solution irrigation timing for P. lanceolata. Therefore, the saturation point of plant N irrigation is recommended when leaf N concentration, SPAD, NVDI, and ∆F/Fm’ ratio are 4.00%, 50.68, 0.64, or 0.137, respectively. The objective of moisture study was to use non-destructive measurements as the precise irrigation indices. Drought stress was imposed on plants for 0, 3, 5, 7, 12, and 16 days by withholding water. Measurements were conducted on the third leaf counted from the apex (upper-leaves) and the third leaf from the bottom. Within the range of soil water content (WC) from 10 to 45%, leaf WP, soil WC, soil matric potential (MP), ChlF, photochemical reflectance index (PRI), adjusted normalized difference vegetation index (aNDVI) and the reflectance (R) at 1950 nm (R1950) were measured. Results show that the plants reached the temporary wilting point exhibited –3.87 MPa for leaf WP, and the maximal fluorescence yield of the light adapted state (Fm’) ratio of upper-to-lower leaves was 1.7. When the Fm’ ratio was 1.3, which corresponded to lower leaf WP < –2.27 MPa, soil WC < 21%, MP < –20 kPa, PRI < 0.0443, aNDVI < 0.0301, and R1950 > 8.904, it was time to irrigate. In conclusion, the Fm’ ratio of upper-to-lower-leaves is shown to be a non-destructive predictor of leaf WP and can be used to estimate irrigation timing. The objective of quality study was to use the non-destructive measurement of ChlF (such as ∆F/Fm’ and qP) and leaf area index (LAI) as SV indices. Plants were grown in potting soil under nature sunlight for 120 d. Plants were separated into 5 root growth potential (RGP) groups based on the number of new roots, and morphological and physiological parameters were also separated into those same levels. The trends and rates of increase from levels 1 to 5 in DQI, LAI, total dry weight, ∆F/ Fm’, and qP were all similar to the RGP index. Although RGP and DQI are frequently used as indices for SV, these measurements are time-consuming and require sample destruction. Consistent and strongly high correlations (R2 =0.59 to 0.93) were observed among DQI and LAI, ∆F/Fm’, and LAI × (∆F/Fm’), demonstrating the applicability of these indices for measuring SV in star cluster. In particular, LAI × (∆F/Fm’) was predicted using multiple variables from validation datasets, predictions were compared to actual DQI and RGP values for star cluster, and SV indices were predicted. Therefore, LAI × (∆F/Fm’) can replace DQI and RGP in the non-destructive estimation of SV