Feed Values for Grassland Species and Method for Assessing the Quantitative and Qualitative Characteristics of Grasslands

Abstract

The tasks and objectives of grassland management have changed significantly in recent decades. One of the key elements of adapting to climatic and economic challenges is the optimal use and future sustainability of grasslands. Ferenc Balázs’s plant stand assessment method is a fast, efficient and widely applicable method for evaluating the quantitative and qualitative characteristics of forage in grasslands, as well as the economic value of pastures. This study is based on a three-dimensional coenological survey which is low-cost, does not require technical infrastructure, and empirically considers the species’ preference by livestock. As a result of our extended criteria approach, we assigned modified forage value (k-value) categories to 2310 vascular plant species. Based on our investigations in the presented case study, the Balázs method was proven to be well suited for estimating the yield of grasslands and determining the relative forage value of grasslands with a high degree of confidence in practice. As this method is non-destructive and involves little trampling, it is particularly suitable for monitoring grassland habitats with a high density of protected plant and animal species.

1. Introduction

Maintaining grasslands is crucial for nature conservation, effective grassland management, and the provision of ecosystem services [1,2,3]. It is therefore essential to determine how many animals can be supported on our grasslands in the most sustainable way possible [4]. To this end, the biomass and yield of grasslands must be determined and estimated as accurately as possible.

The amount and quality of biomass produced and its species composition are closely related to grassland management practices too [5,6]. Numerous studies have found that different grazing intensities and their frequency can influence the quantity, composition and quality of herbaceous biomass available to livestock [7,8,9]. Additionally, the quality of forage is closely related to species composition [10,11]. Diverse grassland plant communities often provide better quality forage due to the diversity of nutrient profiles offered by different species [12]. Mowing influences the vegetative structure and growth cycle of plant species, thus also affecting the palatability and nutritional value of forage. Early mowing tends to reduce the flowering of some key forage species, affecting the availability of good-quality forage later in the growing season [13,14].

According to recent research, environmental stress factors, particularly drought and climate change, are garnering increasing attention due to their impact on biomass production. For example, it has been shown that an increase in the frequency and intensity of drought periods significantly reduces biomass production in Festuca arundinacea-dominated grasslands [15,16,17]. Studies have shown that different management practices can significantly influence biomass yield and nutritional value. For example, fertilisation of grasslands is associated with improved biomass production, which highlights the importance of nutrient management in improving feed quality [18,19].

In addition to the quantity of biomass, its quality is also essential. Therefore, successful management requires maximising biomass while harvesting at the optimum point of quality indicators and, among other things, applying rotational grazing strategies [20,21]. Research has also shown that the feed value of Festuca arundinacea varies depending on maturity and growth stage, particularly in terms of crude protein and digestibility. Early harvesting generally results in a higher protein content, while late harvesting can increase fibre content, which reduces digestibility [22,23]. The use of specific cultivation techniques to enhance digestibility and disease resistance further amplifies the potential of this grass species [24]. Laboratory testing of the feed value of grass involves various mechanical and chemical analyses. The most commonly used techniques include the determination of crude protein, fibre and mineral content. However, hyperspectral techniques are becoming more widespread, which can be used to identify different plant species and, with promising results, to determine specific feed quality indicators such as cellulose and lignin content [25]. These analyses provide important information on feed value and help to improve feed quality and optimise animal husbandry practices [25]. In addition, air pollutants, especially ozone, have been documented to adversely affect the quality of grass feed value, potentially reducing it [26].

Various methods are used to estimate the quantity and quality of pasture biomass. These methods for measuring and assessing the forage value of grasslands cover a broad spectrum, from laboratory analysis to remote sensing techniques, which fundamentally shape the future of forage production and evaluation. Several methods are available to assess biomass by mowing [27], but this is often not feasible in protected areas or when assessing protected species. In detail, for a given sampling area (typically 1 m²), it is necessary to obtain multiple replicate biomass samples to obtain a more accurate estimate of the grassland yield, which can be used to estimate it per hectare and even to collect information on the biomass quality of the area by sorting the cut biomass by grassland components. However, their drawbacks (different stubble heights, high manual labour and equipment requirements, small sample size, inaccurate sampling due to inflexing grasses, etc.) make their application difficult [28].

The practice of estimating the feeding value of grasslands based on the plant species that make up the grassland began in the mid-20th century and was primarily driven by economic interests. Ernst Ludwig Klapp [29] (Germany) and Ferenc Balázs [30] (Hungary) developed seemingly similar evaluation systems, but based on partially different principles [30]. Klapp et al. [29] published the relative forage value of about 350 grassland species. Species were ranked on a 10-point scale, with the most valuable species assigned a value of +8, sturgeon and other species not grazed by animals assigned a value of 0, and toxic species assigned a value of −1. The forage quality of the plant species was determined based on the following criteria:

-

Protein, fibre and mineral content, as determined by chemical analysis.

-

Tastiness and palatability to livestock.

-

Valuable plant parts (leaf, stem, flower, fruit).

-

Duration of the full value (as feed).

-

Usefulness and harvestability of the species.

-

Harmful and toxic properties.

-

Allowable proportion in the plant population (e.g., for toxic plants).

If worthless and toxic species are present in high proportions in the plant population, the overall value of the population will decrease accordingly. To quantify this, Klapp et al. [29] considered the following:

• Forage value of toxic plants up to 3% cover—1, between 3 and 10%—2, above 10% cover—3.
• The number of dicotyledonous species that contaminate hay is reduced by 1–2 values for cover greater than 10%.
• A separate assessment applies to grasses and weeds that are highly detrimental to feed value.

Some authors (like [29,31,32,33]) consider grasses more valuable forage than legumes (although their protein content is much lower) and apply only one negative category to all harmful species from the forage point of view. It is also unacceptable that the quality score for a species should decrease as its cover increases, because as the cover of the species increases, the species does not become more toxic or thorny. This distorts the estimated feed value of the species and thus also the value of the grassland. The main flaw of these methods is that they ignore the height difference between species. However, this has a significant impact on the quality of grassland forage. Thus, they can only perform the feed value estimation function with a strong bias, since they do not take into account the effect of the height difference of plant species on the relative mass of each species, from which we learn the contribution of each species to the feed value of grassland.

Similar errors have been found in, among others, the technique developed by Nitsche [34] and modified by Briemle [33,35] and Briemle and Ellenberg [36], using a 9-value scale (1–9), which has become known as “Futterwert (FW)”. During the development of the method of Briemle [31,32], in addition to 4 additional 9-valued scales (mowability, grazability, trampling tolerance and fallow deer), a functional trait was added, comprising sedges, other dicotyledonous species, herbaceous legumes, woody legumes, and woody shrubs. A habitat rating (extensive grassland, economic grassland, arable and garden, fallow, woodland edge species) was also made. This method was developed for 680 species. Although this method generally (and according to more recent scientific results) provides a good species relationship [37,38], ignoring the height of plant species may still strongly distort the results obtained.

Our study aims to review the process of forage value determination of species and to provide a dataset of 2310 vascular plant species in the Pannonian biogeographical region. Our approach joins the collection of a series of comparative European indicator values [39,40] with a section on grassland management.

2. Materials and Methods

2.1. Method for Estimating Production and Forage Value of Grasslands

During our revisional work, we went through Balázs’s [30] method step by step. Some parts were fully adopted, while others were supplemented and developed to obtain as accurate a picture as possible of the feed value of grasslands.

The first step is to employed the three-dimensional recording method developed by Balázs [30,41] to determine the amount of hay produced. This method is based on a coenological recording but the average height of the plant species is taken into consideration in addition to their cover. It is therefore a suitable method for estimating the relative yield of each species. The relative yield can be converted to an absolute value and dimension by multiplying by a mass coefficient. This method meets the needs of practical grassland farmers and grassland management specialists in determining the amount of green yield (aboveground biomass) or hay produced by the plant community. Using a classical 4 × 4 m quadrat method, we created coenological records to identify species and estimate their cover as accurately as possible, using D_B values. These were obtained by dividing the quadrat area once or several times. 1 D_B value as a unit is 1/32 of the quadrat area (3.125% cover). One record can have a maximum of 32 D_B values, unless the association is multi-level, in each case each level must be recorded separately. The smallest D_B value is 0.2, which represents 0.625% coverage.

The first step of the method is to calculate each plant species’ relative green mass.

t = D_B × m

t: relative green mass of the plant species, D_B: cover of the species, m: average height of the plant species [cm].

The m value is usually species-specific, but (because of the different growing conditions and weather conditions) it is always advisable to measure it in the field when recording. 1 ‘m’ corresponds to 0.01 m of plant height. Measuring the average height of the shoots (stem + leaf) is essential—protruding stems should be excluded.

Relative green mass of quadrats representing the grassland (T): sum of the relative green mass (t) of each plant species (T = ∑t).

The average height of the grassland (M) is obtained by dividing the sum of the relative green mass of the species in the grassland by the total cover expressed as ∑D_B.

M = T/∑D_B

M: average height of the grassland [cm], T: relative green mass of the grassland, which is the sum of the species relative green mass meaning yield, ∑D_B: total cover of the grassland [D_B].

In pastures, we have the opportunity to solely determine the cover values of each species as it is continuously grazed. In this case, we can only choose the average height of ungrazed plant individuals (e.g., outside the grazed paddock) as a control.

The B numbers are used to convert the relative yields into absolute quantities. B numbers express the real green mass of a t-value, and B_M expresses the real green mass of a 0.01 m high section of 100% total cover grassland per hectare. Therefore, when using the B numbers, the D_B values of the records must be converted to % cover (b).

Balázs (1951) [42] suggests the following values for Carpathian Basin pannonic grasslands:

B_t: for grasses: 0.125 t ha⁻¹

for Medicago sativa: 0.147 t ha⁻¹

B_M: for grass: 0.400 t ha⁻¹

for Medicago sativa: 0.470 t ha⁻¹

There are two methods for calculating the yield.

• The average height of the grassland (M) is multiplied by B_M and then multiplied by the actual cover [D_B].
• The grassland’s relative green mass (T-value) is multiplied by [B_t].

The grassland yield cannot be fully utilised, so the stubble height (s) must be subtracted from the average height. Suppose the yield is to be converted into hay value. In that case, the resulting value must be divided by the drying factor (E), which typically ranges from 2.5 and 3.5, depending on the weather, plant species and state of development. For optimum utilisation time, this value is 2.5 for dry grassland, 3 for mesophilic grassland and alfalfa and 3.5 for red clover.

P = [(M − s) × B_M × b] ×100⁻¹ × E

P: amount of grass yield [kg × ha⁻¹]

M: grassland height [cm]

s: stubble height [cm]

B_M: weight of a 1 cm high grass section at 100% total cover; maturity: 0.4 [t × ha⁻¹]

b: D_B values in cover %

E: drying factor

The productivity of a grassland can only be accurately determined if the yields of all utilisation periods are added together. However, only the quantities calculated for the first growth are also suitable for comparing the yields of several grassland areas. The method used by Balázs (1960) [30] classifies pannonic grasslands into five classes of productivity:

I.

Class I: grassland producing > 6.0 t dry matter × ha−1

II.

Class II: grassland producing between 4.5 and 6.0 t dry matter × ha−1

III.

Class III: grassland producing between 3.0 and 4.5 t dry matter × ha−1

IV.

Class IV grassland producing between 1.5 and 3.0 t dry matter × ha−1

V.

Class V grassland producing between 0.0 and 1.5 t dry matter × ha−1

2.2. Forage Value of Species (k)

The forage value of grassland is determined by the proportion of grassland components (from the most valuable species of legumes and grasses to weed species that threaten animal health) present in it [10,43]. Therefore, each grassland species should be assessed separately for accurate evaluation, and grasslands should be managed as a mixture of these species. Classification should be carried out for both beneficial and harmful species. It is essential to highlight the latter species, as most methods [29,31,32] do not assign more than one quality category to them, despite the significant differences in the harmful effects of these species. Balázs [30,41,42] established the basis for a system to characterise the qualitative differences between different grass species and between different plant communities. The seasonal variations of these are also significant [44,45], and can be influenced by the nutrient supply of the plants [46] as well as by the palatability of the different plant groups [47,48,49,50], which can also be investigated using this method. The species quality value (k) is a relative score that indicates the relative role of plant species in forage production.

The forage value categories were developed based on the following criteria. The forage value of the best quality species is similar to that of the abstract feeds, and therefore, they were given a separate category. These species (+6 and +7) improve the forage value of the grassland. All species eaten by the livestock (regardless of taxonomic affiliation) and whose consumption does not have any detrimental consequences are classified on a scale of 1 to 5. Species that the animal does not eat or whose consumption may have harmful repercussions are considered dangerous. These species are given a negative sign and are scored on a scale of 1 to 3. Neutral species that cannot be classified in either group are assigned a value of 0.

Forage value (k-value) categories by Balázs:

Species of value +7: their presence improves the quality of grass fodder. They are high in protein, have high nutritional value, rich foliage, excellent digestibility (over 75% digestible organic matter in the last 10 days of May), slow senescence, rapid growth, are palatable to animals, very well adapted to different habitats, and are trampling-tolerant.

Species of value +6: Also species with excellent nutritional value (65–75% digestible organic matter in the last 10 days of May), slow to senesce, good yielding, but less adapted to the site (possibly shorter life).

Species of value +5: grassland plants that still provide excellent quality forage (but may contain more firming tissue), with a good leaf-stem ratio and 55–65% digestible organic matter in the last 10 days of May. They are not rough, contain little silica, have excellent palatability, are not flaky and do not contain unpleasant odours and flavours. They are suitable for rough grazing and mowing and can be used to make good-quality hay. They are good post-emergence species, but may have a shorter life span.

Species of value +4: also provides good-quality forage, but their leaf-to-stem ratio is worse than the previous group’s. They contain more solidifying tissue and yield less after grazing and mowing. Legume species which are of minor importance from a forage perspective are also included.

Species of value +3: They have a reduced forage value, but if used at the right time, they provide good-quality forage. They are usually slightly rough, flaky, or leathery, have a lot of firming tissue or are less palatable to animals. This category also includes other dicotyledonous species of the highest forage quality.

Species of value +2: At best forage straw quality, when mown young, they can be used as ballast fodder to “dilute” better-quality forage. The livestock usually grazes on them when young. Their nutritional value is relatively low. They have a relatively poor germination rate and tend to go stunted quickly. Most are rough, flaky, or hairy but lack strong, bulging stems. This group includes many dicotyledonous species that the livestock readily eat when young. In small quantities, they improve the palatability of the feed.

Species of value +1: They provide forage of litter quality at most. They are not grazed when young, senesce quickly and lose their foliage rapidly. They may contain silicic acid or other slightly harmful substances, but in small quantities, they do not cause any harm to the livestock. Their nutrient content is very low.

Species of value 0: These species are suitable for foraging at a particular stage of their development, but do not become particularly damaging later on. They are usually small species and are insignificant from a forage point of view.

Species of value −1: Unpleasant smelling, rough or hairy, stalky species, which multiply rapidly and take up a lot of space in front of valuable species. The livestock never eats them, but their possible consumption is not harmful. They are also only suitable for bedding out of necessity.

Species of value −2: These plants are already highly damaging in sward and forage. They usually also contain toxic substances that cause damage when added to fodder, or they are large, thorny plants that occupy a significant portion of the grassland.

Species of value −3: The most forage-poor plants in our grasslands. They are particularly dangerous in pastures because, unlike species favoured by animals, they can reproduce undisturbed without weed control mowing or due to the lack of grazing.

The following criteria were taken into account by Balázs [30] when determining the k-values of plant species:

-

Proportion of valuable plant parts (leaf, stem, flower, fruit).

-

The duration of wholesomeness (as feed).

-

Usefulness and harvestability of the species.

-

Pests and toxicity.

-

Permissible proportion in the plant population (e.g., for poisonous plants).

-

Grazability and regeneration time.

In addition to the above-mentioned features, during the refinement we also took into account the following characteristics, thereby extending the accuracy:

-

Protein, fibre, mineral content and protein/fibre ratio.

-

Digestibility of the main species and its variation in the first growth.

-

Palatability and preference by livestock.

The digestible organic matter (DOM) content of the samples taken from the trimmings of the plots was determined in vitro using rumen fluid digestion according to the method of Tilley and Terry [51].

As a result of our survey, we assigned modified k-values to 2310 species (Appendix A).

2.3. Relative Economic Value of Species (kt)

If the relative green yield (t) of a species is multiplied by the species-specific quality score, the relative economic value of the species in the grassland is obtained:

kt = k × t

The sign of the kt-value of a species can be + or −, depending on the forage value of the species. Species with a k-value of 0 will have a kt-value of 0. Subtracting the negative kt values from the positive kt values (∑kt⁺ −∑kt⁻) will give the difference in the relative kt value of the grassland (∑kt). If ∑kt⁻ is greater than ∑kt⁺, then the grassland has no forage value. The grassland quality (K) is obtained from the value ∑kt by dividing by the average T-sum.

K = (∑kt × T⁻¹)

Semi-natural species rich grasslands always have a K-value below 5 because even if the dominant species has a K value of 5, the other, sub-ordinated species in the plant community will degrade the K value of the grassland. The quality of the green grass or hay produced can be determined regardless of quantity. Balázs [30] classifies grasslands into the following five classes according to their quality:

I.

Class: very good, high-quality grassland, K-value: >4.

II.

Class: good, quality grassland, K-value: 3–4.

III.

Class: medium, quality grassland, K-value: 2–3.

IV.

Class: poor, poor-quality grassland, K-value: 1–2.

V.

Class: bad, poor-quality grassland, K-value: 0–1.

The value of grass productivity (P)—based on the method of Balázs (1960)

The value of a grassland’s productivity is the grassland’s value-producing capacity divided by 100. This gives the point value of the economic value of the grassland.

P = kt ×100⁻¹

Multiplying this by a given value in a given currency at current feed prices, the economic value produced by the grassland can be calculated.

2.4. Sample Area and Field Survey

In this study, we investigate various habitats where environmental factors differ significantly. One study site is a dry grassland where the thin-leaved Festuca pseudovina is the dominant species, and the other is a wet grassland where the broad-leaved Festuca arundinacea is the dominant species. Festuca pseudovina is a less-studied species, but Festuca arundinacea is a well-studied species due to their widespread use, which is attributed to their adaptability, drought tolerance, and high biomass production potential in temperate grassland ecosystems [52,53]. Festuca arundinacea-dominated grasslands make significant contributions to biodiversity and ecosystem stability. These grasslands are home to a diverse array of herbaceous species and contribute to soil organic carbon sequestration, thereby enhancing the overall ecosystem performance [15,16,17]. This species can improve forage quality in mixed pastures [54,55].

To test the method in our thesis, we selected two study sites (Figure 1):

-

A wetland dominated by Festuca arundinacea (Agrostio-Deschampsietum caesitosae Ujvárosi 1941) (Mende, Hungary).

-

A dry grassland dominated by Festuca pseudovina (Achilleo-Festucetum pseudovinae Soó (1933) 1947 corr. Borhidi 1996 (Bösztör, Hungary).

The 4 × 4 m sample quadrats were set up in a 7 × 7 Latin square layout. The quadrats were harvested by mowing the grass with separate mowing frequencies with a mower, leaving 0.04 m of stubble. Plots mowed 2 times a year were harvested on 30 June and 10 October; plots mowed 3 times a year were harvested on 18 May, 30 June and 10 October; and plots mowed 4 times a year were harvested on 18 May, 30 June, 5 August and 10 October. The test years were from 2016 to 2019.

2.5. Statistics

A non-parametric statistical method was used to analyse the cover value of species across different groups of complementary materials. According to the Shapiro–Wilk test, these variables were not normally distributed (p < 0.05). Therefore, the non-parametric Spearman test was used. All statistical analyses were performed using the XLSTAT statistical and data analysis software version 2024.4.1 [56].

3. Results

3.1. Case Study

The wet grassland dominated by Festuca arundinacea was investigated at the Mende sample site. The average cover was as follows: 55.5% grasses, 17.3% legumes, 18.7% forbs, 5.7% toxic species and 0.4% other species (Figure 2)

Festuca pseudovina is the main constituent of the population in the other sample area in Bösztör. Grasses cover 55.7%, legumes cover 5.7%, forbs cover 13.8%, toxic species cover 6.6%, thorny species cover 0.2%, and other species cover 1.0%. Botriochloa ischaemum, a typical C₄ grass species found in arid ecological sites, is often observed in drier years.

3.2. Result of the Yield Estimates

The Spearman correlation value of the green mass estimated by the Balázs method in the dry grassland at Bösztör was 0.44, indicating a moderately strong correlation with the values of the mown biomass samples. The wet meadow records at Mende had a Spearman correlation value of 0.96, indicating a robust correlation with the biomass sample data (Figure 3).

When examining the digestible organic matter yield, similar results were obtained, but the difference between treatments was even greater (

Table 1. Procedure for calculating the digestible organic matter yield of grassland.

Year	Number of Moving	Number of Growth	∑DB 1	T 2	Σkt 3	K 4	P 5	Sz 6	DM 7	DOM 8	DM × DOM
								(t/ha)	(t/ha)	(%)	(t/ha)
2016	2	1	31.7	4655.8	19,395.2	4.2	194	17.04	4.94	58.2	2.88
		2	31.4	3641.1	13,842.1	3.8	138.4	13	3.43	59.76	2.05
	3	1	31.9	4217.4	17,977.2	4.3	179.8	15.27	3.69	69.49	2.56
		2	31.2	2869.4	11,333.5	3.9	113.3	9.92	2.81	61.19	1.72
		3	31.7	3337.5	13,439.1	4	134.4	11.76	2.62	61.37	1.61
	4	1	31.9	2740.3	11,897.8	4.3	119	9.37	1.62	72.72	1.18
		2	31.8	2522.1	10,534.8	4.2	105.3	8.5	1.86	69.16	1.28
		3	31.3	2628.5	9379.1	3.6	93.8	8.95	2.20	62.8	1.38
		4	31.9	3732.1	15,218	4.1	152.2	13.33	3.18	65.66	2.09
2017	2	1	31.9	4883.4	21,015.1	4.3	210.2	17.94	4.86	62.79	3.05
		2	31.9	3625.1	13,990.7	3.9	139.9	12.91	3.52	64.89	2.28
	3	1	31.9	3357.4	14,903.2	4.4	149	11.83	2.81	75.4	2.12
		2	31	3874.7	15,398.2	4	154	13.95	4.26	63.68	2.71
		3	31.8	3338.8	13,395.8	4	134	11.77	2.69	66.72	1.80
	4	1	32	2828.4	11,676.8	4.1	116.8	9.72	1.4	77.78	1.09
		2	31.7	2772.6	11,722.1	4.2	117.2	9.5	2.01	67.23	1.35
		3	31.3	2080	7060.7	3.4	70.6	6.76	1.68	70.91	1.19
		4	31.8	3210.3	13,085.9	4.1	130.9	11.25	2.51	68.58	1.72
2018	2	1	31.3	4121.5	18,399.7	4.5	184	14.92	6.30	70.16	4.42
		2	29.7	3349.4	13,384.6	4	133.8	11.91	5.04	67.11	3.38
	3	1	29	2168.2	8886.2	4.1	88.9	7.22	2.85	75.72	2.16
		2	31.6	3101.3	12,528	4	125.3	10.83	4.15	62.2	2.58
		3	32	3487	13,062.8	3.7	130.6	12.35	5.60	64.65	3.62
	4	1	28.1	2013.7	8678	4.3	86.8	6.65	2.64	75.33	1.99
		2	30.7	3754.6	15,741	4.2	157.4	13.48	4.48	71.16	3.19
		3	27.3	2275.7	8867.7	3.9	88.7	7.74	3.09	70.8	2.19
		4	29.8	2903.5	11,667	4	116.7	10.12	4.42	70.96	3.14
2019	2	1	32	5095.4	22,195.1	4.4	222	18.78	6.64	55.7	3.70
		2	31.3	4347.2	17,119.1	3.9	171.2	15.82	5.32	69.02	3.67
	3	1	31.8	4151.4	16,261.4	3.9	162.6	15.02	3.45	68.38	2.36
		2	31.8	3372.7	12,108.5	3.6	121.1	11.9	3.26	69.73	2.28
		3	31.3	4652.2	17,651.6	3.8	176.5	17.04	3.96	68.09	2.69
	4	1	31	3902.3	15,072.3	3.9	150.7	14.06	4.29	80.75	3.46
		2	30.8	2326.7	8891.7	3.8	88.9	7.77	2.39	73.64	1.6
		3	32.1	2287.1	8372.9	3.7	83.7	7.54	2.37	74.21	1.76
		4	31.5	3734.3	14,149.3	3.8	141.5	13.36	2.98	74.03	2.21

Legend: ¹. Coverage according to Balázs; ². Sum of relative yields; ³. Sum of the relative yield and forage value of the species; ⁴. Quality score of the plant population surveyed according to Balázs; ⁵. Productivity of the plant population recorded: Σkt × 100⁻¹; ⁶. The quantity of green plant yield (without 4 cm stubble) (t × ha⁻¹); ⁷. dry matter content; ⁸. digestible organic matter.

). The productivity (P) in the table combines qualitative and quantitative indicators of the grassland yield in one number, allowing the quantitative and qualitative values of the yield of different grasslands to be standardised, and thus making it suitable for comparing grass or hay from different grasslands.

Table 1 shows the biomass estimation process. If the organic matter digestibility of the grassland is known, this method can also be used to calculate the yield of digestible organic matter per hectare of grassland. Our results showed that among the treatments tested, the 4 × mowed plots gave the highest production, while the 2 × mowed plots gave the lowest production.

3.3. Estimation of Forage Value of the Grassland

Table 2. Forage value estimates.

Species	b (DB)	m (cm)	t	k	kt
Agrostis stolonifera	1.6	33	52.8	5	264
Festuca arundinacea	19	48	921.6	4	3686.4
Poa pratensis	2.9	32	92.2	6	553
Trifolium hybridum	0.3	37	11.8	7	82.9
Trifolium pratense	3.2	40	128	7	896
Achillea collina	1	46	44.2	2	88.3
Cichorium intybus	0.6	44	28.2	3	84.5
Daucus carota	0	44	1.4	1	1.4
Pastinaca sativa	0	40	1.3	−1	−1.3
Plantago lanceolata	0.1	23	1.2	3	3.7
Plantago media	0.6	15	9.6	2	19.2
Ranunculus acris	0	35	1.1	−2	−2.2
Ranunculus repens	0.6	22	14.1	−2	−28.2
Rumex acetosa	0.1	50	5.3	−1	−5.3
Taraxacum officinale	1.6	23	36.8	3	110.4
Σ	32		1349.6		5752.7
M		42.3
K				4.3
P					57.5

Legends: DB: coverage of plant species; m: height of plant species, cm; t: relative mass of plant species = D_B × m; T: ∑t, relative green matter of the quadrat/the grassland; k: quality of plant species; kt: relative economic/forage value of plant species (t × k); +kt: sum of positive values; −kt: sum of negative values; K: forage value of the grassland; P: economic/forage value of the grassland (∑kt/1000).

shows the procedure for calculating the forage value of the grassland. The relative yield (t) obtained by multiplying the species cover (b) by its average height (m) is multiplied by the relative forage value (k) of the species. This is done for each species and the sum of the obtained values (Σkt) is divided by the sum of the relative yields (T) to obtain the average forage value (K) of the grassland, which in this case is 4.3, corresponding to a very good quality. Dividing Σkt (5752.7) by 100 gives the productivity of the grassland, which for the grassland under study was 57.5.

4. Discussion

Determining the forage value of a grassland is a difficult task because, in addition to objective factors, it is also influenced by many subjective factors, such as habituation. Among the objective indicators, Balázs [30] considers the following to be the most important: the degree of development (age), nutritive value, protein content, starch value, fibre and silica content, digestibility, pungency, roughness, fluffiness, hardness, taste, odour, acidity, bitterness, toxicity, etc. The above characteristics indicate that the quality of grassland plants can significantly vary over a broad range. It is also clear that purely chemical analysis alone is insufficient to determine the value of forage, since, for example, species with negative morphological characteristics (e.g., stinginess) are not consumed by livestock even if they have excellent nutritional properties.

Our results demonstrate that the Balázs method is suitable with a high degree of confidence in practice. By changing the minimum D_B value from 0.2 to 0.05 (0.156%). The process can be used to estimate the cover of rare, small species more accurately, which is particularly important in conservation and diversity studies. The method outlined above can also be used in conjunction with % cover estimation and altimetry. However, for those less experienced in cover estimation, D_B value-based estimation is recommended for greater accuracy. During the course of our work, we have modified the Balázs k-values for several species based on recent results and have also significantly expanded the list of species with k-values based on this criterion to 2310 species (Appendix A) (Balázs reported 401 values).

The corrected Balázs three-dimensional recording method is well suited for estimating yield and grassland quality, as verified by direct yield estimation and the laboratory tests described above. The process requires virtually no equipment and is inexpensive to implement. The calculated productivity values represent the economic production of the studied grassland, providing a valuable tool for agricultural planning and management. Thanks to its quick and straightforward application, it can also be used for grassland mapping. It enables the monitoring of the temporal and spatial variation in stands as well as the effects of various treatments. In the future, the plan is to develop indicators that will allow for the expression of more accurate grasslands’ economic and forage value than hitherto. The method is also helpful for conservation purposes, as the three-dimensional coenological releves make it suitable for monitoring the management of protected areas [57]. The technique can measure the proportion of aboveground species production, which is an essential indicator of the carbon balance in grassland ecosystems, especially in lush grasslands [58,59,60]. It can therefore be used effectively in carbon balance studies without damaging the grasslands.

Natural grasslands, particularly in temperate and semi-arid regions, exhibit significant seasonality in biomass production, which can vary substantially due to climatic factors such as temperature and rainfall, as confirmed by our experiment.

The productivity of grassland is also closely related to the frequency of mowing. While increased mowing can increase direct biomass production by minimising interspecific competition, it can also reduce the overall long-term productivity of the grassland by depleting the necessary resources stored in the root system. In our study, 4 × mowing gave the highest productivity and the best forage quality. Regular mowing has been shown to minimise biomass accumulation due to the continuous removal of green tissue, which is essential for photosynthesis and energy storage [61]. Research also shows that optimal productivity can be achieved at specific mowing frequencies, such as every two years, which allows for the maintenance of adequate herbaceous cover while promoting biodiversity [62,63]. The correlation follows a bell curve where very low and very high mowing frequencies can reduce productivity, suggesting a “sweet spot” that balances the competitive dynamics between grassland species [64].

The frequency of mowing has a direct impact on the species richness and composition of grasslands. Studies show that higher mowing frequency is often associated with decreased species richness. In particular, grasslands subjected to intensive and frequent mowing may have reduced plant species abundance and diversity. This occurs because frequent mowing tends to favour disturbance-tolerant species over species less resistant to regular mowing, resulting in a homogenised plant community dominated by a few species [65,66]. For example, Socher et al. (2012) [67] found that more frequently mowed grasslands had lower species richness than those that were mowed less often, confirming a consistent trend observed in different studies [67]. Furthermore, Binder et al. 2018 [66] reported that while mowing intensity was positively correlated with species richness in controlled experiments, more frequent mowing tended to reduce overall diversity [66]. The interaction between mowing frequency and plant diversity is complex, and an intermediate approach, such as mowing every other year, has been proposed to maintain a richer plant community [62,68].

Natural arid grasslands, such as those found in semi-arid regions, typically show a high allocation of belowground biomass, and studies suggest that root mass accounts for 67% of total plant biomass in these ecosystems [69]. Belowground biomass is vital for drought resilience and resource acquisition, thereby increasing the overall resilience of the grassland [70]. For example, in alpine regions, aboveground biomass (AGB) averages around 68.8 g/m², which varies with annual rainfall, suggesting that drier years result in decreased AGB due to limited water availability [71]. Furthermore, the dynamics of aboveground and belowground biomass production can be significantly influenced by climate, especially precipitation, which plays a crucial role in shaping productivity levels [71,72]. In a study of a northern dry grassland ecosystem, water availability was the primary limiting factor in dry years, reducing net primary productivity (NPP) under these conditions [73]. Dry grassland productivity is often resilient to short-term changes in precipitation, highlighting the importance of moisture patterns in determining biomass output [74,75].