In-season assessment of wheat crop health using Vegetation Indices based on ground measured hyper spectral data

Transcription

1 In-season assessment of wheat crop health using Vegetation Indices based on ground measured hyper spectral data Dr. K.A. Al-Gaadi 1, Dr. V.C. Patil 2, E. Tola 2, Dr. M. Rangaswamy 2, and Dr. S.A. Marey 2 1 Department of Agricultural Engineering, Precision Agriculture Research Chair, College of Food and Agriculture Sciences, King Saud University, Riyadh, Saudi Arabia. 2 Precision Agriculture Research Chair, King Saud University, Riyadh, Saudi Arabia. KEYWORDS: remote sensing, spectral reflectance, vegetation indices, wheat. ABSTRACT Vegetation Indices (VI) determined at various crop growth stages help in understanding the response of crop to management practices. An investigation was carried out to determine the right VI that are helpful in assessing the in-season performance of wheat crop treated with graded levels of irrigation water and fertilizers. An experiment on a 50 ha center pivot field was laid out in Split Plot design with four irrigation and three fertilizer levels. Irrigation levels at 100, 90, 80 and 70% evapotranspiration (ETc) formed the main-plot treatments and three fertilizer levels of N:P:K, kg/ha: 300:150:200 (low); 400:250:300 (medium) and 500:300:300 (high), formed the sub-plot treatments. The crop was sown on January 1 st and harvested on May 9 th, Temporal data on biophysical parameters and reflectance of wheat crop in hyper spectral bands ( nm) using Spectroradiometer (Model- Field Spec 3 of ASD Inc. Colorado, USA) were collected at jointing and heading growth stages (February 17 th and April 5 th, 2012). The results of this study revealed that many of the tested spectral indices showed significant response to irrigation levels. But, only two spectral indices showed significant response to fertilizer levels (Plant Senescence Reflectance -PSRI and the Photochemical Reflectance -PRI). PSRI and PRI were sensitive to both irrigation and fertilizer levels. The Middle Infrared-Based Vegetation (MIVI) showed a significant response to the implemented irrigation levels for both sampling dates. Among the tested spectral indices, Normalized Difference Infrared (NDII) and the Normalized Difference Nitrogen (NDNI) exhibited the highest correlation to LAI; and five indices showed the most response to wheat grain yield: Near Infrared band (NIR), Water Band (WBI), Normalized Water -1 (NWI-1), Normalized Water -3 (NWI- 3), and Normalized Water -4 (NWI-4). 1. INTRODUCTION The use of remote sensing applications for crop growth monitoring is becoming an essential part of today's agriculture as it enhances the efficient management of agricultural resources. The key factor for precision crop growth monitoring is the selection of the efficient devices and methods for the accurate measurements of crop growth parameters. Crop spectral reflectance measurements are being used efficiently for crop growth monitoring as they reflect wide range of biochemical and physiological measures. Thus, remote sensing applications based on the measurement and interpretation of the spectral reflectance of agricultural crops are considered as the most important tools for assessment of crop growth and yield performance (Aparicio et al., 2000). vegetation indices, determined mathematically at various spectral bands, are considered as semi-analytical measures of vegetation activity and appropriate for the assessment of spatial variability in agricultural fields (Vina et al., 2011). These vegetation indices will enhance the interpretation process of the crop spectral reflectance measured at different growth Page 58

2 stages, and hence will help in understanding the response of agricultural crops to management practices. Therefore, remote sensing based on various vegetation indices (VI) is expected to provide important information that will help in the efficient assessment of within-field spatial variability of various agronomic parameters (Hatfield and Prueger, 2010). Leaf Area (LAI) is considered as one of the most important biophysical parameters for the assessment of crop performance (vegetation health, biomass, and photosynthesis) and for estimating crop yield. Monitoring of LAI during the cropping season will help in understanding the spatial variability in crop productivity. On the other hand, LAI can serve as indicator of stress in vegetation (Duchemin et al., 2006). The objective of this study was to determine the right vegetation indices (VI) that are helpful in assessing the in-season performance of wheat crop cultivated under different levels of irrigation water and fertilizers. 2. MATERIALS AND METHODS 2.1 Study area This study was conducted on a center pivot field in Todhia Farm located in the Eastern Province of Saudi Arabia within the latitudes of 24º10' 22.77" and 24º12' 37.25" N and longitudes of 47º56' 14.60" and 48º05' 08.56" E (Figure 1). 2.2 Experimental Layout An experiment was conducted on a field of 50 ha cultivated with wheat crop (Triticum aestivum L., cv. Yecora Rojo) sown at 250 kg/ha on January 1 st, Experimental treatments were laid out in split plot design with four irrigation levels as the main treatments and three fertilizer levels as the subtreatments (Figure 2). The main treatments were I1 (irrigation at 100% ETc), I2 (Irrigation at 90% ETc), I3 (Irrigation at 80% ETc), and I4 (Irrigation at 70% ETc). While, the sub-treatments were F1(300:200:200 N:P 2 O 5 :K 2 O kg/ha), F2(400:250:250 N:P 2 O 5 :K 2 O kg/ha), and F3(500:300:300 N:P 2 O 5 :K 2 O kg/ha). Irrigation Levels: I1 Irrigation level 1 I2 Irrigation Level 2 I3 Irrigation level 3 I4 Irrigation level 4 Figure 2: Layout plan of the experiment. Fertilizer Levels: F1 Fertilizer level 1 F2 Fertilizer level 2 F3 Fertilizer level reflectance measurements Wheat crop samples were collected on two different dates (February 17 th, and April 5 th, 2012) that coincided with jointing and heading stages. The crop spectral reflectance was measured in the Laboratory by a Spectroradiometer (Model- Field Spec 3 of ASD Inc. Colorado, USA), using the contact probe. The spectral reflectance was measured for wavelengths between 350 and 2500 nm, in an increment of 1 nm. Figure 1: Location of the study site. 2.4 Calculation of spectral vegetation indices The measured wheat crop spectral reflectance was used to calculate 22 spectral vegetation indices. The vegetation indices tested in this study are presented in Table 1. Page 59

3 Description Formula Reference RED Red band NIR NIR band MIVI Middle Infrared-Based Vegetation (MIR1 RED)/(MIR1 + RED) Thenkabail et Where MIR1 (TM5): 1550 to 1750 nm al., 2002 NDVI 705 Red Edge Normalized Difference Vegetation (ρ ρ 705)/( ρ ρ 705) mndvi 705 Modified Red Edge Normalized Difference Vegetation (ρ ρ 705)/( ρ ρ 705 2ρ 445) SIPI Structure Independent Vegetation ρ ρ 445)/( ρ ρ 445) PSRI Plant Senescence Reflectance (ρ ρ 500)/(ρ 750) NDWI Normalized Water Difference (ρ ρ 1241)/( ρ ρ 1241) MSI Moisture Stress (ρ 1599)/( ρ 819) NDVI Normalized Difference Vegetation (ρ NIR - ρ RED)/(ρ NIR + ρ RED) PRI Photochemical Reflectance (ρ ρ 570)/( ρ ρ 570) WBI NIR band (ρ 900)/( ρ 970) msr 705 Modified Red Edge Simple Ratio (ρ 750 ρ 445)/(ρ 705 ρ 445) VOG3 Vogelmann Red Edge 3 (ρ 734 ρ 747)/(ρ 715 ρ 720) NDNI Normalized Difference Nitrogen [log(1/ρ 1510) log(1/ρ 1680)]/[log(1/ρ 1510) + log(1/ρ 1680)] NDLI Normalized Difference Lignin [log(1/ρ 1754) log(1/ρ 1680)]/[log(1/ρ 1754) + log(1/ρ 1680)] NDII Normalized Difference Infrared (ρ 819 ρ 1649)/(ρ 819 ρ 1649) NWI-1 Normalized Water -1 (R 970 R 900)/(R R 900) NWI-2 Normalized Water -2 (R 970 R 850)/(R R 850) NWI-3 Normalized Water -3 (R 970 R 880)/(R R 880) NWI-4 Normalized Water -4 (R 970 R 920)/(R R 920) REIP Red Edge Inflection Point {[(ρ ρ 782)/2] ρ 702}/( ρ 738 ρ 702) ENVI EX User s Guide, 2009 Gutierrez et al Herrmann et al., Leaf Area (LAI) measurements LAI measurements were made on the same dates of wheat sampling for spectral reflectance measurements (i.e. on February 17 th, and April 5 th, 2012) using the Plant Canopy Analyzer (Model: PCA 2200 of LI-COR Biosciences, USA). At each sampling point, one above canopy and five below canopy readings were recorded to compute a single LAI value. 2.6 Wheat crop yield data collection Wheat grain yield was recorded (t/ha) by harvesting each experimental unit separately using a combine harvester. 3. RESULTS AND DISCUSSION 3.1 Assessment of the calculated spectral vegetation indices vegetation indices were calculated from the measured spectral reflectance of wheat crop samples representing four irrigation levels and three fertilizer levels on two sampling dates (February 17 th, and April 5 th, 2012). To assess the response of the calculated spectral vegetation indices to the irrigation and fertilizer levels applied to wheat crop, the collected data were subjected to statistical analysis (ANOVA). Vegetation indices that showed significant response to irrigation and/or fertilizer treatments are presented in Table 2. It was observed that many of the tested spectral indices showed significant response to irrigation levels. But, only two spectral indices showed significant response to fertilizer levels (Plant Senescence Reflectance -PSRI and the Photochemical Reflectance -PRI). The results presented in Table 2 revealed that the Middle Infrared-Based Vegetation showed a significant response to the implemented irrigation Page 60

4 levels for both sampling dates. PSRI and PRI were sensitive to both irrigation and fertilizer levels. Pr>F Irrigation Levels Fertilizer Levels Indices Feb. 17 th Apr. 5 th Feb. 17 th Apr. 5 th RED NIR MIVI NDVI SIPI PSRI NDWI NDVI PRI WBI NWI NWI NWI NWI REIP Table 2: Response of vegetation indices to irrigation and fertilizer levels. 3.3 Relationship between spectral indices (VI) and leaf area index (LAI) Leaf area index (LAI) was measured on the same sampling dates (February 17 th, and April 5 th, 2012). The data on the relationship between the various spectral vegetation indices and the leaf area index (LAI) are presented in Table 3. R 2 Feb. 17 Apr. 5 Feb. 17 R 2 Apr. 5 RED NWI NIR NWI MIVI NWI NDVI NWI SIPI REIP PSRI msr NDWI mndvi MSI VOG NDVI NDNI PRI NDLI WBI NDII Table 3: R 2 values between vegetation indices and LAI. The Normalized Difference Infrared (NDII) on both sampling dates and the Normalized Difference Nitrogen (NDNI) on February 17 th, exhibited the highest R 2 values (Figure 3). Similarly, Gupta et al. (2006) also reported lower R 2 values ( ) between wheat crop LAI and spectral indices such as Ratio Vegetation (RVI), Normalized Difference Vegetation (NDVI), and Soil-Adjusted Vegetation (SAVI). While, Haboudane et al. (2004) reported higher R 2 ( ) between wheat crop LAI and spectral indices such as Modified Chlorophyll Absorption Ratio index 2 (MCARI2), and Modified Triangular Vegetation (MTVI2). VI NDII Feb NDII Apr 5 3- NDNI Feb 17 y 2 = x R 2 = y 1 = x R 2 = y 3 = x R 2 = LAI Figure 3: Relationship of LAI with NDII and NDNI. 3.4 Relationship between spectral indices and crop yield The vegetation indices, calculated from wheat crop spectral reflectance measurements taken on February 17 th, and April 5 th, 2012, were correlated with wheat grain yield. From the results presented in Table 4, it was observed that, among the tested indices, five indices showed the most response to wheat grain yield with the highest R 2 in the range to These five indices included: Near Infrared band (NIR) and Water Band (WBI) for February 17 th (Figure 4) Normalized Water -1 (NWI-1), Normalized Water -3 (NWI-3), and Normalized Water -4 (NWI-4) for April 5 th (Figure 5). Page 61

5 RED NIR MIVI NDVI 705 SIPI PSRI NDWI MSI NDVI PRI WBI R 2 R 2 Feb. 17 Apr. 5 Feb. 17 Apr NWI-1 NWI-2 NWI-3 NWI-4 REIP msr 705 mndvi 705 VOG3 NDNI NDLI NDII Table 4: Relationship between vegetation indices and crop yield. VI Yield (ton/ha) Figure 4: Relationship of wheat grain yield with WBI and NIR. VI y WBI = x R 2 = y NIR = x R 2 = Figure 5: Relationship of wheat grain yield with NWI-1, NWI-3, and NWI-4. NIR WBI 3- NWI-4 1- NWI-1 2- NWI y 1 = 0.003x R 2 = y 2 = 0.003x R 2 = Yield (ton/ha) y 3 = x R 2 = CONCLUSIONS Many of the tested spectral indices showed significant response to irrigation levels. But, only two spectral indices showed significant response to fertilizer levels (Plant Senescence Reflectance -PSRI and the Photochemical Reflectance - PRI). The Middle Infrared-Based Vegetation (MIVI) showed a significant response to the implemented irrigation levels for both sampling dates. The Plant Senescence Reflectance (PSRI) and the Photochemical Reflectance (PRI) were sensitive to both irrigation and fertilizer levels. Among the tested spectral indices, higher R 2 values with LAI were observed with Normalized Difference Infrared (NDII) and the Normalized Difference Nitrogen (NDNI). Higher R 2 values with wheat grain yield were observed with Near Infrared band (NIR), Water Band (WBI), Normalized Water -1 (NWI-1), Normalized Water -3 (NWI-3), and Normalized Water -4 (NWI-4). 5. ACKNOWLEDGEMENTS This research work was carried out under the project (10 SPA ) funded by the National Plan for Science and Technology (NPST) through King Abdul- Aziz City for Science and Technology (KACST). The assistance provided by the graduate students, namely, Eng. Mohammed Elsiddig Ali Abass, Eng. Ahmed Galal Kaiad and Eng. Ahmed Hassan Zeyada in the field was quite valuable. The unstinted cooperation and support extended by Mr. Jack King and Mr. Alan King in carrying out the research work are gratefully acknowledged. 6. REFERENCES N. Aparicio, D. Villegas, J. Casadesus, J.L. Araus and C. Royo, "Canopy reflectance indices: A new tool for assessing durum wheat LAI and yield," In : C. Royo (ed.),m. Nachit (ed.), N. Di Fonzo (ed.), J.L. Araus (ed.), Durum wheat improvement in the Mediterranean region: New challenges,zaragoza: CIHEAM, p , Page 62

6 B. Duchemin, R. Hadria, S. Erraki, G. Boulet, P. Maisongrande, A. Chehbouni, R. Escadafal, J. Essahar, J.C.B. Hoedjes, M.H. Kharrou, S. Khabba, B. Mougenot, A. Olioso, J.C. Rodriguez, and V. Simonneaux, Monitoringwheatphenology and irrigation in Central Morocco: on the use of relationships between evapotranspiration, crops coefficients, leaf area index and remotely sensed vegetation indices, Agricultural Water Management (79): 1 27, ENVI EX User s Guide, Retrieved on October 28 th, 2013, from: NVI_EX_User_Guide.pdf R.K. Gupta, D. Vijayan, and T.S. Prasad, The relationship of hyper-spectral vegetation indices with leaf area index (LAI) over the growth cycle of wheat and chickpea at 3 nm spectral resolution, Advances in Space Research (38): , M. Gutierrez, M.P. Reunolds, and A.R. Klatt, Association of water spectral indices with plant and soil water relations in contrasting wheat genotypes, Journal of Experimental Botany 61(12): , D. Haboudane, J.R. Millera, E. Patteyc, P.J. Zarco- Tejadad, and I.B. Strachane, Hyperspectral vegetation indices and novel algorithms for predicting green LAI of crop canopies: Modeling and validation in the context of precision agriculture, Remote Sensing of Environment (90): , 2004). J.L. Hatfield and J.H. Prueger, "Value of using different vegetative indices to quantify agricultural crop characteristics at different growth stages under varying management practices," Remote Sensing (2): , I. Herrmann, A. Pimstein, A. Karnieli, Y. Cohen, V. Alchanatis, D.J. Bonfil, Assessment of leaf area index by the red-edge inflection point from VENμS bands, Proceedings Hyperspectral Workshop, Frascati, Italy, March P.S. Thenkabail, R.B. Smith, and E. De Pauw, Evaluation of narrow-band and broad-band vegetation indices for determining optimal hyperspectral wavebands for agricultural crop characterization, Photogrammetric Engineering & Remote Sensing 68(6): , A. Vina, A.A. Gitelson, A.L. Nguy-Robertson, and Y. Peng, Comparison of different vegetation indices for the remote assessment of green leaf area index of crops, Remote Sensing of Environment (115): , Page 63