J. Hort. Sci. Vol. 1 (1): 28-32, 2006 Development of diagnostic leaf nutrient norms and identification of yield limiting nutrients using DRIS in rose grown under protected conditions K. Anjaneyulu Division of Soil Science & Agricultural Chemistry Indian Institute of Horticultural Research Hessaraghatta Lake Post, Bangalore-560 089, India E-mail: anjaney@iihr.emet.in ABSTRACT Leaf samples collected from protected cultivation units of rose around Bangalore (Karnataka) and Hosur (Tamil Nadu), when flower buds were at pea size, were processed and analyzed for various nutrients and thus, the data bank was established. By using Diagnosis and Recommendation Integrated System (DRIS), nutrient expressions, which have shown higher variance and lower coefflcient of variation, were selected as norms viz, N/P(13.02), K/N(0.85), N/S(11.10), P/S(0.853), K/P(11.0), N/Ca(2.18), N/Mg(7.18) etc. In addition, five nutrient ranges have been derived using mean and standard deviation as low, deficient, optimum, high and excess for each nutrient to serve as a guide for diagnostic purpose. The optimum N ranged from 2.53 to 2.96% , P from 0.19 to 0.23%, K from 2.23 to 2.72%, Ca from 1.15 to 1.59%, Mg from 0.41 to 0.55% and S from 0.21 to 0.27%. Among the micronutrients, the optimum Zn ranged from 28 to 64 ppm, Fe from 176 to 240 ppm, Mn from 107 to 175 ppm and Cu from 13 to 21 ppm for roses under protected conditions. The diagnosis of nutrient imbalance (DRIS indices) indicated that the most common yield limiting nutrients were potassium and magnesium among the macronutrients and iron and zinc among the micronutrients in protected cultivation units of rose. Key words: Nutrients, norms, DRIS indices, rose, protected conditions INTRODUCTION Nadu, 320 leaf samples were collected from various With the introduction of liberalization policy in P™tected cultivation units of rose during the years 2002 agriculture, rose cultivation in poly-houses became popular ^^^ 2003. At each site, composite samples of recently in India especially in Karnataka and Tamil Nadu for the matured 5*leaflets (Jones et al, 1991) were collected from export of quality cut flowers. As the nutrient removal is 50 plants to develop leaf nutrient norms/standards. It is very high under protected cultivation due to removal of long essential to select a specific part of the same physiological stalks/stems for export purpose, its nutrient requirements age at a definite location on the plant at definite stage of have to be carefully monitored through plant/leaf analysis growth (when flower buds were at pea size). for high productivity and quality. For efficient fertilizer r̂ j >. i • ^ <- , , r \ • , \ A A • , , Decontamination and Analysis of Samples programme, leaf nutrient standards are required to be J J r developed, as no established norms are available for roses Tissue samples were decontaminated by washing grown under protected conditions. There are a few reports first with tap water to remove the dirt or soil, then in 0.2% in the literature on the use of Diagnosis and detergent solution and in N/10 HCl solution to remove Recommendation Integrated System (DRIS) in ornamental residues of chemical spray materials on the leaf surface plants ( Mourao Filho 2004)^ Therefore, an investigation ^^^^^^^ j , ^^^^ .^ ĵ ,g ^^^ ^^^^^^ ^j^^jjl^^ ^^^^^ was earned out to develop leaf nutnent norms using DRIS, , „ , , r, 7 , • ,̂ r̂ -,̂ T̂ which in addition predicts the yield-limiting nutrients in (^hargava and Raghupathi, 1993). Excess water on the the order of importance (Beaufils, 1973). surface of the leaves was removed by pressing between the MATFRIAI AND METHODS ^"''̂ ^ °^ blotting paper and the leaves were dried in an oven at 70-75°C for 48 h. After complete drying, the samples Establishment of Data bank impling In a survey conducted in Karnataka and Tamil analyzed for different nutrients (except nitrogen) by were powdered in a Cyclotec Mill, and were stored in Leaf Sampling polycarbyl containers for analysis. The samples were mailto:anjaney@iihr.emet.in Anjaneyulu digesting Ig tissue in di-acid mixture (9:4 ratio of nitric acid and perchloric acid) by using standard analytical methods (Jackson, 1973). Nitrogen was estimated by micro- kjeldhal method whereas phosphorus, potassium and sulphur by vanado-molybdate, flame-photometer and turbidity methods, respectively. Calcium, magnesium and the micronutrients Fe, Mn, Cu and Zn were analyzed by using Atomic Absorption Spectrophotometer (Perkin- Elmer-A-Analyst-200). Thus, the data bank was established for the whole population. Computation of DRIS norms By using DRIS, the whole population was divided into two sub-groups namely low and high yielding (Beaufils, 1973; Bailey et al, 1997; Walworth and Sumner, 1987) based on the production of number of flowers/plant/year. From the experience of the growers, 175 flowers/plant/year was considered as the cut off yield between low and high performing gardens. The cut off yield was positioned in such a way that the high yielding sub-population reflects conditions that are deemed desirable (Beaufils, 1973). However, Letzsch and Sumner (1984) have shown that the actual cut off value used has little effect on the norms developed as long as it is not too low. Each parameter was expressed in as many forms as possible, e.g. N/P, P/N, NxP etc. A X^-test was run for each form of expression to ensure that each sub-population is normally distributed after which the mean, variance, standard deviation for all forms of expressions were calculated together with the coefficient of variation. Expressions having comparable means (by T- test) and significant variance ratios between the two sub- populations were retained as being discriminatory. Among different forms of expressions, the one showing higher variance ratio (Variance of low yielding / variance of high yielding) was selected as norm (Letzsch, 1985; Walworth and Sumner, 1987). DRIS indices DRIS provides a means of ordering nutrient ratios into meaningful expressions in the form of indices. The DRIS indices were calculated as described by Walworth and Sumner (1987) by using the following formula, as an example for one nutrient is shown below: N = l/10[-f (P/N)-f(K/N)+f(N/Ca)+f(N/Mg)-f(S/N)-f(Fe/ N)-i-f(N/Mn)-Hf(N/Zn)- f(Cu/N)-i-f(N/dw)] and f(N/P) = Where, f (N/P) = N/P - 1 n/p 1000 1 - n/p N/P 1000 cv when N/P < n/p cv hen N/P > n/p where N/P, the actual value of the ratio of N and P in the plant under diagnosis; n/p, the value of the norm (which is the mean value of the high yielding unit); CV, coefficient of variation for population of high yielding units. Similarly, the indices for other nutrients have been calculated using appropriate formulae. The absolute sum values of the nutrient indices generate an additional index called NBI, the nutritional balance index (Walworth and Sumner, 1987). Leaf nutrient guides/ranges derived from mean and standard deviation Five leaf nutrient guides/ranges have been derived using mean and standard deviation as deficient, low, optimum, high and excess for each nutrient. The optimum nutrient range is the value derived from "mean -4/3SD (standard deviation) to mean + 4/3SD". The range "low" was obtained by calculating "mean - 4/3 SD to mean - 8/ 3SD" and the value below "mean - 8/3 SD" was considered as deficient. The value from "mean •¥ 4/3 SD to mean + 8/ 3 SD" was taken as high and the value above "mean -i- 8/3 SD" was taken as excessive (Bhargava and Chadha, 1993). RESULTS AND DISCUSSION Leaf nutrient concentrations The leaf nutrient status under different protected cultivation units varied differently. For the entire population of roses, the leaf N concentration varied from 2.16 to 3.11% indicating that the nitrogen content was low at least in some of the low yielding gardens when compared to optimum value. P ranged from 0.138 to 0.266% whereas potassium varied from 1.15 to 2.80% for the entire population (Table 1). Among the micronutrients, Fe ranged from 110 to 268 ppm and was generally considered high but was not high under protected cultivation as most of the plants were showing deficiency symptoms in the low yielding gardens though the lowest iron concentration recorded was 110 ppm in a garden, which is normal for many crops. DRIS ratio norms Forty-five nutrient expressions chosen as diagnostic norms from high yielding population are presented in J. Hort. Sci. Vol. 1(1): 28-32, 2006 29 DRIS norms for rose grown under protected conditions Table 1. Mean and range of leaf nutrient concentrations for the wliole population Nutrient N P K Ca Mg S Fe Mn Zn Cu Table 2. Unit % % % % % % ppm ppm ppm ppm Range 2.16-3.110 0.14-0.266 1.15-2.800 0.68-1.770 0.22-0.560 0.17-0.320 110-268 38-193 18-136 02-022 Mean 2.558 0.198 2.194 1.179 0.358 0.229 174 97 43 10 DRIS ratio norms for rose grown under conditions Selected Ratios N/P K/N N/Ca N/Mg N/S N/Fe N/Mn N/Zn Cu/N K/P P/Ca P/Mg P/S P/Fe P/Mn P/Zn Cu/P K/Ca K/Mg K/S Fe/Zn Mn/Zn Cu/Zn Norms 13.020 0.853 2.182 7.183 11.110 0.015 0.030 0.069 4.304 11.000 0.170 0.556 0.853 0.001 0.002 0.005 56.070 1.873 6.155 9.467 4.505 2.397 0.252 CV (%) 14 16 24 28 19 29 45 48 47 15 29 27 12 29 42 46 47 32 31 23 45 48 35 Selected Ratios K/Fe K/Mn K/Zn K/Cu Ca/Mg S/Ca Ca/Fe Mn/Ca Ca/Zn Cu/Ca S/Mg Fe/Mg Mn/Mg Mg/Zn Cu/Mg S/Fe Mn/S S/Zn Cu/S Mn/Fe Cu/Fe Cu/Mn — Norms 0.013 0.026 0.059 0.293 3.324 0.201 0.007 83.120 0.031 8.850 0.657 483.800 262.200 0.009 28.260 0.004 438.400 0.006 46.800 0.552 0.059 0.112 — protected CV (%) 35 44 51 39 20 29 27 40 41 42 30 22 28 37 38 28 43 44 44 28 39 42 — Table 2. It was established that a particular nutrient expression should have a high variance and low coefficient of variation, to be chosen as diagnostic norm between high and low yielding population, for greater diagnostic precision (Walworth and Sumner, 1987 and Angeles et al, 1993). Based on this principle, these 45 nutrient expressions were chosen as diagnostic norms viz. N/P (13.02), K/N(0.85), N/S(n. 10), P/S(0.853), K/P(l 1.0), N/Ca(2.18), N/Mg(7.18) etc. and have shown lower CV values compared to others and may have greater physiological rationale (Raghupathi et al, 2004). Maintaining the ratios of some expressions at optimum when they were with large coefficient of variation was much less critical for the performance of the crop. Therefore, the nutrients considered as yield building components, need to be kept in a state of relative balance for maximum efficiency of dry matter production (Raghupathi et al, 2004). Identification of yield limiting nutrients by DRIS indices The nutrient imbalance in plants was diagnosed through DRIS indices. DRIS provides a means of ordering nutrient ratios into meaningful expressions in the form of indices. A DRIS index is a mean of the deviations of ratios containing a given nutrient from their respective normal or optimum values. Thus, the relative abundance of each nutrient was evaluated by comparing all ratios containing that nutrient (N/P, NxK, Ca/N etc.) with the DRIS norms. As the value of each ratio function was added to one index sub-total and subtracted from another prior to averaging, all indices were balanced around zero (Table 3). Thus, the nutrient indices sum to zero indicating an optimum level, negative values a relative deficiency and positive values a relative excess of that nutrient (Beverly, 1991). The yield limiting nutrients were differing from garden to garden though some of the nutrients were more prominent. The order in which the nutrients were limiting the yield indicated that most often more than one nutrient was limiting the yield (Table 3). However, the diagnosis of nutrient imbalance indicated that the most common yield limiting nutrients are found to be potassium and magnesium among the macronutrients and zinc and iron among the micronutrients in the protected cultivation units growing rose crop. Nutritional balance index (NBI) The absolute sum values of the nutrient indices generate an additional index called nutritional balance index (Walworth and Sumner, 1987). The overall imbalance of the nutrient was assessed based on the sum of indices irrespective of sign (Table 3). Higher the sum value, larger will be the plant nutritional imbalance and, therefore, the lower will be the yield. However, the yield cannot be predicted from sum of indices irrespective of the sign because of the influence of unmeasured factors that affect the yield but not included in the calculation of DRIS indices (Sumner, 1977). Leaf nutrient guide and classification of gardens Five leaf nutrient guides/ranges have been derived using mean and standard deviation as deficient, low, optimum, high and excess for each nutrient and presented (Table 4). The optimum leaf N for rose ranged from 2.53 to 2.96% but the variation in the range was not wide. However, Jones et al, (1991) reported a wider range (3 to 5%) for roses grown outside. The classification of the gardens J. Hort. Sci. Vol. 1 (1): 28-32, 2006 30 Anjaneyulu Table 3: DRIS indices for various nutrients in selected low yielding rose gardens S.No 1 2 3 4 5 6 N 166 -89 9 -27 38 4 P 109 151 46 26 64 -17 K -101 -16 -109 -174 28 135 Ca -72 -22 23 24 15 184 Mg -78 -152 -10 -10 -116 -154 S 227 -60 33 35 -64 10 Table 4. Leaf nutrients guide for rose under protected conditions Nutrient N P K Ca Mg S Unil % % % % % % t Deficient <2.08 <0.13 <1.75 <0.69 <0.25 <0.15 Low Fe -179 -23 -18 125 -72 -18 2.09-2.52 0.14-0.18 1.76-2.22 0.70-1 .14 0.26-0.40 0.16-0.20 Mn 9 2 52 32 178 5 Zn -89 88 -51 -146 -217 -155 Optimum 2.53-2.96 0.19-0.23 2.23-2.72 1.15-1.59 0.41-0.55 0.21-0.27 Cu 8 121 25 115 146 6 Sum order of limiting 1038Fe>K>Zn>Mg 724Mg>N>S>Fe 376K>Zn>Fe>Mg 714K>Zn>N>Mg 938Zn>Mg>Fe>S 688Zn>Mg>Fe>P High 2.97-3.39 0.24-0.27 2.73-3.17 1.60-2.03 0.56-0.69 0.28-0.33 nutrients Excess >3.40 >0.28 >3.18 >2.04 >0.70 >0.34 Fe Mn Zn Cu ppm ppm ppm ppm 305.00 >244.00 >93.00 >31.00 surveyed indicated that nitrogen was at optimum in 85% of the gardens whereas it was limiting only in 9% of the gardens (Table 5). The optimum P ranged from 0.19 to 0.23% indicating a lower requirement of P compared to N (Table 4). It was observed that P was rarely a limiting factor for flower production in roses. Thus, the P concentration was at optimum in 86% of the gardens and was higher in 11% of the gardens. The optimum K ranged from 2.23 to 2.72%. The requirement of K is always next only to nitrogen for roses as it is involved not only in the production of flowers but also in improving the quality. Potassium was found to be low in 21% of the gardens indicating that K was a yield-limiting factor. The optimum concentration range for calcium was from 1.15 to 1.59% and was similar to the results reported by Jones et al (1991). Among the gardens surveyed, 25% of the gardens were low in magnesium. Thus, magnesium status of many individual gardens was low when compared to optimum range (Table 4). Only 6% of the gardens had recorded lower sulphur Table 5. Classification of rose gardens into various nutrient categories Nutrient N P K Ca Mg S Fe Mn Zn Cu Deficient 0 0 2% 0 0 0 5% 0 5% 0 / Hort. Sci. Vol. 1(1): 28-32, 2006 Low 9% 3% 21% 8% 25% 6% 23% 0 30% 0 Optimum 85% 86% 74% 78% 72% 77% 66% 91% 62% 88% High 6% 11% 3% 14% 3% 14% 6% 9% 3% 7% Excess 0 0 0 0 0 3% 0 0 0 5% content indicating that sulphur was not a yield- limiting factor in most of the gardens. Jones ef a/ (1991) reported a wider optimum ranges for many nutrients for roses grown outside. Among the micronutrients, zinc was found to be low in 30% of the gardens whereas iron was low only in 23%of the gardens. The optimum zinc concentration ranged from 28 to 64 ppm whereas iron ranged from 176 to 240 ppm. Thus, zinc was found to be the most yield- limiting factor in most of the protected cultivation units followed by iron. Most of the gardens recorded optimum levels for manganese and copper. It can be concluded that the leaf nutrient standards developed can be used for efficient fertilizer programming and to correct the nutrients in question for obtaining optimum flower yield and quality in rose crop grown under protected conditions. REFERENCES Angeles, D.E., Sumner, M.E. and Lahav, E.1993. Preliminary DRIS norms for banana. J. PL Nutrition, 16:1059-1070 Bailey, J.S., Beattie, J.A.M. and Kilpatrick, D.J. 1997. The diagnosis and recommendation integrated system (DRIS) for diagnosis of nutrient status of grassland swords. 1 Model establishment. PL & Soil 197:127- 135 Beaufils, E.R. 1973. Diagnosis and recommendation integrated system (DRIS). Soil ScL BulL, 1:1-132. University of Natal Pitermariburg, South Africa. Beverly, R.B.1991. A practical guide to the DRIS. Athens: Micro-Macro Publ., USA, 87p Bhargava, B.S. andChadha, K.L.I 993. Leaf nutrient guides 31 DRIS norms for rose grown under protected conditions for fruit crops. In: Advances in Horticulture Fruit Crops, Vol. 2, pp 973-1030, Chadha, K.L. and Pareek,O.P. (eds.), Malhotra Publ. House, New Delhi. Bhargava, B.S. and Raghupathi, H.B.1993. Analysis of plant material for macro and micronutrients. In: Methods of Analysis of Soils , Plants, Waters and Fertilizers. pp 48-82, Tandon, H.L.S (ed.). Fertilizer Development and Consultation Organization, New Delhi. Jackson, M.L.1973. Soil Chemical Analysis, Prentice Hall of India, New Delhi. 498 p Jones, J.B. Wolf, B and Mills, H.A.I991.Plant Analysis Handbook. Micro-Macro Publ., USA Letzsch, W.S.I985. Computer program for selection of norms for use in the diagnosis and recommendation integrated system (DRIS). Commun. Soil Sci. PL Analysis, 16:339-347 Letzsch, W.S. and Sumner, M.E.1984. Effect f population size and yield level in selection of DRIS norms. Commun. Soil Sci. PI. Analysis, 15:997-1006 Mourao Filho, F.A.A. 2004. DRIS: Concepts and Applications on nutritional diagnosis in fruit crops, Sci. Agric, (Piracicaba, Braz.) 61:550-560 Raghupathi, H.B., Reddy,Y.T.N., Kurian Reju and Bhargava, B.S. 2004. Diagnosis of nutrient imbalance in mango by DRIS and PCA approaches. J. PI. Nutrition, 27:1131-1148 Sumner, M.E.I977. Application of Beaufils diagnostic indices to corn data published in literature irrespective of age and condition. PI. & Soil, 46:359- 363 Walworth, J.L. and Sumner, M.F.1987. The diagnosis and recommendation integrated system. In: Adv. Soil 5d., 6:149-188 (MS Received 3 April, 2006 Revised 24 May, 2006) J. Hon. Sci. Vol. 1 (1): 28-32, 2006 32