Zoning of “Protected Designation of Origin La Mancha Saffron” According to the Quality of the Flower

Abstract

The quality of Crocus sativus L. flowers, beyond their stigmas, is influenced by the presence of bioactive metabolites also in their floral bio-residues. Given the effect of climatic and soil variables on these bioactive compounds, the aim of this research was to develop an agroecological zoning of saffron crop areas within the Protected Designation of Origin (PDO) La Mancha region (Castilla-La Mancha, Spain) by integrating the floral metabolite content with climatic and soil variables. To achieve this, a total of 173 samples were collected during the 2022 and 2023 harvests and analyzed via RP-HPLC-DAD to determine crocins, picrocrocin, kaempferols, and anthocyanins. Two new indices, Cropi (crocins + picrocrocin) and Kaeman (kaempferols + anthocyanins), were defined to classify flowers into four quality categories (A–D). High-quality classifications (A and B) were consistently associated with plots grouped in the meteorological stations of Ontur, El Sanchón, and Bolaños, indicating favorable edaphoclimatic conditions and climatic parameters, such as moderate temperatures and reduced humidity, for metabolite biosynthesis. In contrast, plots included in the meteorological stations of Tarazona and Pedernoso were mostly assigned to lower categories (C and D). Spatial analysis using thematic maps revealed that areas with an intermediate carbonate content, less calcareous soils, and higher organic matter levels were linked to higher flower quality. These findings highlight the influence of soil characteristics and climate, with distinct seasonal contrasts, that positively influence metabolite synthesis and flower quality.

1. Introduction

Saffron (Crocus sativus L.) is currently one of the most representative traditional crops in the region of Castilla-La Mancha, particularly in the La Mancha area, where its production is closely linked to the region’s agricultural and cultural heritage. The saffron flower, composed of the stigmas and the remaining floral structures (tepals, style, and stamens), has significant economic and functional value due to its richness in bioactive metabolites such as crocins, picrocrocin, anthocyanins and kaempferols [1].

In Crocus sativus L., the appearance of flowers may occur before (hysteranthemum), simultaneously (synanthemum) or after the appearance of leaves. This phenological behavior is influenced, among others, by soil moisture levels and is usually triggered by rainfall in late summer or early autumn [2,3]. The agronomic cycle of saffron begins in early summer, when the corms are planted. Over the following campaigns, it will produce new corms that will bloom next autumn, while the original corm dies. Thus, the plant’s life cycle spans two years: one for growth and development, and the other for vegetative reproduction. During its cultivation, corms typically remain in the soil for several years, until high planting density and nutrient competition limit their growth, a period that depends on soil fertility. In Spain, for instance, corms are left buried for 4 to 5 years.

As a sterile plant, it reproduces exclusively through budding [4], which limits the rapid expansion of the crop due to its reliance on the availability of corms. This reproductive strategy, closely tied to soil conditions and nutrient storage within the corm, directly influences the physiological state of the flower produced in the subsequent campaign. Therefore, the metabolites detected in the flower reflect the quality of the corm and the nutrients of the previous year.

The stigmas, the most valued part of the flower, are used as a spice for their unique organoleptic properties: the intense red color attributed to crocins, the bitter taste from picrocrocin, and the characteristic aroma produced by safranal [5,6]. This latter compound, which plays a key role in the sensory profile of saffron, is formed during the traditional toasting process, an artisanal practice that distinguishes saffron from La Mancha to others produced. However, beyond the stigmas, the saffron floral bio-residues, often discarded, contain significant amounts of bioactive compounds, particularly kaempferols and anthocyanins, which are well known for their strong antioxidant activity [7]. These components hold great potential for applications in industries such as cosmetics, pharmaceuticals, and functional foods [8]. It is known that saffron quality is directly related to the concentration and proportion of these metabolites, so, it is essential to understand how edaphoclimatic factors influence its development in one of the most emblematic cultivation areas: Castilla-La Mancha Spanish region. The Crocus sativus L. crop is highly dependent on specific climatic factors that determine its suitability in different geographic zones. Thus, a mediterranean-continental climate, characterized by cold winters and hot, dry summers, supports the phenological cycle of the crop, allowing for adequate bulb dormancy and synchronized, high-quality flowering. This fact has led to the cultivation of saffron being traditionally linked to the Mediterranean basin. More specifically, optimal flowering temperatures range between 23 and 27 °C, while excessive summer humidity can compromise bulb health by promoting rot [9,10].

Previous studies have shown that factors such as ground elevation or soil permeability have a great effect on the correct development of the crop [11,12,13,14]. Altitude also plays a crucial role, as elevations between 600 and 1200 m offer favorable thermal variations and humidity levels [11]. As for the characteristics of the soil, these play a decisive role in identifying suitable areas for the cultivation of Crocus sativus L. In this line, this crop requires deep, loose soils with excellent drainage, as excess moisture can lead to corm rot [15]. Ideal textures include loam or sandy loam, which promote proper aeration and root development [15]. In relation to optimal pH values, the range between 6.5 and 7.5 [16] ensures the maximum availability of essential macro- and micronutrients. Also, a high organic matter content further enhances soil structure and water retention while supporting microbial activity, which has been associated with improved saffron quality [15]. Conversely, soil compaction restricts root penetration and nutrient uptake, while salinity—although moderately tolerated—can induce osmotic and nutritional stress [15].

The agroclimatic zoning of saffron crop enables the identification of optimal areas through the analysis of key edaphoclimatic variables (temperature, soil texture, altitude). This approach enhances the phenological adaptation of Crocus sativus L., increases production efficiency, and mitigates biotic and abiotic risks, promoting sustainable systems with high phytochemical quality. However, despite its importance, there is little research on saffron zoning. These studies have explored the agroclimatic zoning of Crocus sativus L. using multidisciplinary approaches. Kothari et al. [17] demonstrated that variables such as altitude and precipitation significantly influence yield and phytochemical quality in the Western Himalayas. In Iran, Yaghoubzadeh et al. [11] and Falahat et al. [18] applied multi-criteria decision analysis (AHP and ANP) integrated with GIS to identify suitable cultivation zones in Sarab, West Azerbaijan, and Semnan, highlighting the relevance of edaphoclimatic factors. Similarly, Bengouga et al. [19] evaluated saffron adaptability in Algerian mountain oases, attributing phenological variations to altitude and soil–climate conditions. These studies underscore the importance of zoning for precise and sustainable agronomic planning. According to data from the “Protected Designation of Origin (PDO) La Mancha Saffron”, the general trend is that the yields of saffron crops in the region are decreasing, hence the need for zoning of the crop. PDO is an official certification granted to agricultural or food products whose quality, characteristics, or reputation are directly linked to their geographical origin, and whose production, processing, and preparation take place entirely within that specific region.

In this context, the spatial zoning of saffron flowers in Castilla-La Mancha emerges as a strategic tool to identify areas with the greatest aptitude for producing high-quality flowers, optimize production resources, and promote the comprehensive use of the crop. Therefore, the main objective of this study is to establish a zoning for saffron cultivation in the region, integrating data on metabolite content and edaphoclimatic conditions to support decision-making for the sustainable management and value enhancement of the product.

2. Materials and Methods

2.1. Chemicals and Reagents

All solvents and reagents used were of HPLC purity or analytical grade. Acetonitrile (ACN) (CAS: 75-05-8), trifluoroacetic acid (TFA) (CAS: 7605-1), and hydrochloric acid (HCl) (CAS: 7647-01-0) were obtained from Panreac (Barcelona, Spain). Ultra-high-purity water was generated using a Milli-Q system (Millipore, Bedford, MA, USA).

Regarding the standards, crocetin esters, including trans-crocetin di-(β-D-gentiobiosyl) ester (trans-4-GG) (CAS: 55750-85-1) and trans-crocetin (β-D-glucosyl) -(β-D-gentiobiosyl) ester (trans-3-Gg) (CAS: 55750-84-0), quercetin-3-O-β-sophoroside (CAS: 18609-17-1), delphinidin-3,5-O-β-glucoside (CAS: 6906-38-3), delphinidin-3-O-β-glucoside (CAS: 6906-38-3), and malvidin-3,5-di-O-β-glucoside (CAS: 16727-30-3), were supplied by Phytolab GmbH & Co. KG (Erlangen-Höchstadt, Germany). Picrocrocin was synthesized in the laboratory following the method described by [20], while kaempferol-3-O-β-sophoroside (CAS: 19895-95-5) was purchased from BLD Pharmatech (Shanghai, China). To prepare for the standards, a stock solution was initially prepared, from which successive dilutions were carried out to obtain various calibration levels. The dilutions were made using a water/hydrochloric acid mixture (100:1, v/v), following the same conditions applied during the extraction process. All prepared solutions were stored at −20 °C.

2.2. Plant Material

A total of 173 Crocus sativus L. flower samples were collected from different places within the” PDO La Mancha Saffron” (Castilla-La Mancha, Spain) during the 2022 and 2023 harvesting seasons (Figures S1 and S2). Harvesting for the first campaign (2022) began in September 2021 and will continue until December 2022, while the second campaign (2023) will continue from October 2022 to December 2023.

In 2022, plots had corms that were one, two, three, and four years old. In 2023, most plots had corms in their second, third, and fourth years of development. The plots had a minimum homogeneity (0.1–3 m²). The sample fields are actual production fields in which random sampling was performed. From each homogeneous plot 3 sectors were made. From each sector 18 flowers were taken at random.

The freshly harvested flowers from each sector, which comprises both saffron flower and stigma, were freeze-dried for five days using a LyoAlfa 6-50 freeze-dryer (Telstar, Terrassa, Spain) to ensure complete dehydration, as indicated by a stable final weight. The freeze-drying process was conducted under control conditions of −50 ± 2 °C and a pressure of 10⁻³ mbar. After drying, the flowers were finely ground with a mortar to produce a uniform powder and stored at ambient temperature (18 ± 3 °C) in a silica gel-containing chamber until further analysis.

To evaluate moisture content, a halogen moisture analyzer (XM-120 T model, Cobos, Barcelona, Spain) was used, operating at 105 °C. The sample was considered to have reached a constant weight when the moisture loss remained below 0.1% for a duration of 180 s.

2.3. Extracts of Crocus Sativus L. Flowers

Crocus sativus L. flower extracts from each sector were obtained following the procedure described by Moratalla-Lopez et al. [20]. For this, 200 mg of freeze-dried and finely ground flower powder from each sample was combined with 25 mL of a water/HCl solution (100:1, v/v) and stirred at 500 rpm for 1 h in a dark environment. The mixture was then centrifuged at 3500 rpm for 5 min using a Selecta centrifuge (Barcelona, Spain). The resulting supernatant was passed through a hydrophilic polytetrafluoroethylene (PTFE) membrane filter with a 0.45 μm pore size (Millipore, Bedford, MA, USA). Finally, the filtered extract was transferred into a vial for analysis using Reversed-Phase High-Performance Liquid Chromatography with Diode Array Detection (RP-HPLC-DAD). The extracts from each plot were analyzed separately (n = 3), obtaining CV less than 10 in all cases.

2.4. Reversed-Phase High-Performance Liquid Chromatography with Diode Array Detection (RP-HPLC-DAD) Conditions

The metabolite content in Crocus sativus L. flowers was analyzed following the method developed by Moratalla-López et al. [21]. The RP-HPLC-DAD analysis was conducted using an Agilent 1200 HPLC system (Agilent, Palo Alto, CA, USA). A Develosil ODS-HG-5 chromatographic column (250 × 4.6 mm, 5 µm) from Teknokroma (Sant Cugat Del Vallès, Barcelona, Spain) was used, and the column temperature was maintained at 40 °C.

The mobile phase consisted of two solvents: ultra-high-purity water with 0.5% trifluoroacetic acid (TFA) (solvent A) and acetonitrile (ACN) (solvent B). The elution was performed using a gradient system, beginning with 100% solvent A. After 30 min, the proportion of acetonitrile increased to 30%, while solvent A decreased to 70%. Between 40 and 45 min, the acetonitrile content was raised to 80%, reducing solvent A to 20%. From 50 to 55 min, the acetonitrile percentage was lowered to 20%, with solvent A increasing back to 80%. Finally, between 55 and 59 min, the system returned to its initial condition of 100% solvent A. The flow rate was set at 1.0 mL/min, and the injection volume for each sample was 30 μL.

For all the compounds, identification was carried out using the DAD by comparing their corresponding UV–visible spectra and retention times with those of pure standards. The DAD detector was set to 440 nm for the identification of crocetin esters, while picrocrocin and HTCC (hydroxy-2,6,6-trimethyl-1-cyclohexen-1-carboxaldehyde) were detected at 250 nm, flavonols at 266 nm, and anthocyanins at 520 nm. Quantification was performed using calibration curves for the corresponding commercial standards at five different concentrations, based on UV–visible signals (R² = 0.9900–0.9999). All analyses were performed in triplicate, with two measurements for each sample replicate [22].

2.5. Climatic Data

Climatic data were collected from meteorological stations close to the sampled plots. The data were extracted from the Agroclimatic Information System for Irrigation (SIAR). Daily climatic information was obtained, and monthly averages were made. As the samples were from the 2022 and 2023 seasons, the climate data collected corresponds to the period September 2021-December 2022 for the 2022 season, and for the 2023 season, it covers the period October 2022-December 2023. The parameters collected include temperature (mean, maximum and minimum), precipitation and relative humidity (maximum and minimum).

2.6. Maps

To visualize the spatial variability of saffron flower quality across the PDO La Mancha region, geostatistical interpolation techniques were applied. Specifically, Inverse Distance Weighting (IDW) was employed to estimate values for unmeasured locations based on the proximity of surrounding measured points. This method was chosen for its ability to generate smooth and interpretable spatial surfaces, allowing a detailed representation of the distribution patterns of the Cropi (crocins + picrocrocin) and Kaeman (kaempferols + anthocyanins) indices throughout the territory. IDW interpolation was applied independently for the 2022 and 2023 campaigns, using the centroid coordinates of each sampling site linked to the corresponding meteorological station.

This approach facilitated the generation of continuous maps that reflect the relative concentration of bioactive metabolites in saffron flowers with a high degree of spatial resolution. All maps were generated using ArcGIS PRO software (version 3.0.0, Esri Inc., Redlands, CA, USA) under an academic license provided by the University of Castilla-La Mancha. The resulting thematic maps were used not only to illustrate the classification of quality categories (A–D) per station, but also to identify spatial patterns and clusters of high or low agroecological suitability for saffron cultivation across the region.

For the comparative study, sampling plots were assigned to the corresponding meteorological stations by proximity.

2.7. Data Analysis

Statistical analyses were performed with IBM SPSS Statistics software (version 29.0, IBM Corp., New York, NY, USA) and Statgraphics Centurion statistical program (version 19.4.02; StatPoint, Inc., The Plains, VA, USA).

The data of metabolites determined by RP-HPLC-DAD was examined using one-way analysis of variance (ANOVA) at a 95% probability level, according to Tukey’s test, to determine the differences between plots grouped according to the meteorological station using IBM SPSS Statistics software. The Shapiro–Wilk test was applied to confirm the normal distribution of the data before performing the ANOVA analysis.

Pearson’s correlation analysis, performed with and Statgraphics Centurion statistical program, was carried out to investigate the relationship between flower metabolites content and climatic parameters.

3. Results and Discussion

3.1. Crocus Sativus L. Flower Metabolite Content

The content of the metabolites identified in saffron flowers by RP-HPLC-DAD are summarized in

Table 1. Content (dry weight) of picrococin, HTCC and crocins of Crocus sativus L. flowers from different locations from 2022 season.

Meteorological Stations	Picrocrocin	HTCC	trans-5-tG	trans-5-ng	trans-4-GG	trans-4-ng	trans-3-Gg	trans-2-gg	cis-4-GG	trans-2-G	cis-3-Gg	cis-2-gg	cis-2-G	Cropi
	mmol/kg	mmol/mg	mmol/kg	mmol/kg	mmol/kg	mmol/kg	mmol/kg	mmol/kg	mmol/mg	mmol/kg	mmol/kg	mmol/kg	mmol/kg	mmol/kg
Albacete	41.13 ± 7.49	49,574.20 ± 19,324.77	0.39 ± 0.33	0.27 ± 0.01	26.29 ± 1.51	0.33 ± 0.01	51.98 ± 5.21	0.46 ± 0.08	1003.80 ± 25.48	3.64 ± 0.54	0.89 ± 0.12	0.12 ± 0.08	0.08 ± 0.01	182.33 ± 13.53
Albacete	55.96 ± 16.61	28,715.30 ± 5524.71	0.29 ± 0.24	0.25 ± 0.01	22.25 ± 2.49	0.22 ± 0.02	52.81 ± 8.86	0.46 ± 0.23	765.91 ± 112.73	2.00 ± 0.29	0.78 ± 0.42	0.08 ± 0.11	0.07 ± 0.00	170.66 ± 26.69
Albacete mean	48.55 ± 12.05 a	39,144.75 ± 12,424.74 a	0.34 ± 0.28 c	0.26 ± 0.01 bc	24.27 ± 2.00 c	0.28 ± 0.01 b	52.39 ± 5.21 ab	0.46 ± 0.15 a	884.85 ± 69.10 a	2.82 ± 0.41 c	0.84 ± 0.27 a	0.10 ± 0.10 a	0.07 ± 0.00 a	176.49 ± 20.11 a
Bolaños	26.90 ± 2.42	46,365.45 ± 27,416.14	0.32 ± 0.23	0.24 ± 0.03	23.79 ± 2.22	0.24 ± 0.01	47.30 ± 2.43	0.48 ± 0.08	1152.13 ± 51.82	3.12 ± 0.04	0.98 ± 0.03	0.11 ± 0.0	0.10 ± 0.00	150.18 ± 3.46
Bolaños	112.73 ± 1.70	6622.03 ± 3882.18	0.34 ± 0.32	0.27 ± 0.00	26.93 ± 0.68	0.24 ± 0.01	70.68 ± 4.38	1.02 ± 0.11	1094.00 ± 115.64	2.59 ± 0.21	1.13 ± 0.62	0.09 ± 0.08	0.08 ± 0.00	255.66 ± 11.22
Bolaños mean	69.82 ± 2.06 a	26,493.74 ± 15,649.16 a	0.33 ± 0.27 bc	0.26 ± 0.01 abc	25.36 ± 1.45 c	0.24 ± 0.01 ab	58.99 ± 3.40 b	0.75 ± 0.10 a	1123.06 ± 83.73 ab	2.86 ± 0.12 c	1.06 ± 0.32 a	0.10 ± 0.08 a	0.09 ± 0.00 a	202.92 ± 7.34 a
La Gineta	44.38 ± 10.13	37,756.09 ± 39,695.86	0.32 ± 0.26	0.25 ± 0.01	22.86 ± 3.05	0.23 ± 0.01	58.60 ± 14.92	0.56 ± 0.14	1076.11 ± 545.79	1.99 ± 0.15	1.14 ± 0.08	0.08 ± 0.08	0.07 ± 0.00	166.41 ± 19.11
La Gineta mean	44.38 ± 10.13 a	37,756.09 ± 3969.86 a	0.32 ± 0.26 bc	0.25 ± 0.01 abc	22.86 ± 3.05 c	0.23 ± 0.01 ab	58.60 ± 14.92 b	0.56 ± 0.14 a	1076.11 ± 545.79 ab	1.99 ± 0.15 abc	1.14 ± 0.08 a	0.08 ± 0.08 a	0.07 ± 0.00 a	166.41 ± 19.11 a
Herencia	71.97 ± 5.15	32,090.31 ± 19,139.99	0.26 ± 0.32	0.25 ± 0.01	19.42 ± 0.33	0.22 ± 0.01	38.98 ± 6.51	0.84 ± 0.04	659.05 ± 32.86	1.63 ± 0.20	2.96 ± 0.09	0.12 ± 0.10	0.08 ± 0.00	203.01 ± 7.56
Herencia	18.99 ± 14.48	8757.78 ± 5319.09	0.27 ± 0.26	0.20 ± 0.01	15.79 ± 0.71	0.22 ± 0.01	38.52 ± 8.72	0.36 ± 0.06	1251.03 ± 62.04	2.10 ± 0.09	0.66 ± 0.30	0.08 ± 0.08	0.08 ± 0.00	114.89 ± 13.93
Herencia	53.69 ± 12.02	5830.68 ± 3403.34	0.26 ± 0.26	0.21 ± 0.01	15.53 ± 2.61	0.22 ± 0.01	52.63 ±7.12	0.65 ± 0.01	2460.82 ± 318.58	1.12 ± 0.13	1.31 ± 0.32	0.08 ± 0.11	0.11 ± 0.00	147.43 ± 18.04
Herencia	56.07 ± 12.92	5418.75 ± 271.63	0.24 ± 0.30	0.20 ± 0.01	13.53 ± 2.05	0.20 ± 0.01	35.83 ± 4.42	0.54 ± 0.11	1076.72 ± 36.26	1.00 ± 0.20	1.17 ± 0.06	0.08 ± 0.06	0.08 ± 0.01	51.06 ± 20.16
Herencia	28.77 ± 8.70	3479.34 ± 2336.90	0.25 ± 0.38	0.23 ± 0.01	17.78 ± 2.59	0.21 ± 0.04	46.78 ± 5.37	0.62 ± 0.07	2027.84 ±366.26	1.71 ± 0.93	2.44 ± 0.13	0.11 ± 0.12	0.11 ± 0.01	67.66 ± 15.73
Herencia mean	45.90 ± 10.65 a	11,115.37 ± 6094.19 a	0.26 ± 0.30 ab	0.22 ± 0.01 ab	16.41 ± 1.66 ab	0.21 ± 0.01 a	42.69 ± 6.40 ab	0.60 ± 0.05 a	1495.09 ± 163.19 ab	1.51 ± 0.31 ab	1.71 ± 0.18 ab	0.09 ± 0.09 a	0.09 ± 0.00 a	116.81 ± 15.08 a
Ontur	69.66 ± 2.20	11,014.08 ± 6372.03	0.33 ± 0.32	0.30 ± 0.00	25.08 ± 1.05	0.22 ± 0.01	69.51 ± 6.25	0.76 ± 0.07	1022.35 ± 202.17	2.04 ± 0.22	1.16 ± 0.28	0.07 ± 0.07	0.07 ± 0.01	202.76 ± 14.05
Ontur	48.97 ± 3.13	34,493.68 ± 3828.69	0.31 ± 0.25	0.29 ± 0.01	23.57 ± 0.89	0.22 ± 0.01	61.59 ± 11.28	0.64 ± 0.08	1844.72 ± 955.02	2.00 ± 0.47	2.08 ± 0.18	0.09 ± 0.07	0.09 ± 0.01	181.52 ± 10.39
Ontur	72.06 ± 13.17	8518.27 ± 7209.98	0.30 ± 0.23	0.27 ± 0.00	23.12 ± 2.56	0.23 ± 0.01	60.52 ± 7.16	0.44 ± 0.18	661.44 ± 56.47	1.71 ± 0.23	0.72 ± 0.33	0.07 ± 0.10	0.06 ± 0.00	189.40 ± 16.42
Ontur mean	63.56 ± 6.16 a	18,008.68 ± 17,287.57 a	0.31 ± 0.26 bc	0.29 ± 0.00 c	23.92 ± 1.50 c	0.22 ± 0.01 ab	63.87 ± 8.23 b	0.61 ± 0.11 a	1176.17 ± 404.55 ab	1.92 ± 0.31 abc	1.32 ± 0.26 a	0.08 ± 0.08 a	0.07 ± 0.00 a	191.22 ± 13.62 a
Pozo Cañada	51.02 ± 5.33	48,008.50 ± 31,297.79	0.25 ± 0.30	0.21 ± 0.00	18.38 ± 1.75	0.20 ± 0.01	39.03 ± 4.67	0.35 ± 0.24	623.32 ± 137.51	2.58 ± 0.12	0.57 ± 0.23	0.08 ± 0.07	0.07 ± 0.00	146.89 ± 17.79
Pozo Cañada	74.39 ± 5.25	21,887.28 ± 12,687.34	0.33 ± 0.31	0.28 ± 0.03	25.59 ± 3.19	0.24 ± 0.02	64.12 ± 6.37	0.57 ± 0.06	946.57 ± 455.66	2.42 ± 0.95	1.05 ± 1.08	0.08 ± 0.09	0.122 ± 0.00	206.18 ± 8.64
Pozo Cañada mean	62.71 ± 5.29 a	34,947.89 ± 21,992.57 a	0.29 ± 0.30 abc	0.25 ± 0.01 abc	21.99 ± 2.47 bc	0.22 ± 0.01 ab	51.58 ± 5.52 ab	0.46 ± 0.15 a	784.94 ± 296.58 a	2.50 ± 0.53 bc	0.81 ± 0.66 a	0.08 ± 0.17 a	0.09 ± 0.00 a	202.92 ± 13.21 a
La Puebla de Almoradiel	29.51 ± 6.46	4998.09 ± 2917.48	0.23 ± 0.32	0.24 ± 0.01	14.73 ± 4.89	0.21 ± 0.02	43.98 ± 6.48	0.51 ± 0.25	2209.62 ± 284.84	1.19 ± 1.09	2.83 ± 0.50	0.10 ± 0.07	0.10 ± 0.00	140.19 ± 23.27
La Puebla de Almoradiel mean	29.51 ± 6.46 a	4998.09 ± 2917.48 a	0.23 ± 0.32 a	0.24 ± 0.01 abc	14.73 ± 4.89 a	0.21 ± 0.02 a	43.98 ± 6.48 ab	0.51 ± 0.25 a	2209.62 ± 284.84 b	1.19 ± 1.09 a	2.83 ± 0.50 b	0.10 ± 0.07 a	0.10 ± 0.00 a	140.19 ± 23.27 a
El Sanchón	77.76 ± 1.47	18,842.75 ± 10,985.34	0.30 ± 0.26	0.28 ± 0.00	24.72 ± 2.05	0.21 ± 0.01	61.95 ± 9.24	0.77 ± 0.06	801.33 ± 134.39	2.48 ± 0.20	0.85 ± 0.06	0.08 ± 0.08	0.07 ± 0.00	204.24 ± 3.33
El Sanchón mean	77.76 ± 1.47 a	18,842.75 ± 10,985.34 a	0.30 ± 0.26 abc	0.28 ± 0.00 c	24.72 ± 2.05 c	0.21 ± 0.01 ab	61.95 ± 9.24 b	0.77 ± 0.06 a	801.33 ± 134.39 a	2.48 ± 0.20 bc	0.85 ± 0.06 a	0.08 ± 0.08 a	0.07 ± 0.00 a	204.24 ± 3.33 a
Tarazona	42.26 ± 7.07	9265.78 ± 5538.56	0.27 ± 0.29	0.21 ± 0.02	15.99 ± 2.57	0.20 ± 0.01	35.81 ± 2.83	0.70 ± 0.10	1651.60 ± 395.27	1.13 ± 0.24	1.62 ± 0.10	0.08 ± 0.07	0.08 ± 0.00	136.75 ± 13.00
Tarazona mean	42.26 ± 7.07 a	9265.78 ± 5538.56 a	0.27 ± 0.29 abc	0.21 ± 0.02 a	15.99 ± 2.57 ab	0.20 ± 0.01 a	35.81 ± 2.83 a	0.70 ± 0.10 a	1651.60 ± 395.27 ab	1.13 ± 0.24 a	1.62 ± 0.10 ab	0.08 ± 0.07a	0.08 ± 0.00 a	136.75 ± 13.00 a
F P	2.06	2.36	6.74 ***	8.19 ***	12.11 ***	3.59 **	5.42 ***	1.73	3.10 **	6.88 ***	3.93 **	2.17	2.08	2.60

trans-5-tG: trans-crocetin (β-D-triglucosyl)-(β-D-gentiobiosyl) ester; trans-5-ng: trans-crocetin (β-D-neapolitanosyl)-(β-D-; trans-4-GG: trans-crocetin di-(β-D-gentiobiosyl) ester; trans-4-ng: trans-crocetin (β-D-neapolitanosyl)-(β-D-glucosyl) ester; trans-3-Gg: trans-crocetin (β-D-glucosyl)-(β-D-gentiobiosyl) ester; trans-2-gg: trans-crocetin di-(β-D-glucosyl) ester; cis-4-GG: cis-crocetin di-(β-D-gentiobiosyl) ester; trans-2-G: trans-crocetin (β-D-glucosyl)-(β-D-gentiobiosyl) ester; cis-3-Gg: cis-crocetin (β-D-glucosyl)-(β-D-gentiobiosyl) ester; cis-2-gg: cis-crocetin di-(β-D-glucosyl) ester; cis-2-G: cis-crocetin (β-D-gentiobiosyl) ester); Cropi: sum of crocins and picrocrocin. The mean values (n = 3) are shown with their standard deviation. For each compound, different letters indicate significant differences between samples grouped by meteorological stations according to Tukey test; ** p value < 0.01; *** p value < 0.001) and typed in bold.

,

Table 2. Content (dry weight) of picrococin, HTCC and crocins of Crocus sativus L. flowers from different locations from 2023 season.

Meteorological Station	Picrococin	HTCC	trans-4-GG	trans-3-Gg	trans-2-G	cis-3-Gg	cis-2-gg	cis-2-G	Cropi
	mmol/kg	mmol/kg	mmol/kg	mmol/kg	mmol/kg	mmol/kg	mmol/kg	mmol/kg	mmol/kg
Albacete	77.56 ± 1.05	0.15 ± 0.02	12.51 ± 2.13	0.99 ± 1.41	2.37 ± 0.14	0.84 ± 0.38	0.04 ± 0.0004	0.15 ± 0.05	101.16 ± 13.57
Albacete	88.57 ± 14.01	0.13 ± 0.06	3.97 ± 5.62	0.76 ± 1.08	2.19 ± 1.37	0.97 ± 0.43	0.04 ± 0.003	0.25 ± 0.02	101.56 ± 2.97
Albacete mean	83.06 ± 7.53 a	0.14 ± 0.04 a	7.69 ± 3.87 a	0.88 ± 1.24 a	2.28 ± 0.99 b	0.90 ± 0.40 a	0.04 ± 0.001 a	0.20 ± 0.03 ab	101.36 ± 79.98 a
Herencia	98.39 ± 9.74	0.19 ± 0.06	16.58 ± 5.29	3.422 ± 1.08	2.59 ± 0.71	3.79 ± 1.06	0.10 ± 0.02	0.42 ± 0.1	154.22 ± 26.85
Herencia	78.23 ± 11.24	0.155 ± 0.02	13.41 ± 1.3	3.07 ± 0.35	2.35 ± 1.02	3.60 ± 0.75	0.04 ± 0.004	0.47 ± 0.09	126.80 ± 16.86
Herencia	91.09 ± 23.45	0.16 ± 0.04	14.69 ± 4.05	2.97 ± 1.02	0.71 ± 0.48	3.54 ± 0.84	0.06 ± 0.03	0.41 ± 0.06	139.88 ± 36.61
Herencia mean	89.23 ± 14.81 a	0.17 ± 0.04 a	14.89 ± 3.54 abc	3.15 ± 1.77 bc	1.88 ± 0.73 ab	3.64 ±0.88 cd	0.07 ± 0.018 a	0.44 ± 0.083 bc	140.30 ± 26.77 abc
Juanaco	105.30 ± 22.41	0.19 ± 0.05	19.83 ± 2.18	3.91 ± 0.42	1.74 ± 0.06	3.54 ± 0.36	0.04 ± 0.0003	0.51 ± 0.03	148.82 ± 29.05
Juanaco	108.93 ± 10.59	0.22 ± 0.04	19.11 ± 3.08	3.77 ± 0.62	2.81 ± 0.72	2.70 ± 0.54	0.04 ± 0.001	0.36 ± 0.05	114.41± 4.17
Juanaco	116.78 ± 23.07	0.11 ± 0.06	8.50 ± 5.75	2.08 ± 1.19	1.77 ± 1.14	1.83 ± 0.53	0.04 ± 0.002	0.34 ± 0.06	168.84 ± 17.51
Juanaco	105.61 ± 17.36	0.07 ± 0.01	16.01 ± 2.77	3.28 ± 0.59	2.67 ± 0.52	2.37 ± 0.57	0.05 ± 0.02	0.34 ± 0.05	169.33 ± 19.13
Juanaco	100.75 ± 8.37	0.20 ± 0.02	17.56 ± 1.91	3.99 ± 0.27	1.49 ± 0.42	3.20 ± 0.46	0.04 ± 0.002	0.34 ± 0.15	148.54 ± 4.69
Juanaco	70.97 ± 2.16	0.18 ± 0.02	12.08 ± 1.38	2.74 ± 0.36	0.09 ± 0.01	3.20 ± 0.81	0.04 ± 0.003	0.38 ± 0.15	157.63 ± 9.27
Juanaco mean	101.39 ± 13.93 a	0.16 ± 0.03 a	15.51 ± 2.84 abc	3.29 ± 0.57 bc	1.76 ± 0.47 ab	2.80 ± 0.54 bc	0.04 ± 0.004 a	0.37 ± 0.08	151.26 ± 13.97 abc
Motilleja	106.43 ± 29.45	0.07 ± 0.03	17.09 ± 6.39	3.70 ± 1.40	2.79 ± 1.47	2.21 ± 0.66	0.06 ± 0.04	0.37 ± 0.03	163.57 ± 50.37
Motilleja mean	106.43 ± 29.45 a	0.07 ± 0.03 a	17.09 ± 6.39 bc	3.70 ± 1.40 bc	2.79 ± 1.47 b	2.21 ± 0.66 b	0.06 ± 0.04 a	0.37 ± 0.03 bc	163.57 ± 50.37 bc
Ontur	82.46 ± 22.98	0.12 ± 0.66	28.63 ± 6.22	6.66 ± 1.45	0.09 ± 0.01	4.48 ± 0.78	0.04 ± 0.003	0.04 ±0.01	182.97 ± 44.21
Ontur mean	82.46 ± 22.98 a	0.12 ± 0.66 a	28.63 ± 6.22 d	6.66 ± 1.45 d	0.09 ± 0.01 a	4.48 ± 0.78 d	0.04 ± 0.003 a	0.04 ±0.01 a	182.97 ± 44.21 c
Pedernoso	57.38 ± 10.69	0.09 ± 0.02	6.64 ± 2.08	1.83 ± 0.42	0.87 ± 0.19	2.56 ± 0.39	0.04 ± 0.002	0.54 ± 0.03	85.76 ± 17.27
Pedernoso	86.07 ± 12.29	0.15 ± 0.03	12.14 ± 1.62	3.05 ± 0.47	1.75 ± 0.31	3.21 ± 0.43	0.04 ± 0.002	0.61 ± 0.04	132.96 ± 18.19
Pedernoso	84.50 ± 17.14	0.11 ± 0.02	9.28 ± 1.40	2.33 ± 0.24	1.29 ± 0.20	2.33 ± 0.59	0.04 ± 0.01	0.38 ± 0.04	120.15 ± 19.07
Pedernoso mean	75.98 ± 13.37 a	0.12 ± 0.02 a	9.35 ± 1.7 ab	2.41 ± 0.37 ab	1.31 ± 0.223 ab	2.70 ± 0.47 bc	0.05 ± 0.004 a	0.51 ± 0.03 c	112.95 ± 41.81 ab
Pozo Cañada	79.33 ± 17.70	0.17 ± 0.01	14.78 ± 2.41	3.28 ± 0.38	1.37 ± 0.33	3.16 ± 0.87	0.04 ± 0.002	0.51 ± 0.24	131.48 ± 25.01
Pozo Cañada	79.66 ± 10.80	0.15 ± 0.01	12.43 ± 0.58	3.10 ± 0.23	2.05 ± 1.54	3.07 ± 0.13	0.04 ± 0.01	0.63 ±0.14	155.59 ± 13.24
Pozo Cañada	114.86 ± 31.93	0.18 ± 0.03	19.97 ± 2.80	4.28 ± 0.49	1.29 ± 0.03	2.28 ± 0.71	0.04 ± 0.002	0.45 ± 0.14	182.97 ± 42.30
Pozo Cañada	100.42 ± 14.21	0.48 ± 0.01	8.63 ± 0.25	1.94 ± 0.06	2.25 ± 0.89	2.39 ± 0.20	0.04 ± 0.003	0.39 ± 0.03	119.94 ± 14.87
Pozo Cañada	104.6 ± 20.01	0.17 ± 0.06	15.03 ± 5.63	3.16 ± 1.21	1.92 ± 0.89	3.50 ± 0.74	0.04 ± 0.001	0.37 ± 0.09	124.46 ± 38.58
Pozo Cañada mean	95.78 ± 18.92 a	0.23 ± 0.02 a	14.16 ± 2.33 abc	3.15 ± 0.47 bc	1.78 ± 0.73 ab	2.88 ± 0.52 bc	0.04 ± 0.003 a	0.47 ± 0.10 c	142.88 ± 26.8 abc
El Sanchón	107.60 ± 9.50	0.21 ± 0.02	21.82 ± 2.05	4.47 ± 0.48	0.09 ± 0.01	3.66 ± 0.69	0.06 ± 0.04	0.48 ± 0.14	178.86 ± 17.00
El Sanchón	99.02 ± 3.77	0.21 ± 0.02	20.52 ± 2.13	3.74 ± 0.31	2.55 ± 0.71	3.87 ± 0.17	0.04 ± 0.002	0.37 ± 0.07	161.63 ± 7.54
El Sanchón	89.33 ± 0.49	0.22 ± 0.09	19.05 ± 0.99	4.06 ± 0.07	2.36 ± 0.14	4.12 ± 0.28	0.10 ± 0.02	0.30 ± 0.15	153.93 ± 0.55
El Sanchón mean	98.65 ± 4.58 a	0.21 ± 0.04 a	20.46 ± 1.72 c	4.09 ± 0.29 c	1.67 ± 0.28 ab	3.89 ± 0.38 cd	0.07 ± 0.02 a	0.38 ± 0.12 bc	164.80 ± 8.36 bc
Tarazona	82.40 ± 6.04	0.13 ± 0.06	11.07 ± 1.15	2.26 ± 0.13	1.20 ± 0.27	3.10 ± 0.23	0.04 ± 0.002	0.41 ± 0.14	119.94 ± 8.11
Tarazona	70.21 ± 9.80	0.12 ± 0.01	8.67 ± 2.12	2.14 ± 0.43	0.93 ± 0.39	2.53 ± 0.41	0.04 ± 0.02	0.45 ±0.22	103.56 ± 9.68
Tarazona mean	76.30 ± 7.92 a	0.12 ± 0.035 a	9.87 ± 1.63 ab	2.20 ± 0.28 ab	1.07 ± 0.33 ab	2.81 ± 0.32 bc	0.04 ± 0.01 a	0.43 ± 0.18 bc	111.75 ± 8.89 ab
Villanueva de la Jara	84.71 ± 8.51 a	0.13 ± 0.02	10.96 ± 0.48 ab	2.43 ± 0.06 ab	1.43 ± 0.13 ab	3.72 ± 0.33 cd	0.05 ± 0.002 a	0.40 ± 0.06 bc	124.46 ± 15.15
Villanueva de la Jara mean	84.71 ± 8.51 a	0.13 ± 0.02 a	10.96 ± 0.48 ab	2.43 ± 0.06 ab	1.43 ± 0.13 ab	3.72 ± 0.33 cd	0.05 ± 0.002 a	0.40 ± 0.06 bc	124.46 ± 15.15 ab
F ***	2.31	1.82	10.31 **	13.42 **	2.20	10.74 ***	2.73	5.78 **	5.23 ***

trans-4-GG: trans-crocetin di-(β-D-gentiobiosyl) ester; trans-3-Gg: trans-crocetin (β-D-glucosyl)-(β-D-gentiobiosyl) ester; trans-2-G: trans-crocetin (β-D-glucosyl)-(β-D-gentiobiosyl) ester; cis-3-Gg: cis-crocetin (β-D-glucosyl)-(β-D-gentiobiosyl) ester; cis-2-gg: cis-crocetin di-(β-D-glucosyl) ester; cis-2-G: cis-crocetin (β-D-gentiobiosyl) ester); Cropi: sum of detected crocin + picrocrocin. The mean values (n = 3) are shown with their standard deviation. For each compound. different letters indicate significant differences between samples grouped by meteorological stations according to Tukey test ( ** p value < 0.01; *** p value < 0.001) and typed in bold.

,

Table 3. Content (dry weight) of anthocyanins and flavonols of Crocus sativus L. flowers from different locations from 2022 season.

Meteorological Stations	D-3,5-di-O-g	D-3-O-g	M-3,5-di-O-g	P-3,5-di-O-g	P-3-O-g	K-3 sophoroside-7-glucoside	K-aglycona	K-3-O-sophoroside	Kaeman
	mmol/kg	mmol/kg	mmol/kg	mmol/kg	mmol/mg	mmol/mg	mmol/mg	mmol/kg	mmol/kg
Albacete	12.85 ± 0.49	2.47 ± 0.17	5.51 ± 0.90	0.53 ± 0.07	334.12 ± 294.05	6610.13 ± 129.27	114.33 ± 99.40	59.89 ± 2.10	90.29 ± 2.02
Albacete	14.08 ± 1.26	2.43 ± 0.25	3.96 ± 0.07	0.48 ± 0.06	289.29 ± 250.59	8151.47 ± 559.03	429.13 ± 284.39	63.65 ± 1.55	87.77 ± 3.65
Albacete mean	13.46 ± 0.87 b	2.45 ± 0.17 b	4.73 ± 0.48 a	0.50 ± 0.06 b	311.70 ± 34.35 a	7380.8 ± 344.15 ab	271.73 ± 191.39 a	61.77 ± 1.82 b	89.03 ± 2.83 b
Bolaños	12.61 ± 0.94	2.38 ± 0.13	3.76 ± 0.19	0.46 ± 0.05	443.17 ± 51.16	8085.53 ± 409.02	342.60 ± 297.12	68.56 ± 2.80	94.12 ± 2.24
Bolaños	13.39 ± 0.66	2.11 ± 0.07	3.68 ± 0.20	0.38± 0.01	403.71 ± 10.81	9527.57 ± 94.67	394.55 ± 182.85	61.65 ± 2.74	90.75 ± 2.47
Bolaños mean	13.00 ± 0.80 ab	2.25 ± 0.10 ab	3.72 ± 0.19 a	0.42 ± 0.03 ab	423.44 ± 30.98 a	8806.55 ± 251.84 b	368.57 ± 239.98 a	65.11 ± 2.77 b	92.43 ± 2.35 b
La Gineta	13.92 ± 2.38	2.17 ± 0.09	1.73 ± 3.01	0.38 ± 0.01	131.22 ± 227.27	7524.36 ± 848.23	302.94 ± 153.53	57.78 ± 2.31	83.55 ± 2.76
La Gineta mean	13.92 ± 2.38 ab	2.17 ± 0.09 ab	1.73 ± 3.01 a	0.38 ± 0.01 a	131.22 ± 227.27 a	7524.36 ± 848.23 ab	302.94 ± 153.53 a	57.78 ± 2.31 ab	83.55 ± 2.76 ab
Herencia	12.68 ± 0.13	2.43 ± 0.08	4.77 ± 0.67	0.42 ±0.02	434.16 ± 34.63	6682.66 ± 420.16	360.21 ± 285.12	62.49 ± 0.57	89.58 ± 3.44
Herencia	9.70 ± 0.83	2.02 ± 0.06	4.95 ± 1.28	0.52 ± 0.04	406.84 ± 8.68	5998.11 ± 696.82	348.62 ± 215.25	46.17 ± 3.74	69.16 ± 6.45
Herencia	13.09 ±0.94	2.14 ± 0.03	3.11 ± 0.11	0.41 ± 0.00	402.73 ± 5.90	7150.33 ± 1326.68	299.36 ± 252.45	53.11 ± 2.09	79.00 ± 4.07
Herencia	10.37 ± 0.67	1.96 ± 0.13	0.00 ± 00	0.40 ± 0.00	396.10 ± 6.68	5469.55 ± 367.69	308.10 ± 267.21	49.08 ± 2.26	67.31 ± 3.07
Herencia	10.22 ± 0.80	2.15 ± 0.04	3.76 ± 0.19	0.42 ± 0.02	402.33 ± 13.53	7830.61 ± 526.51	608.86 ± 145.82	65.49 ± 3.14	92.51 ± 3.89
Herencia mean	11.21 ± 0.67 ab	2.14 ± 0.06 ab	3.31 ± 0.45 a	0.43 ± 0.016 ab	408.43 ± 13.88 a	5546.45 ± 667.57 b	385.03 ± 233.17 a	55.27 ± 2.36 ab	79.51 ± 4.81 ab
Ontur	12.85 ± 0.49	2.09 ± 0.14	2.30 ± 0.62	0.44 ± 0.01	402.28 ± 15.44	7972.84 ± 500.77	542.26 ± 71.41	61.42 ± 3.37	87.03 ± 4.27
Ontur	15.63 ± 3.18	2.32 ± 0.33	3.78 ± 0.59	0.39 ± 0.03	391.61 ± 22.97	8521.04 ± 992.77	195.09 ± 175.37	55.37 ± 5.41	94.44 ± 8.89
Ontur	14.12 ± 0.83	2.19 ± 0.12	2.45 ± 0.72	0.42 ± 0.05	395.95 ± 8.28	7106.88 ± 527.19	221.63 ± 91.03	63.75 ± 1.18	81.65 ± 9.61
Ontur mean	14.20 ± 1.5 ab	2.20 ± 0.19 ab	2.84 ± 0.64 a	0.41 ± 0.03 a	396.61 ± 15.56 a	7866.92 ± 673.57 ab	319.66 ± 112.60 a	60.18 ± 3.32 ab	91.04 ± 77.59 ab
Pozo Cañada	13.04 ± 1.18	2.10 ± 0.06	3.66 ± 0.55	0.40 ± 0.01	130.09 ± 225.32	7273.77 ± 325.20	260.60 ± 308.55	55.89 ± 2.60	78.71 ± 8.15
Pozo Cañada	13.78 ± 0.79	2.08 ± 0.13	2.31 ± 1.95	0.35 ± 0.04	128.51 ± 222.58	8115.71 ± 306.16	334.93 ± 117.12	59.05 ± 1.44	85.73 ± 5.93
Pozo Cañada mean	13.4 ± 0.98 ab	2.10 ± 0.09 ab	2.98 ± 1.25 a	0.37 ± 0.02 a	129.30 ± 222.58	7694.74 ± 315.68 ab	297.76 ± 212.83 a	57.47 ± 2.02 ab	82.22 ± 7.04 ab
La Puebla de Almoradiel	12.44 ± 1.27	2.13 ± 0.10	2.59 ± 1.85	0.38 ± 0.03	421.88 ± 54.17	7225.67 ± 321.84	362.89 ± 266.06	60.12 ± 3.60	84.96 ± 6.53
La Puebla de Almoradiel mean	12.44 ± 1.27 ab	2.13 ± 0.10 ab	2.59 ± 1.85 a	0.38 ± 0.03 b	421.88 ± 54.17 a	7225.67 ± 321.84 ab	362.89 ± 266.06 a	60.12 ± 3.60 ab	84.9 ± 6.3 ab
El Sanchón	13.75 ± 0.88	2.13 ± 0.12	3.39± 2.49	0.39 ± 0.00	400.63 ± 2.72	7495.02 ± 692.49	570.41 ± 328.25	59.80 ± 5.41	86.97 ± 9.24
El Sanchón mean	13.75 ± 0.88 ab	2.13 ± 0.12 ab	3.39 ± 2.49 a	0.39 ± 0.00 ab	400.63 ± 2.72 a	7495.02 ± 692.49 ab	570.41 ± 328.25 a	59.80 ± 5.41 ab	86.97 ± 9.24 ab
Tarazona	9.33 ± 0.89	1.93 ± 0.10	3.39 ± 2.49	0.4 ± 0.02	394.75 ± 3.59	6257.37 ± 312.79	290.30 ± 196.97	49.21 ± 4.51	71.26 ± 5.39
Tarazona mean	9.33 ± 0.89 a	1.93 ± 0.10 a	3.39 ± 2.49 a	0.4 ± 0.02 a	394.75 ± 3.59 a	6257.37 ± 312.79 a	290.30 ± 196.97 a	49.21 ± 4.51 a	71.26 ± 5.39 a
Fp	2.16 *	3.16 **	1.16	3.79 **	3.33	3.80 **	0.63	3.15 **	2.98 **

D-3,5-di-O-g: delphinidin-3,5-di-O-β-glucoside; D-3-O-g: delphinidin-3-O-β-glucoside; M-3,5-di-O-g: malvidin-3,5-diO-β-glucoside; P-3,5-di-O-g: petunidin-3,5-di-O-β-glucoside; P-3-O-g: petunidin-3-O-β-glucoside; K-3-O-s-7-O-g: kaempferol-3-O-β-sophoroside-7-O-β-glucoside; K-aglycona: kaempferol aglycone; K-3-β-O-s: kaempferol-3-O-β-sophoroside. Kaeman: sum of kaempferols and anthocyanins. The mean values (n = 3) are shown with their standard deviation. For each compound, different letters indicate significant differences between samples grouped by meteorological stations according to Tukey test (* p value < 0.05; ** p value < 0.01 and typed in bold.

and

Table 4. Content (dry weight) of anthocyanins and flavonols of Crocus sativus L. flowers from different locations from 2023 season.

Meteorological Station	D-3.5-di-O-g	D-3-O-g	K-3 sophoroside-7-glucoside	K-3-O-sopphoroside	Kaeman
	mmol/kg	mmol/kg	mmol/kg	mmol/kg	mmol/kg
Albacete	7.74 ± 0.50	3.14 ± 0.18	3.61 ± 0.01	49.70 ± 1.27	68.34 ± 2.40
Albacete	8.62 ± 0.37	3.51 ± 0.14	4.52 ± 0.55	53.58 ± 0.69	74.33 ± 2.13
Albacete mean	8.18 ± 0.43 a	3.33 ± 0.16 ab	4.08 ± 0.28 a	51.64 ± 0.98 a	71.33 ± 2.27 a
Herencia	9.43 ± 0.50	3.39 ± 0.10	4.65 ± 0.26	64.50 ± 1.77	86.75 ± 3.05
Herencia	10.65 ± 1.30	3.31 ± 0.27	5.00 ± 0.52	70.12 ± 9.29	94.72 ± 12.63
Herencia	11.47 ± 0.70	3.38 ± 0.22	5.35 ± 0.85	66.78 ± 8.43	93.09 ± 10.59
Herencia mean	10.51 ± 0.83 ab	3.36 ± 0.19 ab	5.00 ± 0.54 ab	67.13 ± 6.48 b	91.53 ± 8.75 b
Juanaco	7.94 ± 0.37	2.84 ± 0.12	4.71 ± 0.20	62.28 ± 0.68	81.01 ± 1.31
Juanaco	9.22 ± 0.71	3.44 ± 0.17	4.61 ± 0.27	59.44 ± 5.49	81.33 ± 7.19
Juanaco	9.61 ± 0.52	3.19 ± 0.05	5.05 ± 0.17	63.33 ± 0.42	85.75 ± 0.84
Juanaco	8.46 ± 0.98	3.19 ± 0.24	3.66 ± 0.08	54.77 ± 4.18	74.89 ± 6.45
Juanaco	8.90 ± 0.93	3.07 ± 0.22	5.04 ± 0.20	66.04 ± 3.35	86.91 ± 3.63
Juanaco	10.78 ± 1.85	3.11 ± 0.37	6.02 ± 0.69	75.48 ± 6.24	100.15 ± 10.87
Juanaco mean	9.15 ± 0.89 a	3.14 ± 0.2 a	4.85 ± 0.27 ab	63.56 ± 3.39 ab	84.96 ± 5.05 ab
Motilleja	9.81 ± 0.21 a	3.58 ± 0.04 ab	4.44 ± 0.25 a	63.23 ± 4.18 ab	86.44 ± 2.98 ab
Motilleja mean	9.81 ± 0.21 a	3.58 ± 0.04 ab	4.44 ± 0.25 a	63.23 ± 4.18 ab	86.44 ± 2.98 ab
Ontur	13.30 ± 1.47 d	3.57 ± 0.13 ab	4.62 ± 0.51	89.08 ± 9.96 b	119.25 ± 12.98 c
Ontur mean	13.30 ± 1.47 d	3.57 ± 0.13 ab	4.62 ± 0.51 a	89.08 ± 9.96 c	119.25 ± 12.98 c
Pedernoso	8.44 ± 0.34	2.89 ± 0.17	5.34 ± 0.31	63.04 ± 4.01	82.82 ± 5.15
Pedernoso	9.19 ± 0.52	3.29 ± 0.09	5.62 ± 0.49	70.57 ± 6.79	92.24 ± 8.34
Pedernoso	11.74 ± 1.23	3.80 ± 0.34	5.90 ± 1.18	72.94 ± 7.30	100.22 ± 10.81
Pedernoso mean	9.79 ± 0.69 a	3.33 ± 0.2 ab	5.62 ± 0.66 ab	68.85 ± 6.03 b	68.85 ± 8.1 b
Pozo Cañada	9.05 ± 0.81	3.04 ± 0.42	3.29 ± 2.39	59.26 ± 5.25	80.41 ± 9.46
Pozo Cañada	10.51 ± 1.46	3.20 ± 0.57	4.69 ± 0.49	65.19 ± 5.66	89.41 ± 9.42
Pozo Cañada	8.80 ± 0.40	2.85 ± 0.09	4.35 ± 0.12	64.31 ± 3.16	84.76 ± 3.96
Pozo Cañada	10.83 ± 0.37	3.61 ± 0.19	4.22 ± 0.45	65.34 ± 5.46	90.62 ± 6.64
Pozo Cañada	10.54 ± 0.33	3.45 ± 0.18	3.56 ± 0.18	57.38 ± 4.18	81.91 ± 4.29
Pozo Cañada mean	9.94 ± 0.67 a	3.23 ± 0.29 a	4.02 ± 0.72 a	62.29 ± 4.74 ab	85.42 ± 6.75 ab
El Sanchón	10.69 ± 2.32	3.42 ± 0.50	5.88 ± 1.01	75.68 ± 10.91	100.50 ± 16.94
El Sanchón	9.07 ± 0.41	2.94 ± 0.13	5.30 ± 0.31	67.65 ± 2.83	88.74 ± 3.73
El Sanchón	9.32 ± 1.12	3.07 ±0.37	4.11 ± 1.03	64.82 ± 7.05	86.53 ± 10.60
El Sanchón mean	9.69 ± 1.28 a	3.14 ± 0.33 a	5.10 ± 0.78 ab	69.38 ± 6.93 b	91.92 ± 10.49 b
Tarazona	9.76 ± 0.61	3.59 ± 0.23	5.08 ± 0.18	61.69 ± 3.87	84.81 ± 5.43
Tarazona	8.78 ± 1.90	3.00 ± 0.66	4.68 ± 0.52	62.69 ± 5.81	83.27 ± 10.77
Tarazona mean	9.27 ± 1.25 a	3.29 ± 0.44 a	4.79 ± 0.35 ab	62.19 ± 4.84 ab	84.04 ± 8.1 ab
Villanueva de la Jara	12.70 ± 0.84 bc	4.01 ± 0.30 b	6.50 ± 0.36 b	70.93 ± 5.34 b	100.34 ± 7.07 b
Villanueva de la Jara mean	12.70 ± 0.84 bc	4.01 ± 0.30 b	6.50 ± 0.36 b	70.93 ± 5.34 b	100.34 ± 7.07 b
F *	5.77 ***	2.49 *	4.71 ***	7.84 ***	6.73 ***

D-3,5-di-O-g: delphinidin-3,5-di-O-β-glucoside; D-3-O-g: delphinidin-3-O-β-glucoside; K-3-O-s-7-O-g: kaempferol-3-O-β-sophoroside-7-O-β-glucoside; K-3-O-s: kaempferol-3-O-β-sophoroside. Kaeman: sum of kaempferols and anthocyanins. The mean values (n = 3) are shown with their standard deviation. For each compound, different letters indicate significant differences between samples grouped by meteorological stations according to Tukey test (* p value < 0.05; *** p value < 0.001) and typed in bold.

, where samples has been divided among campaigns studied (2022 and 2023) and compounds related to the saffron stigma quality (picrocrocin, HTCC and crocins) and saffron floral bio-residues (kaempferols and anthocyanin). Also, for each of these divisions, the sampled plots were grouped by the nearest meteorological station, among which ANOVA analysis has been performed.

The levels of picrocrocin and crocins compounds showed notable variations both among meteorological stations and between the campaigns (Table 1 and Table 3). If it compared with the research of Maryam Kabiri et al. [23], it can be observed that the levels of crocin sum together with the picrocrocin obtained, although within the range of that paper, were in general higher.

For picrocrocin, no significant differences were found for either of the two campaigns. In 2022, the highest picrocrocin values were observed in the plots belonging to the meteorological stations of El Sanchón and Bolaños, while in 2023, those grouped under the Motilleja meteorological station stood out with an average of 106.43 mmol/kg, followed by Juanaco and El Sanchón. Although the levels of this compound were slightly higher in the second campaign, in general, values obtained in both campaigns were like those obtained by Moratalla-López et al. [21] when entire saffron flowers were studied with the lower limit of concentration in 2022 being lower than those in that study (26.90 < 46.38 mmol/kg). When comparing the values obtained with those from trials where the spice alone has been analyzed, it is necessary to make the transformation from spice to flower [24].

For HTCC, no significant differences were observed for either of the two campaigns studied (Table 1 and Table 3). However, there were differences between the values obtained from one campaign to another, with those obtained in the second one being higher, as is the case with picrocrocin, which is a precursor of HTCC [25]. Such differences among campaigns could be attributed, among others, to factors such as temperature, since it has been seen that it affects its formation.

Related to the crocins, the trans-4-GG showed statistically significant differences between meteorological stations in both campaigns, 2022 and 2023 (Table 1 and Table 2). According to the one-way analysis of variance (ANOVA), in 2022, the plots grouped at the Albacete, Bolaños, La Gineta, Ontur and El Sanchón meteorological stations were significantly different from stations like La Puebla de Almoradiel, which presented the lowest concentration. In 2023, the pattern was similar for plots from Ontur meteorological station, since presented the highest concentration, being significantly different from stations such as Albacete and Tarazona, indicating the highest accumulation of trans-4-GG in flower samples from plots near to Ontur meteorological station over both years. Comparing both campaigns, there was also a difference in concentration ranges, with the highest range for the 2023 campaign (3.97–70.69 mmol/kg). Despite this, the values obtained for trans-4-GG crocin were in a similar range to those obtained by Moratalla-López et al. [21] (Table 1 and Table 2), with a higher maximum value of 28.63 mmol/kg found in 2023. For trans-3-Gg, comparing the concentrations obtained for both campaigns were significantly different and within each campaign, statistically significant differences were observed (Table 1 and Table 2). In 2022, Ontur, Bolaños, la Gineta and El Sanchón presented the highest concentration, being significantly higher than Tarazona. In 2023, Ontur continued to present a significantly higher value than the rest of the stations, with Albacete showing the lowest values. The results for both campaigns confirm that the plots grouped at the meteorological station of Ontur consistently promotes higher trans-3-Gg accumulation. The maximum values obtained for trans-3-Gg crocin were higher than those obtained in Moratalla-López et al. research [21] (15.96 mmol/kg), with a maximum value of 70.69 mmol/kg observed in 2022 season. The compound trans-2-G presented statistically significant variation in 2022, with the plot grouped in the meteorological stations of Albacete and Bolaños, the ones with the higher concentration, contrary to those in Tarazona and La Puebla de Almoradiel. On the contrary, in 2023, no significant differences were detected, indicating greater homogeneity in the concentration in that year. Also, like the previous ones, the concentrations found for this compound were in a similar range to that found by Moratalla-López et al. [21]. Comparing both campaigns, some of them, such as trans-4-GG, trans-3-Gg, trans-2-G, and their cis forms, showed higher concentrations in 2023 in the plots belonging to the meteorological station of Ontur (Table 3), contrasting with the lower levels observed in those from Tarazona and El Pedernoso. Regarding trans-5-tG, trans-5-ng, and trans-4-ng crocins, compounds listed only in the 2022 season also presented significant differences: trans-5-tG showed a higher concentration in plots grouped in the Albacete meteorological station compared to ones related to Herencia or La Puebla de Almoradiel; trans-5-ng also showed a wide significant difference among plots belonging to Ontur and El Sanchón meteorological stations and those from Herencia and Tarazona. These differences underline location-specific biosynthetic variability.

For the cis forms, the compound cis-3-Gg exhibited significant differences in both years (Table 1 and Table 2). In 2022, the concentrations shown by the different flowers samples were very close. Despite this, the plots that presented the higher significant values were included under the La Puebla de Almoradiel meteorological station, whereas in 2023, the values observed for this compound were higher and the plots include in Ontur meteorological station was clearly differentiated showing significantly higher value than stations like Albacete or Juanaco, suggesting a marked enhancement of cis-3-Gg biosynthesis in flowers from Ontur localization. For cis-2-gg no significant differences were observed for neither campaign. Regarding cis-2-G compound, 2023 campaign revealed significant differences: plots belonging to Ontur meteorological station had significantly lower value than those from stations like Pedernoso and Pozo Cañada. Although, as previously mentioned, no statistically significant differences were observed in 2022, it was observed that the lowest contents for this compound were detected in Ontur, confirming a consistently lower cis-2-G content in flower samples from this localization. Finally, cis-4-GG, reported only in 2022, did not show statistically significant differences, suggesting no major differences across locations for this compound that year. For all these crocins (cis-3-Gg, cis-2-gg, cis-2-G and cis-4-GG) the concentration values found were like those found by Moratalla-López et al. [21].

To develop a term that encompasses the characteristic metabolites of the saffron flower stigma quality, the Cropi parameter is defined, which results from the sum of crocins and picrocrocin. Such parameter highlights, in a global way, the significant individual differences indicated above. In this line, Cropi values showed highly significant differences between meteorological stations in both 2022 and 2023. In 2022, no significant differences were observed. However, values from the plots grouped in meteorological stations such as Ontur and El Sanchón were the highest. The Cropi values for other stations like Herencia and La Puebla de Almoradiel were numerically lower, suggesting that despite statistical overlap, Ontur and El Sanchón consistently perform better in terms of total metabolite content. In 2023, this trend was reinforced, since significant differences were observed and Ontur and El Sanchón showed significantly higher values than Albacete and Tarazona. This grouping pattern highlights Ontur as the most favorable location for the accumulation of saffron metabolites related to stigma (crocins and picrocrocin) followed by El Sanchón, while locations such as Tarazona and Villanueva de la Jara remained in the lower statistical groups with consistently reduced Cropi values across both years. These consistent interannual results suggest that Ontur and El Sanchón possess edaphoclimatic conditions particularly conducive to the biosynthesis and accumulation of crocins and picrocrocin, making them key reference sites for high-quality saffron production and metabolic research.

Among the flavonol and anthocyanin compounds analyzed in Crocus sativus L. flowers from different meteorological stations during the 2022 and 2023 seasons, many of them showed consistent and significant differences across locations according to the one-way analysis of variance (ANOVA) and their content exhibited a more marked interannual variation compared to the previous metabolite group (Table 3 and Table 4). For the anthocyanins, D-3,5-di-O-glucoside presented the clearest pattern in both years, observing statistically significant differences for both campaigns. In 2022, plots grouped in Albacete meteorological station have higher levels than Tarazona ones, while in 2023, Ontur presented the highest accumulation of this compound, significantly above from plots grouped in stations like Albacete and Juanaco (Table 3 and Table 4). The concentrations of D-3,5-di-O-g were considerably higher than that reported by Zhang et al. [26], where the values ranged between 0.045 and 0.073 mmol/kg. On the contrary, these were like those reported by Moratalla et al. [21]. In the case of D-3-O-g relevant differences were also recorded for both campaigns. In 2022, plots groups in Albacete meteorological station showed the highest significant levels, in contrast to Tarazona, which had the lowest concentrations. In 2023, Villanueva de la Jara stood out, with higher significant concentrations than stations such as Pozo Cañada and Juanaco.

Regarding K-3-sophoroside-7-glucoside and K-3-O-sophoroside, significant variation was detected in both campaigns. For K-3-O-sophoroside plots included in the meteorological stations of Bolaños and Albacete presented significant higher values in 2022, whereas in 2023, plots with the highest concentration were enclosed in Ontur meteorological station (Table 3 and Table 4). For K-3-sophoroside-7-glucoside, in 2022 the higher significant values were observed in plots included in Bolaños and Herencia. In 2023, plots grouped in Villanueva de la Jara presented significant higher values, whereas Albacete and Pozo Cañada presented the lower ones (Table 3 and Table 4). For Compared with the literature, the concentrations of K-3-O-sophoroside obtained were slightly lower than those detected in the study by Zhang et al. [26].

Compounds exclusively reported in 2022, such as M-3,5-di-O-glucoside, P-3,5-di-O-glucoside, and K-aglycone, did not exhibit a different concentration distribution depending on the meteorological stations considered.

To develop a term that encompasses the characteristic metabolites of the saffron floral bio-residues, the Kaeman parameter was defined, which results from the sum of kaempferols and anthocyanins. As can be seen in Table 3 and Table 4, Kaeman values varied significantly. In 2022, the higher significant values were observed in plots grouped in the meteorological station of Albacete, whereas the lower significant values were observed in Tarazona. In 2023, plots belonging to Ontur, Villanueva de la Jara, Sanchón y Pedernoso presented the higher values, whereas those include in Albacete, had the lowest ones. These findings suggest a clear influence of geographic location on the metabolic profile of flowers samples. The results partially align with previous studies, such as that of Zhang et al. [26], which also report variability depending on geographic origin and environmental conditions.

In 2022, Kaeman values were generally moderate and more evenly distributed among meteorological stations, suggesting that environmental conditions during those campaigns may not have strongly induced flavonoid biosynthesis. However, in 2023, there was a notable increase in total Kaeman levels in several locations, particularly in Ontur, which recorded the highest value across both years (Table 3 and Table 4). This sharp increase was especially relevant as both kaempferol glycosides and anthocyanins play critical roles in plant stress response and antioxidant defense [27]. Overall, a slight decrease in the absolute concentrations of compounds such as Kaeman was observed in the second campaign.

The elevated levels of kaempferol derivatives observed in 2022 may be associated with increased solar radiation or water deficit stress during key developmental stages, which are known to upregulate the flavonoid biosynthetic pathway [28,29]. Moreover, the presence of anthocyanins, which contribute to the pigmentation of floral tissues and offer photoprotection, also appeared more prominent in samples from stations like Villanueva de la Jara and El Sanchón, aligning with potentially more intense environmental stressors during the 2022 flowering period.

In both campaigns, the results demonstrate that the specific climatic and edaphic conditions of each meteorological station could directly influence the biosynthesis of saffron secondary metabolites, particularly those responsible for its organoleptic (picrocrocin, crocins) and functional (anthocyanins, flavonoids) properties.

With the aim of classified the quality of the saffron flowers based on its bioactive metabolites summarized in the Cropi (sum of crocins and picrocrocin) and Kaeman (sum of kaempferols and anthocyanins) parameters, four quality categories (A, B, C, and D) were defined based on empirical thresholds for these values (Table S1), which reflect both the organoleptic and functional properties of the saffron flowers.

Category A encompasses samples considered to be very high quality, where Cropi values were equal to or greater than 180 mmol/kg and Kaeman values were equal to or greater than 85 mmol/kg. Category B was assigned to the samples of high quality, with Cropi values equal to or greater than 140 mmol/kg and Kaeman values equal to or greater than 75 mmol/kg. Category C included samples of medium quality, with Cropi values equal to or greater than 100 mmol/kg and Kaeman values equal to or greater than 60 mmol/kg. Finally, Category D grouped samples that did not meet the minimum values established for Category C (lower to 100 mmol/kg for Cropi and 60 mmol/kg for Kaeman), being considered as low quality.

This classification provides a quantitative index for each station, facilitating the spatial interpretation of territorial suitability for producing high-quality saffron flowers. In this line and considering the results obtained (Table 1, Table 2, Table 3 and Table 4), the grouped plots of the meteorological stations such as Ontur, El Sanchón, and Bolaños were mostly classified in Categories A and B, confirming their high aptitude for producing saffron flowers rich in bioactive compounds. In contrast, the plots belonging to the meteorological stations like Tarazona and Pedernoso, which had previously shown lower Cropi and Kaeman values, were predominantly classified in Categories C or D.

This information, represented on the surface of Castilla-La Mancha in a more visual way, can be seen in Figure 1, which shows meteorological stations represented by color-coded points, indicating the quality category assigned to each station based on its analytical results. Stations such as Ontur, El Sanchón, and Bolaños predominantly appear in colors corresponding to Categories A and B, which aligns with the textual data suggesting that these areas have a high potential for producing flowers rich in bioactive compounds. The consistency of these classifications across both years indicates a favorable agroecological stability.

In contrast, stations such as Tarazona and Pedernoso are mostly classified within Categories C or D, with colors reflecting low Cropi and Kaeman values. This observation is consistent with the data presented in the text and supports the hypothesis that these locations face environmental or management limitations that negatively impact saffron flower quality.

3.2. Climate Data and Its Relationship with Metabolite Content in Saffron Flowers

Figure 2 shows the most relevant climatic parameters corresponding to the meteorological stations assigned to the samples from the 2022 and 2023 campaigns. The variables presented include average temperature, temperature difference, average of relative humidity, minimum relative humidity and precipitation.

Climatic data from the meteorological stations associated with the saffron flowers sampled in PDO La Mancha Saffron were analyzed for the years 2022 and 2023. Climate data were collected for the first campaign from September 2021 to December 2022 and for the second campaign from October 2022 to December 2023. The months were selected based on those most influential for flower cultivation and quality, according to the literature. To confirm that the differences in climatic conditions among the meteorological stations in the experimental seasons (2022 and 2023) correspond to differences in the climatic conditions of these locations, the average of mean temperature and precipitation for a period of 10 years has been included in Table S2, with the aim of characterizing the area.

The study of these parameters is essential to interpret potential environmental influences on the accumulation of secondary metabolites in Crocus sativus L. For the average temperature, the El Pedernoso meteorological station exhibited the highest values in 2023, especially around the flowering period, whereas Motilleja remained among the coolest stations. In 2022, Bolaños showed high average temperature values during the flowering period. In this same campaign, in Ontur meteorological station the highest temperature differences were recorded, particularly during spring and summer, suggesting greater thermal stress compared to other meteorological stations. The one for Albacete, in contrast, showed the most stable temperature profiles across both years. The average level of relative humidity was higher and more stable in 2022, with Villanueva de la Jara meteorological station having the highest value overall, while Herencia and Ontur registered lower values in 2023. Similarly, the minimum relative humidity was lowest at the Ontur and El Sanchón meteorological stations, particularly during the summer months. Precipitation data indicated that 2022 was generally wetter, with the records of La Gineta and Pozo Cañada showing the highest precipitation levels. In 2023, Tarazona and El Sanchón had the lowest rainfall, potentially contributing to moderate water stress during key phenological stages of saffron.

To know the influence of climatic conditions during the agronomic cycle, a correlation analysis was done between Cropi (sum of crocins and picrocrocin) and Kaeman (sum of kaempferols and anthocyanins) parameters (Table 1, Table 2, Table 3 and Table 4) and among the climate parameters obtained from meteorological stations. Since, the aim is to highlight the influence of the climatic parameters on the accumulation of these compounds, the correlation was made for both seasons (2022 and 2023) (

Table 5. Pearson correlation coefficients (r) between Cropi and Kaeman parameters and climatic conditions from the 2022 and 2023 harvest seasons.

	Tm (°C)		TMA (°C)		Tma (°C)		Dif TMA-Tma (°C)		Hr (%)		Hrmin (%)		P (mm)
	Cropi	Kaeman	Cropi	Kaeman	Cropi	Kaeman	Cropi	Kaeman	Cropi	Kaeman	Cropi	Kaeman	Cropi	Kaeman
October 2021–2022	−0.639 ***	−0.818 ***	−0.561 ***	−0.685 ***	−0.634 ***	−0.802 ***	0.085	0.137	0.337 *	0.418 **	−0.095	−0.020	0.447 **	0.679 ***
November	−0.639 ***	−0.803 ***	−0.669 ***	−0.798 ***	−0.546 ***	−0.722 ***	−0.410 *	−0.378 *	0.242	0.381 *	−0.112	−0.030	0.269	0.254
December	−0.432 ***	−0.628 ***	−0.284	−0.390 **	−0.457 **	−0.681 ***	0.359 *	0.560 ***	−0.160	−0.052	−0.347 *	−0.292	−0.642 ***	−0.830 ***
January	0.162	0.113	0.674 ***	0.804 ***	−0.217	−0.345 *	0.595 ***	0.762 ***	−0.336 *	0.382 **	−0.577 ***	−0.699 ***	−0.281	−0.301 *
February	0.728 ***	0.829 ***	0.702 ***	0.841 ***	0.597 ***	0.605 ***	0.421 ***	0.588 ***	0.059	0.009	−0.351 *	−0.474 **	0.394 **	0.485 ***
March	−0.643 ***	−0.797 ***	−0.716 ***	−0.829 ***	0.536 ***	0.501 ***	−0.722 ***	−0.801 ***	0.747 ***	0.861 ***	0.752 ***	0.858 ***	0.352 *	0.464 **
April	−0.721 ***	−0.852 ***	−0.732 ***	−0.836 ***	−0.478 ***	−0.678 ***	−0.710 ***	−0.758 ***	0.751 ***	0.858 ***	0.757 ***	0.859 ***	0.196	0.180
May	0.641 ***	0.783 ***	0.667 ***	0.809 ***	0.295 *	0.350 *	0.611 ***	0.744 ***	−0.210	−0.358 *	−0.411 **	−0.553 ***	−0.634 ***	−0.775 ***
June	0.707 ***	0.825 ***	0.734 ***	0.861 ***	−0.397 **	−0.547 ***	0.710 ***	0.854 ***	−0.716 ***	−0.846 ***	−0.728 ***	−0.871 ***	−0.685 ***	−0.787 ***
July	0.154	0.119	0.225	0.213	0.018	−0.129	0.221	0.286	−0.169	−0.268	−0.056	−0.099	−0.749 ***	−0.882 ***
August	−0.247	−0.265	−0.172	−0.161	0.006	−0.024	−0.152	−0.125	0.308 *	0.326 *	0.302 *	0.312 *	0.607 ***	0.698 ***
September	0.405 **	0.375 *	0.456 **	0.483 ***	−0.318 *	−0.507 ***	0.479 ***	0.607 ***	−0.676 ***	−0.787 ***	−0.657 ***	−0.735 ***	−0.735 ***	−0.730 ***
October 2022–2023	0.572 ***	0.688 ***	0.627 ***	0.748 ***	0.257	0.227	0.531 ***	0.672 ***	−0.477 ***	−0.528 ***	−0.600 ***	−0.669 ***	−0.404 ***	−0.428 ***

Tm (°C): average of mean temperature; TMA (°C): absolute maximum temperature; Tma (°C): absolute minimum temperature; Hr (%): average of relative humidity; Hrmin (%): absolute minimum relative humidity; P (mm): precipitation. Cropi: sum of crocins and picrocrocin (mmol/kg); Kaeman: sum of kaempherols and anthocyanins (mmol/kg). Significant correlation values are printed in bold according to Fisher’s LSD test (* p value < 0.05; ** p value < 0.01; *** p value < 0.001).

). Therefore, correlation values typed in bold were significant, at least, with a p value < 0.05 (r > 0.51 or r < −0.5), where positive correlations indicated an increase in the analyzed variable, whereas the negatives ones indicated a decrease.

Significant positive correlations between both Cropi and Kaeman and mean temperature (Tm), absolute maximum temperature (TMA), and the difference between maximum and minimum temperatures (Dif TMA-Tma) were observed in February, May, June, and October (2022–2023). This suggests that increased temperatures during these months promote the accumulation of precursors of secondary metabolites in saffron flowers. Flower induction in Crocus sativus L. occurs most effectively within a temperature range of 23 to 27 °C, typically between late spring and mid-summer, following the withering and disappearance of the aerial parts of the plant [10]. Also, in September and October (2022–2023) months, both Cropi and Kaeman exhibited a positive correlation with temperature parameters. During this period, when temperatures reach approximately 17 °C, flowering is initiated [9]. Conversely, negative correlations were found for both parameters in October (2021–2022) and November, confirming that elevated autumn temperatures delay flowering, ultimately leading to lower levels of these metabolites. This aligns with previous studies [30] indicating that low temperatures during the dormant stage inhibit flowering, resulting in reduced Kaeman levels.

Among the monitored meteorological stations, Ontur recorded the average temperature closest to 17 °C in September, October, and November. Notably, this station also exhibited the highest Kaeman content, reflecting increased flowering, which was likely facilitated by mean temperatures of 20.84 °C in September, 17.29 °C in October, and 12.77 °C in November (Figure 2). The temperature decline during this period appears to be a key factor in promoting flowering [10,31].

Related to the difference in maximum and minimum temperature, correlations have been found for Cropi and Kaeman, which is consistent with previous literature on other crops such as grapes, where temperature difference is observed to affect anthocyanin levels [32]. These correlations were highly significant and positive for the months of January, February, June, July, September and October (2022–2023) and negative for the months of March and April.

Regarding relative humidity (Hr and Hrmin), in March and April, a significant positive correlation was observed, indicating that higher humidity favors the presence of these metabolites in saffron flowers. In contrast, in June and September, relative humidity shows a negative correlation with Cropi and Kaeman parameters, which indicates that water stress in these months may enhance the synthesis of certain metabolites [28]. The meteorological station with the lowest relative humidity during June and July of 2022 was Bolaños, which reflected the highest Cropi content compared to the rest of the stations (Table 1 and Table 2). The Kaeman parameter was also higher in the samples corresponding to this station. Relative humidity in October 2022–2023 was lower at the Ontur meteorological station (Figure 2) with a negative correlation with Cropi and Kaeman levels in the saffron flower samples. The samples grouped under this station showed the highest Kaeman content compared to the rest of plots sampled (Table 3 and Table 4). The Pozo Cañada meteorological station in December 2023 was the second station with the lowest relative humidity recorded (Figure 2) and the samples corresponding to this station showed the highest Cropi content (Table 1 and Table 2).

For precipitation, it also showed an interesting relationship with Cropi and Kaeman. In December, May, June, July, August, and September, the correlation was negative and significant, indicating that a lower precipitation may induce an increase in the concentration of these compounds in the flower. The samples corresponding to the Pozo Cañada meteorological station were among those with the lowest precipitation in December 2022. Conversely, in October 2021–2022, the correlation was positive, suggesting that rainfall during this month may be associated with an increase in metabolite synthesis. Comparing both campaigns, the average annual rainfall was higher in the 2021–2022 season compared to the 2022–2023 one (Figure 2). The difference in precipitation between years is an important factor as water stress is known to promote the production of crocin and picrocrocin [29]. This difference also leads to a difference in mean temperatures from one year to the next as a higher precipitation rate acts as a thermal buffer by smoothing temperature extremes. Water stress has been shown to lead to an increase in crocin and picrocrocin in the saffron flower [28].

3.3. Soil Data and Its Relationship with Metabolite Content in Saffron Flowers

The data related to the soil is summarized in Figure 3, where different soil characteristics are shown in maps of Castilla-La Mancha. In all of them, the meteorological stations used to group the studied plots have been indicated. The edaphoclimatic conditions in the Castilla-La Mancha region, particularly in the Bolaños area, present a favorable context for saffron cultivation due to a combination of optimal soil and climatic parameters. The prevailing soil pH across most of the territory is alkaline, typically ranging between 8.0 and 8.5 (Figure 3a). However, in the Bolaños area, the pH ranges from neutral to slightly alkaline (7.0–8.0), a range that appears especially conducive to the biosynthesis of key secondary metabolites in saffron such as crocin, picrocrocin, kaempferols, and anthocyanins [2]. Notably, this region aligns with traditional saffron-growing areas in PDO La Mancha Saffron, underscoring the role of alkaline conditions as a favorable factor for saffron development. In terms of soil texture, the clay content (Figure 3b) plays a significant role by influencing water retention capacity. Soils with a clay content between 20 and 35%, as observed in the Bolaños sampling sites, are associated with higher levels of crocin and kaempferol, suggesting a link between soil composition and the quality of saffron yield. Additionally, electrical conductivity (Figure 3c) levels across all sampling areas remained below 0.5 dS/m, considered in the optimal range for saffron growth, as elevated electrical conductivity values above 2.1 dS/m have been linked to physiological stress indicators such as increased electrolyte leakage, proline accumulation, and malondialdehyde content [33]. Collectively, these edaphic parameters indicate that the soils in the studied region provide a suitable environment for saffron crop, both in terms of chemical balance and physical structure.

In the calcium carbonate content map (Figure 3d), meteorological stations such as Tarazona (11), Pedernoso (12), and La Puebla de Almoradiel (3) are located in areas with very high carbonate levels, which may limit the availability of certain essential micronutrients and hinder balanced nutrition for saffron cultivation [34,35]. In contrast, Bolaños (7), Ontur (1), and El Sanchón (4) appear in areas with intermediate calcium carbonate levels, which are more compatible with optimized plant nutrition and, consequently, higher crop quality. The limestone soil map (Figure 3e) reinforces this distinction: meteorological stations in the eastern and northeastern zones, such as Motilleja (13) and Villanueva de la Jara (10), show a marked presence of calcareous soils, whereas in the central-southern zone, where stations like Bolaños (7) and El Sanchón (4) are located, the calcareous influence is less intense or more heterogeneous. This heterogeneity may favor improved agronomic performance of the crop [36].

Regarding organic matter content (Figure 3f), the map indicates that stations such as Bolaños (7), Pozo Cañada (2), and El Sanchón (4) are situated in areas with higher organic matter accumulation. This is advantageous for saffron cultivation [14], as it contributes to better soil structure, increased moisture retention, and higher biological activity. In contrast, stations such as Tarazona (11), Pedernoso (12), and Herencia (6) exhibit low organic matter levels, which may be associated with soils that are poorer in available nutrients and less resilient to hydric or thermal stress conditions.

4. Conclusions

This study enabled the establishment of a zoning system within the PDO La Mancha Saffron area based on the quality of Crocus sativus L. flowers, using the content of bioactive metabolites present in both stigmas and floral bio-residues. The proposed indices (Cropi and Kaeman) allowed the classification of plots into different quality categories, showing a spatial distribution closely linked to the edaphic and climatic characteristics of the region. The meteorological stations of Ontur, El Sanchón, and Bolaños grouped the highest-quality flower samples, while Tarazona and Pedernoso were mainly associated with lower-quality categories. These results highlight not only the influence of soil properties, particularly intermediate carbonate content, higher organic matter levels, and lower active lime, but also the role of climatic conditions. Factors such as mean temperature, thermal amplitude, and controlled relative humidity during key floral development stages were found to favor the biosynthesis of key metabolites. Therefore, the zoning developed in this work may serve as a valuable tool for agronomic planning, helping to optimize resources and enhance the value of the crop through the promotion of differentiated and sustainable production.