Nitrogen fertilization of intercropped cereal-legume: A potassium, sulfur, magnesium and calcium plant acquisition dataset

Cereal-legume mixture is a well-known successful intercrop model for an efficient use of soil nutrients [1,2]. Effects of mineral N gradient on the acquisition of major nutrients: potassium (K), calcium (Ca), magnesium (Mg) and sulfur (S) is presented. A greenhouse pot experiment was conducted with wheat (Triticum aestivum L. cv. Lennox) and white lupin (Lupinus albus L. cv. Feodora) grown as sole crops and intercropped along a soil mineral N gradient obtained by 15N addition. Plants were harvested at flowering stage and dry weights of shoots and roots were measured. Potassium, calcium, magnesium and sulfur concentrations in shoots and roots were determined by Inductively Coupled Plasma Mass Spectrometry (ICP-MS).


Specifications
Agricultural sciences Specific subject area Agronomy and crop sciences Type of data Table  figure  How data were acquired Dataset was acquired after a greenhouse experiment and by using Inductively Coupled Plasma Mass Spectrometry for plant tissue analysis Data format Raw data Analysed data Parameters for data collection Mediterranean soil Description of data collection Data collection was acquired in two step: Step 1. A greenhouse pot experiment during 61 days, using a full factorial design (7 replicates) with three N treatments (N1 = 2, N2 = 33 and N3 = 65 mg N kg −1 dry soil added as a 15 N-labelled urea) and three crop treatments: wheat (Triticum aestivum L. cv. Lennox) grown as sole crop; white lupin (Lupinus albus L. cv. Feodora) grown as sole crop. Step

Value of the Data
• Intercropping is an agroecological solution emerging as an alternative to dominant systems requiring high inputs of nutrients. • This dataset can be used to identify critical N dose shifting plant interactions.
• This dataset can be used to design field trials to phenotype plants for their ability to coexist under resource gradient. • Plant breeders and cropping systems designers are the panel targeted by this dataset in order to identify multi-nutrient acquisition trait trade-offs in intercropped systems.

Differences of nutrient accumulation of mono and intercropped species along nitrogen gradient
Within each soil N treatment ( Table 3 ), intercropping systematically induced a positive effect on wheat and a negative effect on lupin, whether in terms of nutrient concentration or accumulation ( Figs. 1 and 2 ; Table 1 ). Except for Mg in roots at N3, intercropped wheat systematically and significantly accumulated more K, Ca, Mg and S in both shoots and roots, whatever the N treatment ( Fig. 1 a, Tables 2 and 3 ) compared to sole wheat. In contrast, intercropped lupin exhibited lower significant nutrient accumulation observed for shoots ( Fig. 1 b, Tables 2 and 3 ) compared to sole lupin. Except for Ca and Mg at N1, a significant decrease of K, Ca, Mg and S accumulation in shoots was observed for intercropped lupin compared to sole lupin, whatever the N treatment. In particular, sulphur accumulation were lower for intercropped S shoots whatever N treatment ( Fig. 1 a and Table 3 ) whereas lupin intercropped roots were only affected at N2 and N3 compared to sole crop. In contrast, S accumulations in intercropped wheat were sys- Values are the mean of 7 replicates. Error bars represent standard errors. Stars stand for significant differences between crop treatments within each N treatment (Tukey's test at p < 0.05). tematically greater than those of sole wheat at all N levels ( Fig. 1 a, Table 3 ). These results were consistent with the greater dry weight of shoots and roots of intercropped wheat is observed compared to sole wheat, whatever the N treatment, and lower shoot dry weight of intercropped lupin at N2 and N3 ( Fig. 3 ) compared to sole lupin.

Plant tissue concentration
The only significant differences between crop treatments for K concentrations in wheat were observed at N3 ( Fig. 2 a, Tables 2 and 3 ). In contrast, for all N treatments, K concentrations in lupin shoots significantly decreased when intercropped with wheat, whereas K concentrations in roots of intercropped lupin significantly decreased only at N3 ( Fig. 2 b, Table 3 ). S concentration was significantly reduced in lupin shoots and roots when intercropped with wheat, except for roots at N1 ( Figs. 1 b and 2 Table 3 ). A significant effect of the intercropping on S concentration of wheat shoots was only observed at N3 ( Fig. 2 a, Table 3 ). In terms of Ca concentrations, the only significant differences were observed for wheat roots, whatever the N treatment ( Fig. 2 a,  Table 3 ). For Mg concentrations, significant differences between crop treatments were only observed for wheat shoots at N2 and N3 ( Fig. 2 a), and for lupin shoots at N3 ( Fig. 2 b). Finally, like for K and S concentrations, Mg concentrations in shoots of intercropped wheat were significantly greater than those of sole wheat only at higher levels of soil mineral N.

Experimental Design, Materials and Methods
A calcareous cambisol was collected (0-15 cm depth) from the INRA experimental station located at Mauguio in Southern France (3 °59 6 E, 43 °37 13 N) three months after pea ( Pisum sativum L.) harvest. The soil was sieved at 1-cm to remove any coarse organic material, air dried and stored in sealed buckets at ambient temperature until the establishment of the greenhouse experiment. Before greenhouse experiment, the soil was re-sieved with a 4 mm mesh and then mixed with perlite to allow better conditions for root development. Pots (4 dm 3 ) were filled with 2 kg of soil and 0.2 kg of perlite. The final soil + perlite mixture exhibited the following mean properties: clay 202 g kg −1 , silt 454 g kg −1 , sand 343 g kg −1 , total CaCO 3 39 g kg −1 , pH w 8.3, CEC Metson 15 cmol + kg −1 , organic C 10.4 g kg −1 , total N 0.95 g kg −1 , ammonium (N -NH 4 + ) 20 mg kg −1 , nitrate (N -NO 3 − ) 39 mg kg −1 , inorganic available P (Olsen extraction method) 33.9 mg kg −1 .
The greenhouse pot experiment was conducted at Center Mondial de l'Innovation (Roullier Group -Saint-Malo, France) during 61 days, using a full factorial design with three N treatments (N1 = 2, N2 = 33 and N3 = 65 mg N kg −1 dry soil added as a 15 N-labelled urea) and three crop treatments: wheat ( Triticum aestivum L. cv. Lennox) grown as sole crop; white lupin ( Lupinus Table 3 Results ( p -values) of one-way ANOVA (crop treatment as factor meaning intercropped or sole crop) performed on potassium (K), calcium (Ca), magnesium (Mg) and sulfur (S) accumulation or concentration in shoots and roots of wheat and lupin as sole crops or as intercrops within each N treatment (N1, N2, N3). Number of stars (one, two or three) indicates the test significance level ( p < 0.05, p < 0.01 or p < 0.001, respectively).  albus L. cv. Feodora) grown as sole crop; wheat-white lupin intercrop. In sole crop treatment, 12 seeds of wheat or 6 seeds of white lupin were sown per pot and thinned to 4 and 2 individuals, respectively, after emergence. For the intercrop treatment, seeds were sown at half the densities of those achieved for sole crops, i.e . 2 individuals of wheat and 1 individual of white lupin per pot. Pots were arranged in a complete randomized design with seven replicates. Throughout the course of the experiment, pots were automatically and daily watered to maintain water content at 75% of Water Holding Capacity (WHC = 28%) of the soil + perlite mixture. The air temperature in the greenhouse was 22.7 ± 2.1 °C, and the incident photosynthetically active radiation was 200 μmol s −1 m −2 on average, with a 16 h photoperiod. Plants (shoots and roots) were harvested at 61 days after sowing, when most individuals of both species were in full flowering stage. For intercropped treatments, the root systems of the two species were carefully separated by hands to avoid excessive breakage. Within a pot, shoots or roots from all plants of a given species were assembled to constitute a unique shoot or root sample per species. Roots were carefully rinsed with deionized water in order to eliminate remaining soil particles. The dry weights of shoots and roots were measured after drying at 70 °C for 3 days. Potassium, Ca, Mg and S concentrations in shoots and roots were determined by Inductively Coupled Plasma Mass Spectrometry (ICP-MS) after digestion of 100 mg of finely ground material in a microwave oven (Multiwave PRO, Anton Paar, Austria) with concentrated HNO 3 (65%) at 200 °C.
The effects of crop (intercropped or sole crop) and N treatments (N1, N2, N3) on K, Ca, Mg and S concentrations and accumulation in shoots and roots were initially tested by twoway ANOVA. In absence of interaction between factors (crop treatment x N treatment), one-way ANOVA was performed to test differences between crop treatments within a given N treatment. When necessary, data were squared root or log-transformed to cope with the ANOVA requirements. Significant differences between means were tested by Tukey's multiple comparison tests ( p < 0.05). All statistical analyses were performed with R software v. 3.5.1 (R Core Team, 2018). The additional packages "car" and "agricolae" were used for ANOVA and Tukey's tests, respectively.