Natural Factors Driving Yield Variability of Camelina sativa L. Crantz and Brassica carinata L. Brown Yield on Sandy-Textured Soils—Case Study from Poland

Abstract

Climate change-induced variability in temperature and precipitation increasingly constrains crop production on sandy-textured soils with low water-holding capacity and limited nutrient retention. Such soils, classified as Brunic Arenosols, are widespread across the temperate climate zone of Central Europe, particularly in post-glacial landscapes, where they constitute a significant proportion of marginal agricultural lands. This study evaluated the relative influence of growing-season weather conditions and selected soil physicochemical properties on the yield of Camelina sativa and Brassica carinata cultivated under low-input management on Brunic Arenosols in northwestern Poland during the 2023 season. Yields varied markedly among sites. Camelina sativa produced yields from 300 to 930 kg ha⁻¹, with the highest yield recorded at the site characterized by higher BS and phosphorus availability. Brassica carinata produced yields from 0 to 370 kg ha⁻¹, including complete yield loss at one location due to severe pathogen infestation. Spearman’s correlation analysis revealed that temperature was a key determinant for both crops (r = 0.77 for C. sativa; r = 0.82 for B. carinata). For Camelina sativa, yield was strongly associated with BS (r = 0.80) and available P (r = 0.69), whereas Brassica carinata was more sensitive to climatic variability, showing a negative relationship with precipitation (r = −0.63). The results indicate species-specific responses to soil fertility and weather conditions under water- and nutrient-limited conditions typical of Central European sandy soils. While Camelina sativa performance was more closely linked to soil chemical status, Brassica carinata appeared predominantly climate-driven. These findings highlight the broader relevance of the study for temperate regions of Central Europe and support the integration of soil fertility management with climate-adaptive strategies when introducing alternative oilseed crops to marginal lands.

1. Introduction

In recent decades, many regions of the world have experienced rising air temperatures, altered precipitation regimes, and an increasing frequency of extreme weather events, which are widely recognized as major consequences of climate change [1]. Across Europe, these shifts are often expressed through greater variability of rainfall distribution during the growing season, combined with higher evaporative demand driven by warming [2]. As a result, drought events—both in terms of frequency and duration—have become more common in several European regions, significantly affecting water availability for agriculture and other sectors [3]. These climate-related changes pose a substantial threat to crop production, as they directly influence plant growth, nutrient uptake, and yield formation, especially in rainfed systems [4,5]. Agricultural vulnerability is particularly high on sandy-textured soils (e.g., Brunic Arenosols), which are characterized by low water-holding capacity, limited organic matter content, and often reduced nutrient retention [6,7,8]. Under such conditions, short-term precipitation deficits can rapidly translate into soil moisture stress, making crop yields highly sensitive to both weather fluctuations and inherent soil properties [9]. Consequently, there is a growing need to identify crop species and management strategies that are more resilient to climate-driven constraints, in order to maintain productivity and reduce the risk of yield losses under future conditions [10,11]. At the same time, agriculture is expected to contribute to climate change mitigation by supporting renewable energy production and reducing dependence on fossil fuels [12,13]. Oilseed crops with potential for biofuel and bioproducts have gained increasing attention in this context [14]. Among them, Camelina sativa L. Crantz, is often described as a low-input crop with a relatively short growing cycle and moderate tolerance to abiotic stress, including drought and nutrient limitations [13,15]. Due to its oil quality and suitability for biofuel, feed, and industrial uses, camelina has been proposed as an alternative crop for sustainable agriculture, particularly in regions where conventional oilseed crops may become less reliable under climate change [15,16]. Another oilseed species of growing interest is Brassica carinata L. Brown is considered a promising industrial oilseed species due to its adaptability, relatively high oil content, and suitability for biodiesel and aviation biofuel production [12,17]. Theses crops are cultivated across a wide range of environments, including Europe, Africa, Australia, Asia, and the Americas, and has been increasingly evaluated for its performance under marginal conditions [12].

Despite the increasing interest in both species, yield performance within marginal lands remains insufficiently characterized, especially in Central European conditions [18]. In Poland, the dominant soil type on marginal lands is rusty soil [9], corresponding to Brunic Arenosols in the international WRB classification [19]. This soil type most often forms on glaciofluvial sandy sediments containing elevated amounts of aluminosilicate minerals [6,7], are characteristic of post-glacial landscapes, widespread in several agricultural regions and are often associated with lower yield potential and greater sensitivity to precipitation shortages [20]. Poland is characterized by considerable seasonal variability in rainfall and temperature, which may interact strongly with soil fertility constraints [21]. Under uniform agronomic management, yield differences between sites can therefore be largely attributed to natural factors such as weather conditions and soil physicochemical properties. Understanding these relationships is crucial for assessing the suitability of Camelina sativa and Brassica carinata as alternative oilseed crops for climate-adaptive agriculture and bioeconomy development in regions of temperate climate.

However, existing studies typically assess either climatic drivers or soil properties in isolation, while their combined effects under low-input conditions remain poorly understood, particularly for emerging oilseed crops such as Camelina sativa and Brassica carinata. Thus, we aimed to evaluate the influence of natural factors—specifically weather conditions (temperature and precipitation) and selected soil physicochemical properties (pH, macronutrient availability, total organic carbon (TOC), and total nitrogen (TN))—on the yield of Camelina sativa and Brassica carinata cultivated on sandy-textured soils, which constitute one of the major components of soil cover in the European temperate climatic zone. The objective of the study was to assess the relative importance of growing-season weather variability and soil fertility status in shaping yield performance across different sites, and to identify whether yield responses differ between the two oilseed crops under the water- and nutrient-limited conditions typical of marginal lands. We hypothesized that (i) in sandy soils with low water retention, yield variability is primarily governed by short-term climatic fluctuations rather than intrinsic soil properties, due to the rapid transmission of precipitation deficits into plant-available water limitations; and (ii) soil fertility acts as a secondary but critical control by regulating the efficiency of nutrient uptake and biomass formation under variable climatic conditions.

2. Materials and Methods

2.1. Study Sites

The study was conducted during the 2023 growing season at four locations in northwestern Poland. The study sites were distributed within Central European Lowlands province across distinct physiographic macroregions (Figure 1), representing typical sandy-textured soils of the temperate climatic zone. The study sites locations were: Site A—Trzebiechów, Warta-Odra Ice Marginal Valley macroregion; Site B—Kozie Laski, Wielkopolskie Lakeland macroregion; Site C—Lubosz, Wielkopolskie Lakeland macroregion; Site D—Płaszewo, Koszalin Coastland macroregion [21]. According to the Köppen–Geiger climate classification, all sites are situated within a fully humid warm temperate climate with warm summers (Cfb) [22]. Despite belonging to the same climatic type, the locations differ in their geographical position and exposure to maritime influences, particularly in the case of the northernmost coastal site (Site D). The geological parent material across the studied regions consists predominantly of glacial tills and fluvioglacial deposits formed during the Pleistocene. In most locations, sandy and gravelly fluvioglacial sediments dominate, providing the substrate for the development of light-textured soils [23], typical for marginal lands.

2.2. Field Trials

At each study site, Camelina sativa and Brassica carinata were cultivated on separate fields with an area of 0.35 ha per crop-large-area experiment. Prior to sowing, organic fertilization (Cattle slurry at a rate of 30 m³ ha⁻¹) was applied in March 2023, approximately two weeks before crop establishment. The slurry was characterized by the following nutrient composition: total nitrogen (0.32%), ammonium nitrogen (0.22%), phosphorus (0.04%), potassium (0.23%), and magnesium (0.02%). Immediately after application, the soil was tilled using a disc harrow to ensure uniform mixing of the slurry with the topsoil layer and to reduce potential nitrogen losses. Sowing of Camelina sativa (variety Olivia) was performed in the first week of April at a density of 320 seeds m⁻², while Brassica carinata (variety Nujet 350) was sown in the first week of May at a density of 60 seeds m⁻², reflecting species-specific thermal requirements and optimal agronomic timing. Plant density was verified in the field based on ten replicate measurements within each plot. Both crops were grown under comparable soil (Brunic Arenosols) and climatic conditions within each site. No chemical plant protection products (herbicides, fungicides, or insecticides) nor mineral fertilization were applied during the growing season, allowing for the assessment of crop performance under low-input management conditions. Harvesting was carried out at full maturity at the turn of July and August 2023. Yield was determined from a single large-area harvest per plot, which is a standard approach for field-scale experiments and reflects practical production conditions.

2.3. Field Survey

In order to ensure a detailed characterization of soil cover variability within each study site, several reconnaissance augerings were performed using a hand auger (Eijkelkamp-type gouge auger). These preliminary borings confirmed the homogeneous sandy character of the soil cover across each study site, indicating limited spatial variability of soil units. On this basis, the most representative locations were selected for soil sampling and profile description. Soil samples were collected from five points within each site at two depth intervals: 0–30 cm and 30–60 cm. Sampling was carried out 1–2 weeks after crop sowing to capture baseline soil conditions at the early stage of plant development. In addition, intact soil cores were collected from each site specifically for bulk density determination. The selected sampling points reflected the dominant soil conditions identified during the reconnaissance survey. Additionally, one soil profile pit was excavated at each site to enable detailed soil classification and morphological assessment. Soil morphology, as well as the identification of diagnostic horizons and materials down to a depth of 100 cm, were described in accordance with the field guide for soil description [19]. Soil classification was performed according to the World Reference Base for Soil Resources 2022 [19]. Meteorological conditions during the growing season, including air temperature and precipitation, were continuously monitored using automatic weather stations located in close proximity to the experimental fields. These data were used to characterize the climatic background of the study period.

2.4. Laboratory Analysis

Prior to laboratory analyses, all soil samples, except for those collected for bulk density determination, were air-dried at room temperature. The dried samples were gently crushed and homogenized in a mortar and subsequently sieved through a 2 mm mesh to obtain the fine earth fraction (≤2 mm) for further analyses. Soil texture was determined using the sieve–hydrometer method. Bulk density (BD) was measured using intact soil cores (100 cm³) and expressed on a dry mass basis. Soil reaction (pH) was measured potentiometrically in both distilled water (pH_H2O) and 1 M KCl solution (pH_KCl) at a soil-to-solution ratio of 1:5. Total organic carbon (TOC) and total nitrogen (TN) contents were determined using a CNS VarioMax elemental analyzer (Elementar Analysensysteme GmbH, Langenselbold, Germany). As the analyzed samples were free of calcium carbonate, the measured total carbon (TC) was assumed to represent TOC. Plant-available forms of phosphorus (P), potassium (K), and magnesium (Mg) were extracted using the Mehlich 3 method and quantified by inductively coupled plasma optical emission spectrometry (ICP-OES). Exchangeable base cations (Ca²⁺, Mg²⁺, K⁺, and Na⁺) were extracted with 1 M ammonium acetate at pH 7.0 and then quantified using atomic absorption spectrophotometry (AAS). Hydrolytic acidity (HA) was determined after extraction with 1 M sodium acetate, followed by potentiometric titration. Based on exchangeable base cations and hydrolytic acidity the total exchangeable baes (TEB), cation exchange capacity (CEC) and base saturation (BS) have been calculated. All analyses were performed according to commonly accepted soil science laboratory procedures [24].

2.5. Data Processes

Basic descriptive statistics, including the mean and standard deviation, were calculated for all measured soil variables to summarize central tendency and variability. Data were initially tested for normality using the Shapiro–Wilk test. As most variables did not meet the assumptions of normal distribution, non-parametric analyses were applied where appropriate. Differences in soil properties among study sites were assessed using the Kruskal–Wallis test, followed by Dunn’s post hoc test. To identify the key soil and environmental variables associated with Camelina sativa and Brassica carinata yield, Spearman’s rank correlation analysis was performed. The analysis included the following parameters: crop yield, precipitation, temperature, BD, pH(H₂O), TOC, TN, P, K, Mg, HA, and BS. The soil parameters represent mean values calculated for two depths: 0–30 cm and 30–60 cm. This approach allowed identification of the main factors explaining variability in soil and climate characteristics and their potential influence on crop performance. All statistical analyses were performed using statistical packages Statistica 13 (StatSoft Inc., Tulsa, OK, USA), and PAST 4.15 [25].

3. Results

3.1. Soil and Weather Conditions Characterization

The study soils belongs to the reference group of Arenosols (Figure 2), with well-developed subsoil horizon of brownish and reddish colour, and coarse texture, thus fulfilling the criteria for Brunic qualifier. The full soil names in accordance with World Reference Base for Soil Resources [18] were as follows: Site A—Dystric Brunic Arenosol (Aric, Ochric); Site B—Eutric Gleyic Brunic Arenosol (Aric, Ochric); and Sites C and D—Eutric Brunic Arenosols (Aric, Ochric). The Eutric qualifier indicates higher base saturation and nutrient availability, while the Dystric qualifier corresponds to lower base saturation. The Gleyic qualifier at Site B indicates periodically saturated conditions in the subsoil. The Ochric qualifier reflects a weakly developed humus horizon, thinner or lower in organic matter content than typical, whereas the Aric qualifier indicates that the humus horizon has structural elements created by ploughing. All soils were developed on very thick sandy deposits of glacial and fluvioglacial origin. The humus horizon was approximately 30 cm thick at Sites A and B, and around 20 cm at Sites C and D, consistent with the weakly developed Ochric horizons. The sandy texture combined with low organic matter and low water-holding capacity makes these soils representative of marginal lands, characterized by limited natural fertility and susceptibility to water and nutrient stress.

During the 2023 growing season, all study sites experienced a gradual increase in air temperature from March to July, reaching maximum values of approximately 20 °C in mid-summer (Figure 2). Mean precipitation showed notable temporal variation throughout the season. Early in the season (March–May), rainfall was generally low, with May being the driest month, which may have limited water availability during early crop growth. From June onwards, precipitation increased steadily, reaching the highest values in August at all sites. Sites B and C received the greatest rainfall (>200 mm) during this late season period (Figure 3). Overall, temperature patterns were highly similar among the sites, indicating uniform thermal conditions. In contrast, the variation in precipitation, especially in the mid-to-late season, represents the main climatic factor that could influence differences in crop growth and yield between the study sites.

3.2. Soil Physical Properties

The study soils were dominated by a sandy texture, consistent with their classification as Brunic Arenosols. Particle size analysis confirmed that the sand fraction ranged from 89% to 94% across all sites and depths, with silt and clay content generally below 8% and 4%, respectively (

Table 1. Soil texture (particle size distribution) and bulk density of the studied soils.

Site	Depth (cm)	Particle Size Distribution			Class USDA	BD (g cm−3)
		Sand (%)	Silt (%)	Clay (%)
A	0–30	90	7	3	sand	1.52 ± 0.03 a
	30–60	90	6	4	sand	1.72 ± 0.04 a
B	0–30	89	8	3	sand	1.48 ± 0.03 a
	30–60	91	6	3	sand	1.54 ± 0.02 b
C	0–30	90	7	3	sand	1.71 ± 0.02 b
	30–60	94	5	1	sand	1.55 ± 0.04 b
D	0–30	90	6	4	sand	1.55 ± 0.04 ab
	30–60	94	4	2	sand	1.48 ± 0.05 b

Explanation: BD—bulk density (mean ± standard deviation); different letters indicate significant differences among study sites according to Dunn’s post hoc test (p ≤ 0.05).

). According to the USDA soil texture classification, all studied soil layers were classified as sand.

Bulk density (BD) values ranged from 1.48 to 1.55 g cm⁻³, with the topsoil layers (0–30 cm) generally showing slightly lower BD than the subsoil layers (30–60 cm), reflecting the influence of organic matter and biological activity in the surface horizon. Statistical analysis indicated significant differences in BD among sites within individual depth intervals. In the 0–30 cm layer, BD at site D was significantly different from the remaining sites (A, B, and C) (p ≤ 0.05). In contrast, in the 30–60 cm layer, site A differed significantly from sites B, C, and D, whereas no significant differences were observed among the other sites within this depth. Field assessment of soil structure revealed that the humus horizons have weakly developed, small to medium subangular structure.

3.3. Soil Chemical Properties

The analyzed soils exhibited significant differences in chemical properties among the four sites (A–D) and the two depth intervals (0–30 cm and 30–60 cm). Dunn’s post hoc test confirmed that surface horizons (0–30 cm) contained significantly higher levels of TOC and TN compared to subsoil layers (

Table 2. Selected chemical properties of the studied soils (mean ± standard deviation).

Site	Depth (cm)	pH		TOC	TN	P	K	Mg
		H2O	1M KCl	%		mg kg−1
A	0–30	5.2 ± 0.1 a	4.4 ± 0.2 a	0.57 ± 0.05 a	0.08 ± 0.01 b	80.5 ± 2.0 b	73.0 ± 2.0 b	11.7 ± 1.0 a
	30–60	5.0 ± 0.2 a	4.2 ± 0.2 a	0.18 ± 0.03 a	0.05 ± 0.01 b	75.2 ± 0.2 b	71.3 ± 2.3 a	12.6 ± 0.6 a
B	0–30	6.2 ± 0.1 bc	5.6 ± 0.1 ab	0.58 ± 0.04 a	0.05 ± 0.01 a	86.9 ± 3.3 a	74.6 ± 2.1 a	10.5 ± 0.9 a
	30–60	6.3 ± 0.3 a	5.6 ± 0.2 ab	0.09 ± 0.02 b	0.01 ± 0.00 a	81.6 ± 4.4 a	77.0 ± 1.7 a	10.2 ± 0.8 b
C	0–30	6.1 ± 0.1 ab	5.3 ± 0.1 ab	0.49 ± 0.04 b	0.04 ± 0.00 ab	87.4 ± 3.6 a	77.0 ± 1.7 a	14.0 ± 1.1 b
	30–60	7.1 ± 0.2 b	6.8 ± 0.2 c	0.03 ± 0.02 b	0.01 ± 0.00 ab	82.3 ± 1.3 a	61.9 ± 1.0 b	10.8 ± 0.7 a
D	0–30	7.2 ± 0.2 c	6.5 ± 0.2 b	0.60 ± 0.02 a	0.05 ± 0.00 a	84.8 ± 2.8 a	76.0 ± 3.4 a	11.7 ± 0.8 a
	30–60	6.9 ± 0.2 ab	6.2 ± 0.2 bc	0.19 ± 0.03 a	0.01 ± 0.00 a	81.5 ± 1.2 a	75.4 ± 3.7 a	12.3 ± 1.0 a

Explanation: TOC—total organic carbon; TN—total nitrogen; P—phosphorus; K—potassium; Mg—magnesium; different letters indicate significant differences among study sites according to Dunn’s post hoc test (p ≤ 0.05).

). The highest surface TOC and TN values were observed in site D, which differed significantly from profiles A and B, where these parameters were lowest.

Soil pH also varied significantly among the study sites. Soil in site A was significantly more acidic than soils in sites B, C, and D, which exhibited a neutral reaction. No significant changes in pH with depth were detected in soil B and C, whereas soil A showed a slight but significant decrease in the subsoil layer. Soil in the site D displayed the highest pH values, differing significantly from site A.

The distribution of plant-available macronutrients revealed clear and statistically significant differences among the sites. Surface soil horizons contained significantly greater amounts of P, K, and Mg than subsoil layers, indicating nutrient accumulation in the topsoil. All profiles exhibited a significant decline in plant-available P with depth, while K and Mg showed more stable distributions, with only slight but significant reductions in subsoil layers. A particularly pronounced and statistically significant decrease in K was observed in the deeper layer of profile C, suggesting enhanced leaching or lower nutrient retention capacity.

Dunn’s test confirmed a distinct gradient of fertility and organic matter content among the studied soils. Profiles B, C, and D exhibited significantly higher fertility and more favourable nutrient status compared to profile A, which had the lowest values of TOC, TN, and plant-available macronutrients (Table 2).

The studied soils exhibited distinct patterns in exchangeable cations, HA, CEC, and BS across the four sites and soil depths. Calcium (Ca²⁺) was the dominant exchangeable cation in all profiles, followed by magnesium (Mg²⁺) and potassium (K⁺), while sodium (Na⁺) contributed minimally. Soil within the sites B and C generally showed higher TEB in the surface horizons compared to sites A and D, reflecting a greater capacity to retain nutrient cations in the topsoil (

Table 3. Sorptive properties of the studied soils.

Site	Depth (cm)	Ca2+	Mg2+	K+	Na+	HA	TEB	CEC	BS
		cmol(+) kg−1							%
A	0–30	2.40 ± 0.19 ab	0.11 ± 0.01 a	0.03 ± 0.00 a	0.02 ± 0.00 a	3.51 ± 0.13 a	2.57 ± 0.19 ab	6.08 ± 0.27 a	42.2 ± 1.7 a
	30–60	2.04 ± 0.16 ab	0.15 ± 0.03 ab	0.03 ± 0.01 a	0.02 ± 0.00 a	3.42 ± 0.08 a	2.24 ± 0.17 ab	5.66 ± 0.25 a	39.9 ± 1.4 a
B	0–30	2.49 ± 0.37 a	0.17 ± 0.05 ab	0.02 ± 0.01 a	0.03 ± 0.01 a	1.21 ± 0.07 b	2.71 ± 0.39 a	3.91 ± 0.39 ab	68.9 ± 3.4 b
	30–60	2.57 ± 0.53 a	0.11 ± 0.01 a	0.02 ± 0.01 a	0.03 ± 0.00 a	1.17 ± 0.06 b	2.72 ± 0.52 a	3.90 ± 0.50 ab	69.4 ± 4.8 b
C	0–30	2.37 ± 0.29 ab	0.28 ± 0.05 b	0.02 ± 0.01 a	0.03 ± 0.00 a	1.86 ± 0.08 ab	2.70 ± 0.28 a	4.56 ± 0.29 ab	59.0 ± 2.8 ab
	30–60	1.79 ± 0.21 ab	0.20 ± 0.02 b	0.03 ± 0.01 a	0.03 ± 0.00 a	1.87 ± 0.05 ab	2.05 ± 0.17 ab	3.92 ± 0.17 ab	52.3 ± 2.4 ab
D	0–30	1.64 ± 0.25 b	0.14 ± 0.02 ab	0.02 ± 0.01 a	0.02 ± 0.00 a	1.08 ± 0.05 b	1.83 ± 0.25 b	2.91 ± 0.20 b	62.6 ± 4.5 ab
	30–60	1.65 ± 0.25 b	0.15 ± 0.02 ab	0.02 ± 0.01 a	0.03 ± 0.00 a	1.18 ± 0.06 b	1.84 ± 0.23 b	3.02 ± 0.23 b	60.8 ± 3.5 b

Explanation: HA—hydrolytic acidity; TEB—total exchangeable bases; CEC—cation exchange capacity; BS—base saturation; different letters indicate significant differences among study sites according to Dunn’s post hoc test (p ≤ 0.05).

).

Hydrolytic acidity decreased in the order of sites A > C > B > D, with surface soil horizons typically exhibiting slightly higher acidity than the subsoil in the more acidic soils. Consequently, BS was lowest in soil A, indicating a more acidic, less fertile soil, and highest in soil B, reflecting a more neutral environment with greater nutrient availability. Soil C displayed a moderate base saturation, whereas soil D showed relatively stable values with depth, suggesting more uniform chemical conditions (Table 3).

Cation exchange capacity (CEC) mirrored these trends: soils with higher HA generally had higher CEC, while TEB contributed to overall BS. Depth-related patterns were also evident; in most profiles, exchangeable bases and BS decreased slightly with depth, consistent with the downward decline in organic matter and nutrient retention. Notably, soil from site C exhibited a marked decrease in TEB and BS in the subsoil, highlighting potential leaching or lower fertility in deeper layers (Table 3).

3.4. Relationships Between Crop Yields, Soil Properties, and Weather Conditions

Yields of both crops varied among sites. The highest yield of Camelina sativa was recorded at site B (930 kg ha⁻¹), whereas the lowest yield occurred at site A (300 kg ha⁻¹). For Brassica carinata, the greatest yield was also observed at site B (370 kg ha⁻¹), while sites C and D produced only 150 kg ha⁻¹ and 0 kg ha⁻¹, respectively. The complete yield loss at site D resulted from severe pathogen infestation, which prevented biomass production (

Table 4. Crop yield of Camelina sativa and Brassica carinata across study sites.

Site	C. sativa	B. carinata
	kg ha−1
A	300	150
B	930	370
C	720	150
D	420	0

).

Spearman’s rank correlation analysis revealed distinct patterns in the relationships between crop yields, soil properties, and meteorological conditions for Camelina sativa and Brassica carinata. For Camelina sativa, yield was strongly and positively correlated with base saturation (BS; r = 0.80), air temperature (T; r = 0.77), and available phosphorus (P; r = 0.69). In contrast, strong negative correlations were observed with total organic carbon (TOC; r = −0.77) and total nitrogen (TN; r = −0.60). Moderate relationships were also noted for bulk density (r = −0.37), hydrolytic acidity (r = −0.46), Mg (r = −0.34), and soil pH (r = 0.38). Precipitation showed only a weak positive association (r = 0.20) (

Table 5. Spearman’s rank correlation coefficients (n = 20, p < 0,05) describing relationships between yields of Camelina sativa and Brassica carinata, soil properties, and weather conditions.

	BD	pH	TOC	TN	P	K	Mg	HA	BS	Pr	T
C. sativa	−0.368	0.385	−0.769 *	−0.598 *	0.695 *	−0.155	−0.341	−0.459 *	0.796 *	0.200	0.775 *
B. carinata	−0.062	−0.314	−0.295	−0.025	0.178	−0.601 *	−0.478 *	0.197	0.337	−0.632 *	0.817 *

Explanation: BD—bulk density; TOC—total organic carbon; TN—total nitrogen; P—phosphorus; K—potassium; Mg—magnesium; HA—hydrolytic acidity; BS—base saturation; Pr—precipitation; T—temperature; *—statistically significant at p < 0.05.

). In the case of Brassica carinata, yield exhibited a very strong positive correlation with temperature (r = 0.82), highlighting the dominant role of thermal conditions in determining productivity. Conversely, significant negative correlations were found with precipitation (r = −0.63) and K (r = −0.60). Moderate negative relationships were also observed with Mg (r = −0.48) and soil pH (r = −0.31), while correlations with TOC (r = −0.29), TN (r = −0.02), and bulk density (r = −0.06) were weak. A weak positive association was noted with hydrolytic acidity (r = 0.20) and BS (r = 0.34).

Compared to Camelina sativa, the yield of Brassca carinata appeared to be more strongly controlled by meteorological factors, particularly temperature and precipitation, while soil chemical properties played a secondary role. The negative relationship with precipitation suggests a potential sensitivity to excessive moisture conditions, which may limit growth or nutrient uptake. Overall, temperature emerged as a key determinant of yield for both species. However, Camelina sativa demonstrated stronger associations with soil fertility parameters, particularly base saturation and phosphorus availability, whereas Brassica carinata showed greater responsiveness to climatic variability. These findings highlight species-specific differences in environmental controls of yield formation and suggest distinct adaptive strategies under varying soil and weather conditions.

4. Discussion

The results of this study demonstrate that yield variability of Camelina sativa and Brassica carinata cultivated on sandy soils is governed by the combined effects of climatic conditions and soil fertility, with clear evidence of species-specific responses. The high sand content, weakly developed structure, and moderate-to-high bulk density indicate low water-holding capacity and restricted nutrient retention, features that are characteristic of marginal soils in the European temperate zone [7,8,20]. This supports the general assumption that, in coarse-textured soils with low water-holding capacity, short-term weather fluctuations can strongly influence crop performance. These physical constraints likely played a central role in shaping crop performance, as both Camelina sativa and Brassica carinata must establish and produce biomass under conditions of limited water and nutrient availability [16,17]. The low water-holding capacity of these soils provided the basis for the hypothesis that yield variability of Camelina sativa and Brassica carinata across the studied sites would be primarily driven by weather conditions during the growing season. In sandy soils, irregular rainfall patterns are expected to translate rapidly into water deficits due to limited retention capacity, thereby directly influencing plant growth. Moreover, temperature-driven increases in evapotranspiration may rapidly exacerbate water deficits, thereby amplifying the sensitivity of crops to climatic variability [5]. The results therefore support the general assumption that soil physical properties interact with climatic variability to influence yield under low-input conditions [20,26].

However, the findings indicate that yield variability was not governed solely by weather conditions. Soil fertility and climatic factors exhibited distinct relationships with crop yield, highlighting species-specific responses to environmental conditions. Camelina sativa yield appeared to benefit from warmer growing-season temperatures and soils with higher base saturation and phosphorus availability. Such relationships, reported also in previous studies [27,28], suggest that thermal conditions and nutrient status were important drivers of productivity, even on sandy soils with inherently low nutrient retention. The association with base saturation may indicate that more neutral chemical conditions facilitated nutrient uptake, whereas phosphorus availability likely contributed to improved metabolic activity and biomass formation. These findings align with the concept that marginal soils, although limited in fertility, can support crop production when essential nutrients are available and climatic conditions are favourable [29].

The negative relationships between yield and TOC and TN require cautious interpretation. Rather than implying a detrimental effect of organic matter, these correlations may reflect the complex dynamics of organic matter transformation in sandy soils. Higher yields could be associated with more advanced organic matter mineralization, resulting in greater availability of nutrients in mineral forms rather than in organically bound pools [30,31]. Such a mechanism is plausible in low-input systems where nutrient cycling depends largely on soil organic matter decomposition. Consequently, the observed patterns may indicate that nutrient availability from mineralized organic matter, rather than total organic content per se, influenced crop performance [31]. The highest TOC and TN concentrations were observed in the uppermost soil horizons, which is a common phenomenon resulting from the accumulation of plant residues at the soil surface, leading to increased contents of these parameters [32].

Overall, the interactions observed in this study provide only partial support for the hypothesis. Although precipitation distribution was expected to be the primary determinant of yield variability due to the low water-holding capacity of sandy soils (determined by weak aggregation), temperature and nutrient availability appeared to exert stronger and more consistent influences. This suggests that, under sandy soil conditions, thermal regimes and soil chemical status may modulate or even override the direct effects of precipitation variability [31]. This may be explained by the fact that total precipitation does not directly translate into plant-available water in sandy soils, where rapid infiltration and low retention promote losses beyond the root zone [33]. The weak aggregate structure and limited water retention capacity likely reduced the buffering potential of the soils [7,20], but at the same time may have amplified the importance of nutrient availability and temperature-driven mineralization processes. Thus, yield variability resulted from the combined action of climatic and edaphic factors rather than precipitation alone.

Similarly, the influence of soil fertility on yield differed between species, reflecting contrasting adaptive strategies under water- and nutrient-limited conditions. Camelina sativa demonstrated greater sensitivity to base saturation and phosphorus availability, both of which affect nutrient uptake efficiency and plant responses to environmental stress. Phosphorus is essential for energy metabolism and physiological regulation, processes that are critical for maintaining plant productivity under variable environmental conditions [34]. In contrast, Brassica carinata appeared more responsive to climatic factors, particularly temperature. Such differences underline the need to consider species-specific requirements when evaluating crop suitability for marginal lands.

From a broader perspective, these findings contribute to the understanding of crop performance in sandy soils typical of temperate regions, where low water retention and limited nutrient availability constrain agricultural productivity [20,26,28]. The ability of Camelina sativa and Brassica carinata to produce biomass under these conditions demonstrates their potential as alternative crops for marginal lands. Similar observations have been reported by other authors, who have highlighted the adaptability of these species to marginal environments characterized by limited nutrient availability and water constraints, emphasizing their suitability for sustainable biomass production under low-input management systems [12,16,17,26,28]. However, the interactions between soil properties, climatic factors, and management practices remain critical determinants of yield. Strategies aimed at enhancing soil organic matter quality, improving nutrient availability, and optimizing water retention may therefore contribute to sustainable productivity on marginal sandy soils [35]. The findings of this study highlight both the opportunities and constraints associated with crop production in such environments, supporting the view that marginal lands can play a role in diversified agricultural systems when appropriate management practices are applied.

5. Conclusions

The study confirmed that yield of both oilseed crops was primarily shaped by interactions between soil fertility and weather conditions on marginal sandy soils. Warmer growing-season temperatures and higher base saturation, together with greater phosphorus availability, were key factors supporting productivity, particularly for Camelina sativa. In contrast, Brassica carinata showed stronger sensitivity to climatic variability, indicating species-specific responses to environmental constraints. Soil physical limitations typical of marginal lands—low water-holding capacity and limited nutrient retention—likely moderated yield potential under low-input conditions. The observed relationships between yield and organic matter parameters suggest that nutrient availability from mineralised organic pools may be more important for short-term productivity than total organic carbon alone. Overall, the results highlight that both soil and weather factors contribute to yield variability, but their influence differs between species. These findings support the potential of oilseed crops on marginal lands, while emphasizing the need for targeted management strategies to improve soil fertility and resource efficiency. Future research should therefore include multi-year experiments and a larger number of locations, as well as consider integrating process-based approaches to better capture the dynamics of soil–plant–climate interactions.