How Do Territorial Relationships Determine the Provision of Ecosystem Services? A Focus on Italian Metropolitan Regions in Light of Von Thünen’s Theorem

Abstract

This study aims to explore the relationship between the provision of ecosystem services (ESs) and other territorial characteristics. Taking Italian Metropolitan Regions (MRs) as case studies, the gradient of specialization providing a set of ESs in different territorial contexts is examined using the National Strategy for Internal Areas (SNAI) territorial classification. The main objective of this research is to understand whether there is a spatial pattern of location of different ESs within metropolitan SNAI areas. Inspired by Von Thünen’s spatial economic theories, this study explores how proximity to urban centers influences land use and ES specialization. Through land use analysis and the calculation of a SI, we evaluate patterns in ES supply, based on a benefit transfer approach. The results show that the MRs provide about EUR 14.6 billion per year in benefits, equivalent to 15% of the national wealth in environmental goods and services. At the SNAI area scale, internal areas have the highest average economic values per hectare, while the central areas have lower economic values. This trend is confirmed by the calculation of the specialization index (SI) in line with Von Thünen’s theorem as follows: central areas are specialized in the provision of bundles of ESs related to intensive land use (e.g., food production), while the peripheral areas are specialized in the supply of regulation ESs related to more natural areas. The findings underline significant policy implications for metropolitan planning, stressing the need for the balanced management of ESs to address urban demands and enhance resilience. This research contributes to understanding the spatial dynamics of ES supply, offering a basis for tailored interventions in metropolitan and national contexts.

1. Introduction

Metropolitan Regions (MRs) are territorial bodies of vast areas consisting of at least 250,000 inhabitants. In Italy, they have been provided for in Constitutional Law No. 3/2001, designated by Law 56/2014 that replaced the pre-existing provinces. The European Commission recently pointed out that the development of MRs will influence the direction of the Union’s future [1]. MRs vary in size and socio-economic characteristics, and the interaction with non-MRs contributes to the nation’s competitiveness and development. MRs also play a significant role in the country’s economy by providing job opportunities and helping generate income. Economic growth and MR expansion [2], along with changes in land use and land cover, have transformed their landscapes. MRs represent the focus of several studies aimed at finding a balance between their land use evolution and the maintenance or improvement of ecosystem services (Ess) [3,4,5,6]. Urbanization represents the most critical component of these transformations, as it is the most important factor for ES variation [7,8,9,10,11]. Land use changes and artificial area expansions are socio-economic processes that have evolved rapidly in MRs since the 19th and 20th centuries, along with industrial development and rural exodus due to the loss of active population in the agro-forestry sector [12,13,14,15]. Recently, with the transition from an industrial to a post-industrial economy [16], the urban–rural boundary is increasingly disappearing as urban areas sprawl [17]. Population decline in rural areas has caused a change in land use and affected some economic activities related to forestry, pastoralism, and agriculture. This has changed the agricultural and forest landscape [18] and ES supply [8,19,20]. Transformations that have affected land use have profoundly altered the structure and functionality of ecosystems, making them more vulnerable to climate change [21] and hydrogeological disruption. Urban areas depend on rural areas because many of the ESs essential for the economic and social sustenance of the population originate in these areas. The increasing population in urban areas generates a greater demand for ESs, and economic activities in urban areas pressure ecosystems and greatly influence service provision. A shift from compact models toward fragmented structures can be observed in Mediterranean cities [22,23,24,25]. The effects of the recent wide urbanization of these territories, from the point of view of ecology and urban metabolism, are generally the increased consumption of resources [22], loss of environmental connectivity and biodiversity [23,24], and dependence on material and energy sources outside urban areas [25]. The assessment of ES alteration triggered by land use change can be supported by spatially explicit approaches such as the concept of the urban-to-rural continuum [26,27,28,29,30]. Although the latter has the limitation of generalizing existing spatial patterns in a MR, it is a useful tool capable of studying major trends in the spatial ES variation related to different types of territories. A recent study based on the urban–rural gradients in the Barcelona MR [28], among others, jointly analyzed air purification and outdoor recreation services demand–supply by dividing Barcelona’s MR into concentric centers with increasing distance from the city center. The results show the relationships between the spatial patterns of ESs and the distances into which the urban-to-rural continuum has been divided. In a study by Marino [19], changes in the supply–demand balance and variations in the economic value of ESs were investigated through an analysis of the processes of permanence and transformation that affected inland areas in Italy from 1990 to 2018. In our work, to explore possible ES zoning through the center–inland gradient of Italian MRs, a similar approach has been used considering the SNAI (National Strategy for Inner Areas) zones. The SNAI, which has been launched in 2012, subdivides the Italian municipalities according to the distance from the pole areas where essential services or rather, education, health, and mobility are concentrated. Based on these premises, the SNAI distinguishes six types of municipalities, half in central areas (A, B, C) and half in inland areas (D, E, F). Instead of using concentric circles, the inner area strategy distinguishes municipalities based on the different travel times required to reach the hub. The inter-municipal centers (B) are less than 20 min from the poles; belt municipalities (C) are 20 min away; intermediate areas (D) up to 40 min, peripheral areas (E) up to 75 min; and finally, ultra-peripheral areas (F) have times even longer than the latter. The SNAI has been developed to increase essential services in inland areas, enhancing the natural, cultural, cognitive capital, and system of production of a place. According to Armondi [14] the interest in inland areas derives from the territorial capital in inner areas being widely unused due to the qualitative and quantitative demographic changes that have occurred. From a landscape ecology point of view, land abandonment in remote areas within MRs generates forest expansion that, in turn, homogenizes the landscape. In this sense, it contributes to the deterioration of the agro-silvo-pastoral mosaic produced by peasants’ activities [31], thus representing a threat to the spatial heterogeneity’s ability to enhance biodiversity and ecosystem functions [32,33]. As ES provision depends on a mix of environmental and socio-economic aspects [34], it is crucial to understand the territory and distribution of SE across the various spatial gradients. The Von Thünen model has been applied in the recent literature in several cases where the zoning of a given phenomenon can be observed [35,36,37,38]. Von Thünen proposed his spatial economic theory at the beginning of the19th century, arguing that the arrangement of crops or the degree of intensity of cultivation around an urban market depends on transport costs. A market town at the center of a fertile plain represents the originally ideal city of the theory. The territory around this city is divided into different concentric circles, based on six agricultural typologies moving from intensive to extensive activities as follows: (a) horticultural products, milk, and flowers; (b) forestry; (c) rotating system for cereals and root crops; (d) rotational crop and pasture systems; (e) three-field systems; and finally, (f) livestock. However, it does not consider the road network when calculating transport costs. Based on this context, our research aims to study the urban–rural dynamics in MRs by revisiting Von Thünen’s theorem. We seek to understand whether there is a model of spatial specialization in the ES supply in MRs and the role of different SNAI areas in providing these benefits. To this end, the present study aims to check if there is spatial zoning of provisioning and regulation ESs in 14 MRs, starting from land use maps and travel distance SNAI zones. Based on relative land use distribution, the ultra-peripheral areas, for example, are expected to be the ones most able to offer high rates of regulation ESs, largely due to woodland abundance and the low urbanization rate. Conversely, in the hubs, where built-up areas and arable land are concentrated, we expect (i) a low rate of regulation ES provision and (ii) a high ES supply. The elaboration of SI for each SNAI zone is used to investigate the characteristics of spatial zoning of provisioning and regulation ESs through the MRs. Our results can contribute to the debate regarding the socio-ecological impact of master plans of land use changes.

2. Materials and Methods

The research was conducted following a methodological framework divided into four phases (Figure 1). First, we studied the land uses of MRs by considering the perimeters of SNAI areas (Step 1). Next, we associated economic coefficients (EUR/ha) to land use classes, following the benefit transfer approach and updating the results of previous work [8] (Step 2). This allowed us to examine the level of specialization in the provision of ESs, on the one hand, and the variability of this specialization within MRs, on the other (Step 3). Finally, in our discussions, we suggest governance proposals to support the provision of ES in urban areas (Step 4).

2.1. Study Area

The study area consists of circa 47 thousand km² associated with the 14 MRs of the following Italian territories: Torino (6824 km²), Rome (5357 km²), Palermo (5001 km²), Bari (3860 km²), Bologna (3701 km²), Catania (3573 km²), Firenze (3514 km²), Messina (3255 km²), Reggio Calabria (3203 km²), Venezia (2461 km²), Genova (1824 km²), Milano (1575 km²), Cagliari (1243 km²), and Napoli (1171 km²). As illustrated in Figure 2, not all regions have MRs, as follows: this is the case of Valle d’Aosta, Valle d’Aosta, Trentino Alto-Adige, Friuli-Venezia Giulia, Umbria, Marche, Abruzzo, Molise, and Basilicata. Conversely, Sicily is the only region with 3 MRs.

To characterize the study areas according to SNAI categories, the shapefiles of the aforementioned 14 study areas were intersected with the perimeters of the 6 SNAI zones proposed by ISTAT as follows [39]: hub (A), intermediate hub (B), or belt (C) municipality, belonging to central areas; and intermediate (D), peripheral (E), or ultraperipheral (F), belonging to inland areas. Rome has the largest hub in Italy (1469 km², corresponding to 27% of its area). The typology of the SNAI areas with the highest representation is the belt type, whereas in the 3 Sicilian MRs, peripheral municipalities prevail. As the specific composition in the SNAI zones of each MR varies depending on the time required to reach the hubs, each metropolis presents a different percentage of central and internal areas (see

Table 1. SNAI typology extensions in each MR (%).

	SNAI
MRs	A	B	C	D	E	F	TOT
Bari	7.1	19.1	44.3	28.4	1.1	0	100
Bologna	12.4	12	46.5	15.4	13.7	0	100
Cagliari	6.6	0	57.4	36	0	0	100
Catania	5.1	0	16.9	15.1	61.8	1	100
Firenze	11.7	10.8	45.7	16.5	15.2	0	100
Genova	12.7	1.4	41.4	28.2	13.3	3	100
Messina	6.5	0	5.3	18	35.6	34.6	100
Milano	15.4	7.6	77	0	0	0	100
Napoli	17	32.1	42.6	3.3	5	0	100
Palermo	3.2	0	18.5	19.1	45.4	13.8	100
Reggio di Calabria	8.2	0	10.6	38.4	36.8	6	100
Roma	27.4	0	12.9	50.1	9.6	0	100
Torino	4.5	2.4	44.2	26.3	22.6	0	100
Venezia	18.7	5.6	37.5	32.7	5.5	0	100

below). For example, if, on the one hand, Milano is composed of only central areas (hubs, intermediate hubs, belts), on the other hand, ultra-peripheral areas can be listed only in Catania, Genoa, Messina, Palermo, and Reggio Calabria.

To study land uses in the study area, the geographical layer of the Corine land cover (CLC), dated 2018 [40], was intersected with the perimeters of the SNAI areas in each MR. According to the methodology applied in our previous work [41], land use and cover classes have been grouped into the following 9 macro-classes: 100 artificial surfaces, 210 arable land, 220 permanent crops, 230 pasture, 240 heterogeneous agricultural areas, 310 woodland, 320 shrubs, 330 sparse or absent vegetation, 500 wetlands and water bodies. Macro-classes have been realized by grouping agricultural and forest classes at the II hierarchical level, while artificial areas and water bodies have been aggregated at the I level. As can be observed in

Table 2. Land macro-class extension in each SNAI typology (%).

			SNAI
		Area	A	B	C	D	E	F
LULC Classes	Class Description	km2	%
100	Urban areas	4423	34	7	41	13	4	0
210	Arable lands	12,658	10	8	40	20	18	3
220	Permanent crops	4043	8	6	22	35	24	10
230	Pastures	1107	2	1	51	39	7	0
240	Heterogeneous agricultural areas	6989	11	5	35	27	19	0
310	Forests	9978	5	3	27	31	28	7
320	Shrub and/or herbaceous vegetation	5494	5	1	20	26	38	13
330	Sparsely vegetated areas	1047	4	1	14	22	57	3
500	Wetland and water bodies	823	40	6	22	23	8	3

, arable land areas cover the largest surface in the total areas and are particularly present in the belt zones. Where the composition of the central and internal SNAI zones is concerned, it can be observed that artificial urban areas characterize more than a third of the hubs and belts (central areas), while in intermediate and peripheral zones (internal areas), landscapes are dominated by a great amount of wooded and semi-natural classes.

The landscape composition of each SNAI zone varies within MRs (

Table A1. Land use classes in each Central SNAI zone and each MR (BA = Bari; BO = Bologna; CA = Cagliari; CT = Catania; FI = Firenze; GE = Genova; ME = Messina; MI = Milano; NA = Napoli; PA = Palermo; RC = Reggio Calabria; RM = Roma; TO = Torino; VE = Venezia).

	BA	BO	CA	CT	FI	GE	ME	MI	NA	PA	RC	RA	TO	VE
LULC	A—Hubs
class	km2
	274	459	82	182	411	232	212	243	199	158	263	1469	309	460
	%
100	26	19	43	35	18	24	20	70	58	57	16	31	45	18
210	8	51	5	33	19			27	4		3	41	27	18
220	42			15	10	1	10		9	8	14	2
231	2											1	1
240	19	26	2	10	15	8	19	1	20	2	28	15	16	7
310	3	2			35	33	16	1	5	10	24	6	8
320	1	1	2	1	1	34	23		3	22	15	3	1
330				1			12		1		1		1
500		1	48	4	1							1	2	56
	B—Intermediate hubs
	km2
	737	445			380	26		119	376				162	138
	%
100	4	3			3	22		49	27				36	15
210	46	85			12			48	24				35	47
220	19				19	6			11
231					1			2					4
240	21	7			15	36		1	23				18	8
310	8	1			49	31			10				6
320	3	2			1	5			2
330									3
500		2												30
	C—belt
	km2
	1711	1722	741	605	1607	755	173	1213	498	927	339	692	3016	923
	%
100	16	7	11	21	7	5	13	27	32	9	7	10	9	11
210	55	56	15	6	15			63		29	9	16	33	64
220	1	1	2	12	18	1	31		5	20	24	16	1	3
231	20				1				18			2	3
240	3	20	20	36	15	14	12	3	31	13	26	22	20	13
310		11	23	7	41	58	19	4	9	7	18	23	25	1
320	5	4	25	6	2	21	23		3	20	12	11	7
330			2	11			2		2		3		1
500		1	3	1	1			1	1	1			1	8

and

Table A2. Land use classes in each Internal SNAI zone and each MR (BA = Bari; BO = Bologna; CA = Cagliari; CT = Catania; FI = Firenze; GE = Genova; ME = Messina; MI = Milano; NA = Napoli; PA = Palermo; RC = Reggio Calabria; RM = Roma; TO = Torino; VE = Venezia).

	BA	BO	CA	CT	FI	GE	ME	MI	NA	PA	RC	RM	TO	VE
	D—Intermediate km2
LULC class	1095	568	44	540	581	514	587		39	954	1230	2682	1795	804
	%
100	3	2	6	6	1	2	8		6	5	4	7	2	7
210	52	8	16	14	12		2			34	8	27	5	63
220	20		2	18	18		16		1	24	32	9		3
231	1			34			27				1	1	2
240	6	29	15	10	15	8	17		43	16	19	24	12	11
310	4	51	33	10	51	76	26		29	3	23	23	40	1
320	13	9	25	7	2	13	5		15	18	11	7	28
330	1	1	2	1					6		2		10
500			1							1		2		15
	E—Periphereal km2
	44	507		2209	535	243	1158		58	2270	1178	514	1542	136
	%
100	1	2		2	1	2	4		24	1	2	2		13
210	96	8		40	9		2			49	6	10		31
220				17			12			9	20	6
231					5	1					2	1	1
240		27		11	6	5	25		35	11	15	21	1	27
310		59		12	66	80	25		26	10	39	42	27
320	1	3		13	13	11	28		13	19	15	15	44
330				5			3		1		2		26
500												4		27
	F—Ultraperipheral km2
			37	0	54	1125			691	193
	%
100			4			1			1	1
210			47		1	8			39	1
220			15			9			12	2
231
240			23		16	9			14	3
310			7		66	44			13	50
320			3		16	28			19	42
330	1	2			1	2
500

). For example, observing the hub composition of the 14 MRs, the following can be deduced: artificial urban areas are the most representative macro-class in Milano’s hub, as they cover 70% of Milano’s hubs; meanwhile, in Bologna, Firenze, Reggio Calabria, and Venezia, artificial areas do not even cover one-fifth (20) of the inner core area. Venezia and Cagliari differ from other metropoles in having approximately half (56% and 48%, respectively) of their central hub area occupied by bodies of water. Venezia, historically built on a lagoon, consists of fragmented islands interconnected by canals, limiting urban expansion, while Cagliari, with its coastal setting, includes marine zones, lagoons, and salt marshes. Roma and Bologna have one feature in common that differentiates them from others in that they have 41 and 51 percent of their hub area covered by arable land, respectively.

2.2. ES Economic Evaluation

For the economic evaluation of ESs, we referred to our previous work [8], where through a review of the scientific literature and the benefit transfer method, unit economic values of ES supply (EUR/ha) had been associated with land use classes. Here, the previous literature review was further expanded with other sources [42,43,44,45,46,47,48], reducing in many cases the statistical variability previously found. The unit economic coefficients used are shown in

Table A3. Economic values (EUR 2022) per hectare and LULC class used for the quantification of ESs. The economic values are based on our previous literature review [8] and other additional sources [45,46,47,48,49,50]. For each ES, the green color varies from more faded green for the minimum values to more intense green for the maximum values. LULC class codes correspond to urban areas (100), arable land (210), permanent crops (220), pastures (230), heterogeneous agricultural areas (240), forests (310), shrub and/or herbaceous vegetation (320), non- and sparsely vegetated areas (330), and wetland and water bodies (500).

LULC Class	Food	Fodder	Wood	Mushrooms	Climate Regulation	Air Purification	Groundwater Recharge	Water Purification	Erosion Protection	Flood Mitigation
100	0.0	0.0	0.0	0.0	0.4	0.1	0.4	0.1	7.0	3.3
210	3208.7	233.9	0.0	0.0	40.4	0.0	295.5	0.0	144.3	137.6
220	3332.4		0.0	0.0	66.3	21.5	246.3	1.9	459.4	251.5
231	0.0	234.6	0.0	0.0	84.0	0.0	591.0	0.0	301.6	142.5
240	1647.4	101.1	6.6	7.2	63.8	375.5	326.0	2.2	295.4	141.8
310	0.0	51.5	167.5	27.1	200.1	1446.3	1148.0	20.7	495.2	334.0
320	0.0	67.1	11.0	11.4	83.1	773.7	557.0	5.3	520.5	212.1
330	0.0	0.0	0.0	0.0	0.0	0.0	416.4	2.2	0.0	155.9
500	0.0	10.1	0.0	0.0	36.5	10.8	640.5	85.8	31.4	298.7

. To obtain the total economic value (EV) of a specific ecosystem service supply (e.g., for a MR), the unit coefficients are multiplied by the area covered by the associated LULC classes (in the MR), and then, the values obtained are added together (see Equation (1) below).

$$EV \left(ESj,MRz\right)={\sum }_{x=1}^{n}EV\left(ESj,LULCx\right)·Area (LULCx,MRz)$$

(1)

2.3. Calculation of Specialization Indices

The general formula for the SI is as follows [49]

$${\displaystyle \frac{A-B}{[\left(1-A\right)\ast B+\left(1-B\right)\ast A]}}$$

(2)

with

• SI > 0 showing greater specialization of the A term compared to the B term;
• SI < 0 showing lower specialization of term A compared to term B.

The SI takes values between −1 and 1. Values close to 1 indicate an area with a more marked characterization of the phenomenon investigated (A) compared to the reference context (B). This formula was selected due to the following advantages: The numerator captures the absolute difference between the two values, indicating whether A dominates over B or vice versa. The denominator provides a weighted normalization based on both terms, constraining the index within a defined and symmetrical range [−1, +1], preventing extreme values, and ensuring that the results remain interpretable and comparable across different contexts. The method results interpret the data because a ratio close to 1 or −1 indicates a large difference, while a ratio close to 0 indicates that the values are similar.

More specifically, we have applied the SI to investigate the specialization of ES supply by comparing, in different ways, the context of the SNAI areas and MRs. In this work, the SI was applied in four different forms, adjusting the A and B terms of the general Formula (2), as shown in

Table 3. Types of SI used in this work and their mathematical construction.

Type of SI	Index Term	Term Calculation
LULC specialization in SNAI areas of all aggregated MRs	A	${\displaystyle \frac{{\sum }_{z=1}^{14}Area (LULCx,SNAIy,MRz) }{{\sum }_{z=1}^{14}Area (SNAIy,MCz)}}$	(3)
	B	${\displaystyle \frac{{\sum }_{z=1}^{14}Area (LULCx,MRz)}{{\sum }_{z=1}^{14}Area \left(MRz\right)}}$	(4)
ES supply economic value (EV) specialization in SNAI areas of all aggregated MRs	A	${\displaystyle \frac{{\sum }_{z=1}^{14}EV(ESj,SNAIy,MRz)}{{\sum }_{z=1}^{14}{\sum }_{j=1}^{10}EV(ESj,SNAIy,MRz)}}$	(5)
	B	${\displaystyle \frac{{\sum }_{z=1}^{14}EV (ESj,MRz)}{{\sum }_{z=1}^{14}{\sum }_{j=1}^{10}EV(ESj,MRz)}}$	(6)
ES supply economic value (EV) specialization in SNAI areas of each MR	A	${\displaystyle \frac{EV (ESj,SNAIy,MRz)}{{\sum }_{j=1}^{10}EV(ESj,SNAIy,MRz)}}$	(7)
	B	${\displaystyle \frac{EV (ESj,MRz)}{{\sum }_{j=1}^{10}EV(ESj,MRz)}}$	(8)
ES supply economic value (EV) specialization in SNAI areas	A	${\displaystyle \frac{EV (ESj,SNAIy)}{{\sum }_{j=1}^{10}EV \left(SNAIy\right)}}$	(9)
	B	${\displaystyle \frac{EV \left(ESj\right)}{{\sum }_{j=1}^{10}EV\left(ESj\right)}}$	(10)

below.

3. Results

3.1. ES Economic Value in Italian MRs

The application of the benefit transfer method makes it possible to quantify the economic value of the investigated ecosystem services of Italian MRs (

Table 4. Total (MIL EUR) and average (EUR/ha) annual economic value of ES supply in Italian MRs. (BA = Bari; BO = Bologna; CA = Cagliari; CT = Catania; FI = Firenze; GE = Genova; ME = Messina; MI = Milano; NA = Napoli; PA = Palermo; RC = Reggio Calabria; RM = Roma; TO = Torino; VE = Venezia).

MRs	Food	Fodder	Wood	Mushrooms	Climate Regulation	Air Purification	Groundwater Recharge	Water Purification	Erosion Protection	Flood Mitigation	Total (MILEUR/year)	EUR/ha/year
BA	939	36	3	1	23	68	122	1	122	71	1387	3592
BO	671	52	14	3	29	155	177	2	95	65	1264	3416
CA	101	10	6	1	11	76	70	1	41	25	342	2749
CT	643	36	7	2	23	110	142	1	100	63	1129	3161
FI	405	27	28	5	43	265	241	4	133	85	1236	3518
GE	38	10	20	4	27	200	158	3	81	49	590	3232
ME	279	20	17	4	33	229	196	3	131	73	985	3026
MI	293	22	1	0,2	5	10	35	0,2	18	15	400	2540
NA	141	7	2	1	6	32	35	0,4	26	15	266	2272
PA	976	61	9	3	34	162	202	2	152	90	1690	3379
RC	407	19	17	4	33	199	185	3	128	73	1068	3335
RM	801	55	20	4	42	231	246	4	149	95	1647	3074
TO	556	61	34	7	63	420	402	5	218	137	1905	2793
VE	479	34	0,4	0,2	10	14	85	5	32	40	699	2839
Total MRs	6731	452	178	39	382	2171	2298	34	1426	898	14,608	3137
National (Nat.) total	41,577	3172	1403	291	2723	16,142	16,580	234	9578	6118	97,819	3243
MRs/Nat. (%)	16.2%	14.2%	12.7%	13.4%	14.0%	13.5%	13.9%	14.4%	14.9%	14.7%	14.9%	−3.4%

). More specifically, it is possible to see that metropolitan regions provide a total of approximately EUR 14.6 billion per year in environmental benefits. Almost half of this value (EUR 6.731 billion) is derived from agricultural production. This is followed by groundwater recharge (EUR 2.298 billion), air purification (EUR 2.171 billion), and erosion protection (EUR 1.426 billion). At the individual level, it can be seen that the MR that produces the most ESs is Torino, followed by Palermo and Roma. If, on the other hand, we relate the economic ES of the RM to its surface area, Bari has the highest average productivity of ESs (3592 EUR/ha), followed by Firenze (3518 EUR/ha) and Bologna (3416 EUR/ha). Compared to the national total (EUR 97.819 billion), the MRs produce about 15% of the national wealth in environmental goods and services, with an average productivity per hectare of EUR 3137 compared to EUR 3243 for the nation as a whole.

Finally,

Table 5. Total (MIL EUR) and average (EUR/ha) annual ES supply economic value of the SNAI areas of the aggregated MRs. Provisioning services include food, fodder, wood, and mushrooms; regulating services include climate regulation, air purification, groundwater recharge, water purification, erosion protection, and flood mitigation. Colors range from light to dark in proportion to increasing values.

MR SNAI Areas	Surface (ha)	All ES Total Economic Value (MIL EUR/year)	All ES Average Economic Value (EUR/ha/year)	Provisioning ES Average Economic Value (EUR/ha/year)	Regulation ES Average Economic Value (EUR/ha/year)
A	495,216	1153	2329	1370	959
B	238,337	777	3261	2158	1103
C	1,489,525	4750	3189	1800	1389
D	1,183,770	3867	3266	1540	1726
E	1,039,434	3355	3228	1392	1835
F	210,065	706	3361	1215	2146
ALL	4,656,347	14,608	3106	1579	1526

below shows economic values of ES supply (total and average per hectare), divided by SNAI area (aggregated). It can be seen from the table that the inner areas have the highest average economic values per hectare (especially, F—ultraperipheral areas), while the central areas have lower economic values (especially, A—polo areas). It can also be seen that the two types of ES investigated, i.e., provisioning ESs and regulating ESs, demonstrate the opposite behavior, as follows: in central areas, provisioning services prevail; meanwhile, in inland areas, regulating services prevail. These results are mainly determined by land management and, more specifically, by the specific land use and land cover of these areas, as will be explained in the following paragraphs.

3.2. LULC SI for the SNAI Areas of MRs

The Land Use and Land Specialization Index (LULC SI) (

Table 6. LULC SI [−1, +1] of the SNAI areas of Italian MRs. Red indicates the minimum values; green indicates the maximum values. Note: Red indicates negative values; green indicates positive values).

LULC Class Code	LULC Class Description	SNAI Areas
		A—Hubs	B—Intermediate Hubs	C—Belt	D—Intermediate	E—Peripheral	F—Ultraperipheral
100	Urban areas	0.62	0.15	0.14	−0.35	−0.69	−0.85
210	Arable lands	−0.05	0.34	0.16	−0.14	−0.13	−0.25
220	Permanent crops	−0.18	0.12	−0.20	0.18	0.05	0.02
230	Pastures	−0.66	−0.64	0.24	0.21	−0.55	−0.99
240	Heterogeneous agricultural areas	−0.01	0.03	0.05	0.05	−0.10	−0.20
310	Forests	−0.43	−0.31	−0.11	0.12	0.14	0.31
320	Shrub and/or herbaceous vegetation	−0.40	−0.75	−0.25	0.02	0.31	0.44
330	Sparsely vegetated areas	−0.50	−0.61	−0.40	−0.07	0.45	−0.28
500	Wetland and water bodies	0.60	0.12	−0.19	−0.04	−0.46	−0.84

below) allows us to identify the specialization in terms of LULC. The results show that the central areas are those in which urbanization dominates, peaking in the poles (A), while inland areas have a predominantly silvo-pastoral character, especially E and F. The pole areas (A) also show a strong specialization in water-related land cover, but it is important to point out that the quantity and quality of water flow depend on factors related to the other SNAI areas further upstream. The agricultural landscape is widespread, mainly in the intermediate areas (B, C, and D), with interesting alternations of specializations between arable land and permanent crops and a tendency towards landscape fragmentation (see class 240). In general, inland areas (D, E, and F) show a substantially reversed specialization compared to central areas (A, B, and C), with a predominant characterization of pasture and forest areas and de-specialization in terms of agricultural (except D for permanent crops and grassland) and urban areas. The specialization highlighted so far influences the type and quantity of ESs provided, as will be explored in the next section.

3.3. Specialization of ES Supply in the SNAI Areas of Aggregated MRs

The specialization index (SI) calculation first returns a clear dichotomy between the types of ESs offered. Comparing the SIs of the SNAI areas in terms of provisioning and regulation Ess (see Figure 3), it is evident that the central areas (A, B, and C) are specialized in provisioning services while the inner areas (D, E, and F) are devoted to regulation services. In the case of regulation services in inland areas, the SI also returns a progressive specialization, as the distance from central areas increases. By deepening the SI for individual ESs (see

Table 7. ES supply SI [−1, +1] of the SNAI areas of aggregated metropolitan regions (Note: Red indicates negative values; green indicates positive values).

SNAI Area	Food	Fodder	Wood	Mushrooms	Climate Regulation	Air Purification	Groundwater Recharge	Water Purification	Erosion Protection	Flood Mitigation
A	0.16	0.08	−0.23	−0.18	−0.09	−0.20	−0.06	0.23	−0.10	−0.02
B	0.30	0.12	−0.28	−0.29	−0.16	−0.34	−0.17	−0.19	−0.17	−0.11
C	0.12	0.02	−0.09	−0.09	−0.05	−0.11	−0.07	−0.11	−0.04	−0.04
D	−0.07	−0.04	0.06	0.05	0.03	0.06	0.03	0.03	0.03	0.02
E	−0.16	−0.03	0.09	0.09	0.06	0.12	0.08	0.02	0.06	0.04
F	−0.30	−0.08	0.20	0.19	0.12	0.23	0.13	0.08	0.11	0.07

below), the specialization of SNAI areas changes depending on the single service examined, although a distinct pattern can be discerned quite clearly between the food and fodder production ESs and the other ESs. For example, food production is significantly more prevalent in central areas (SI_A = 0.16; SI_B = 0.30; SI_C = 0.12) than in inner areas (SI_D = −0.07; SI_E = −0.16; SI_F = −0.30). This is the same for forage production. The other services analyzed show common and opposite patterns compared to the first two ESs analyzed, with a predominant specialization of inland areas and a de-specialization of central areas in the provision of these ESs. This is mainly due to the predominant LULC in the different SNAI areas. In fact, the first two ESs are primarily generated by agricultural land uses (predominant in the central areas), while the second is generated by forest land uses (predominant in the internal areas), as can be seen from the SI of the LULC reported in Section 3.1. The only exception in this regard appears to be that of the water purification ES, which shows high specialization (0.23) in the polo areas (A). Again, the high specialization of polo areas in water purification can be attributed to the fact that these areas are the ones where most of the land uses related to river reticulation and wetlands (macro-class 500) are present. The literature review returned higher values per hectare of this ES for macro-class 500. Thus, the trade-off between food and fodder provisioning services (mainly related to agricultural land use) and other ESs (mainly related to forests) is evident. The analysis presented also shows how ESs follow similar patterns that can be traced back to the concept of ES bundles. The concept of a bundle of ESs refers to the co-occurrence of different ESs in a given geographic area or landscape, often linked together by ecological, management, or social interactions. In other words, a bundle represents a set of services that are simultaneously provided by the same ecosystem, often influencing each other [49]. In the case of our results, and, more specifically, by comparing LULC specialization and ES supply specialization, clusters can be seen between a particular land use and the ES generated. In particular, (i) agricultural areas (specifically arable land) and the ESs of providing food and fodder mainly concentrated in central areas (A, B, and C); (ii) forested areas and the ESs of wood and mushroom supply, climate regulation, air purification, groundwater recharge, erosion, and flooding mainly concentrated in inland areas; and (iii) aquatic environments and water purification ESs predominantly concentrated in the central area of the poles (A).

Finally, Figure 4 is intended to represent the sum value of the ES supply SI in each SNAI area using an ideal single metropolitan form. This representation serves to comment on the aggregated value of the SI at a general level, considering all the cases studied. As can be observed, the highest rate of specialization is attributed to hubs (A) in the water purification ES. Conversely, the lowest rate of specialization can be found in the air quality regulation corresponding to intermediate hubs (B). Generally, central areas are characterized by negative ISP values in timber and mushroom supply, which is the same for global climate, air quality, and erosion regulation. Conversely, internal areas show positive ISP values in the same services. This is particularly evident in air quality regulation ESs.

3.4. ES Supply Specialization of Individual MRs

In

Table 8. ES supply SI [−1, +1] of Italian MRs grouped by C—central (A, B, and C)—and I—internal (D, E, and F)—SNAI areas. Red indicates the minimum values; green indicates the maximum values. BA = Bari; BO = Bologna; CA = Cagliari; CT = Catania; FI = Firenze; GE = Genova; ME = Messina; MI = Milano; NA = Napoli; PA = Palermo; RC = Reggio Calabria; RM = Roma; TO = Torino; VE = Venezia.

MRs	SNAI	Food	Fodder	Wood	Mushrooms	Climate Regulation	Air Purification	Groundwater Recharge	Water Purification	Erosion Protection	Flood Mitigation
BA	C	0.00	−0.13	0.01	0.01	0.02	0.00	−0.02	0.02	0.03	0.02
	I	0.00	0.22	−0.03	−0.03	−0.04	−0.01	0.04	−0.06	−0.09	−0.04
BO	C	0.27	0.09	−0.44	−0.35	−0.18	−0.37	−0.17	−0.29	−0.13	−0.09
	I	−0.62	−0.29	0.43	0.38	0.27	0.46	0.29	0.35	0.23	0.18
CA	C	0.03	0.01	−0.06	−0.03	−0.08	−0.05	0.17	−0.45	−0.07	−0.10
	I	−0.07	−0.03	0.07	0.04	−0.03	0.03	0.18	−0.62	−0.05	−0.12
CT	C	−0.02	−0.07	−0.10	0.03	0.01	0.02	0.01	0.10	0.03	0.03
	I	0.00	0.01	0.02	−0.01	0.00	0.00	0.00	−0.02	−0.01	−0.01
FI	C	0.10	0.00	−0.05	−0.05	−0.03	−0.07	−0.05	−0.04	−0.02	−0.02
	I	−0.24	−0.01	0.09	0.09	0.06	0.12	0.09	0.07	0.05	0.04
GE	C	0.15	0.03	−0.05	−0.02	−0.02	−0.02	−0.02	−0.04	0.02	−0.01
	I	−0.23	−0.04	0.05	0.02	0.02	0.03	0.03	0.04	−0.02	0.01
ME	C	0.16	−0.12	−0.16	−0.10	−0.06	−0.12	−0.06	−0.09	0.00	0.00
	I	−0.02	0.01	0.02	0.01	0.01	0.01	0.01	0.01	0.00	0.00
MI	C	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
	I	/	/	/	/	/	/	/	/	/	/
NA	C	0.06	0.00	−0.08	−0.06	−0.03	−0.07	−0.04	−0.05	−0.02	−0.01
	I	−0.54	−0.04	0.41	0.35	0.22	0.42	0.27	0.31	0.14	0.11
PA	C	−0.06	−0.10	0.01	0.05	0.03	0.06	0.02	0.08	0.07	0.04
	I	0.01	0.02	0.00	−0.01	−0.01	−0.01	0.00	−0.02	−0.02	−0.01
RC	C	0.10	0.06	−0.12	−0.06	−0.05	−0.07	−0.06	−0.08	−0.02	−0.03
	I	−0.02	−0.01	0.02	0.01	0.01	0.01	0.01	0.02	0.00	0.01
RM	C	0.17	0.08	−0.21	−0.16	−0.09	−0.18	−0.10	−0.23	−0.07	−0.06
	I	−0.10	−0.05	0.09	0.07	0.05	0.09	0.05	0.10	0.04	0.03
TO	C	0.49	0.16	−0.37	−0.34	−0.21	−0.42	−0.26	−0.29	−0.24	−0.18
	I	−0.72	−0.18	0.14	0.16	0.11	0.23	0.18	0.15	0.16	0.13
VE	C	−0.14	−0.05	−0.04	−0.03	0.06	−0.01	0.14	0.13	−0.03	0.12
	I	−0.05	−0.02	0.11	0.16	0.03	0.16	0.01	−0.03	0.05	0.01

below, the SI values of each MR were aggregated into two groups, i.e., central areas (A, B, and C) and inner areas (D, E, and F). For a disaggregated view of the SI of single SNAI areas, refer to

Table A4. ES supply SI [−1, +1] of SNAI areas in Italian MRs. Red indicates the minimum values; green indicates the maximum values (BA = Bari; BO = Bologna; CA = Cagliari; CT = Catania; FI = Firenze; GE = Genova; ME = Messina; MI = Milano; NA = Napoli; PA = Palermo; RC = Reggio Calabria; RM = Roma; TO = Torino; VE = Venezia).

	Food	Fodder	Wood	Mushrooms	Climate Regulation	Air Purification	Groundwater Recharge	Water Purification	Erosion Control	Flood Mitigation
SNAI	BA
A	0.02	−0.27	−0.09	−0.05	0.02	−0.07	−0.04	0.01	0.07	0.04
B	−0.03	0.18	0.26	0.13	0.00	0.10	0.05	0.05	−0.10	−0.05
C	0.01	−0.32	−0.14	−0.04	0.02	−0.03	−0.05	0.01	0.08	0.04
D	−0.01	0.21	−0.01	−0.01	−0.03	0.01	0.05	−0.04	−0.08	−0.04
E	0.26	0.39	−0.84	−0.85	−0.24	−0.85	−0.09	−0.77	−0.42	−0.20
	BO
A	0.35	0.10	−0.73	−0.45	−0.24	−0.49	−0.24	−0.55	−0.15	−0.13
B	0.47	0.15	−0.91	−0.81	−0.34	−0.82	−0.29	−0.48	−0.30	−0.17
C	0.20	0.06	−0.30	−0.23	−0.13	−0.25	−0.12	−0.20	−0.09	−0.07
D	−0.60	−0.28	0.41	0.38	0.26	0.45	0.28	0.33	0.24	0.17
E	−0.63	−0.30	0.45	0.39	0.28	0.47	0.30	0.37	0.23	0.18
	CA
A	−0.21	−0.13	−0.87	−0.73	−0.15	−0.73	0.58	0.60	−0.47	0.40
C	0.04	0.01	−0.05	−0.02	−0.08	−0.03	0.15	−0.59	−0.06	−0.13
D	−0.07	−0.03	0.07	0.04	−0.03	0.03	0.18	−0.62	−0.05	−0.12
	CT
A	0.33	0.08	−0.84	−0.62	−0.20	−0.64	−0.20	0.12	−0.16	−0.08
C	−0.12	−0.12	0.02	0.14	0.06	0.14	0.07	0.10	0.08	0.06
D	−0.07	−0.11	0.02	0.09	0.04	0.09	0.03	0.06	0.07	0.03
E	0.02	0.03	0.01	−0.03	−0.01	−0.03	−0.01	−0.04	−0.02	−0.01
F	−0.12	0.13	0.30	0.16	0.07	0.16	0.09	0.12	−0.08	−0.02
	FI
A	0.14	0.09	−0.07	−0.07	−0.05	−0.09	−0.06	−0.06	−0.06	−0.04
B	0.03	−0.06	−0.01	−0.01	−0.01	−0.02	−0.02	−0.02	0.00	0.00
C	0.11	0.00	−0.06	−0.06	−0.04	−0.07	−0.05	−0.03	−0.02	−0.02
D	0.01	−0.06	0.00	0.00	0.00	0.00	−0.01	−0.02	0.00	0.00
E	−0.60	0.05	0.18	0.18	0.13	0.24	0.19	0.15	0.09	0.08
	GE
A	0.03	0.08	−0.14	−0.03	−0.04	−0.02	−0.03	−0.08	0.09	0.01
B	0.72	0.09	−0.24	−0.17	−0.14	−0.25	−0.20	−0.21	−0.09	−0.10
C	0.14	0.02	−0.03	−0.02	−0.01	−0.02	−0.02	−0.02	0.00	−0.01
D	−0.26	−0.04	0.05	0.02	0.02	0.03	0.03	0.04	−0.02	0.01
E	−0.28	−0.03	0.06	0.03	0.03	0.03	0.03	0.04	−0.02	0.01
F	0.16	0.02	−0.01	−0.01	−0.01	−0.02	−0.01	−0.01	−0.02	−0.01
	ME
A	0.02	−0.01	−0.12	−0.04	−0.04	−0.04	0.01	−0.03	0.01	0.03
C	0.28	−0.24	−0.20	−0.17	−0.09	−0.19	−0.14	−0.15	−0.01	−0.02
D	0.19	0.00	−0.21	−0.10	−0.07	−0.12	−0.09	−0.13	−0.01	−0.03
E	0.04	0.01	−0.06	−0.02	−0.02	−0.02	−0.02	−0.04	0.01	−0.01
F	−0.19	0.02	0.14	0.08	0.06	0.09	0.07	0.10	0.00	0.02
	MI
A	0.02	0.01	−0.10	−0.10	−0.01	−0.10	−0.02	0.01	0.00	0.01
B	0.11	0.04	−0.96	−0.81	−0.09	−0.82	−0.07	−0.88	−0.05	−0.04
C	−0.01	0.00	0.04	0.04	0.01	0.04	0.01	0.03	0.00	0.00
	NA
A	0.01	−0.02	−0.10	−0.01	0.00	−0.01	−0.02	−0.03	0.03	0.01
B	0.10	0.12	−0.08	−0.11	−0.06	−0.13	−0.04	−0.14	−0.09	−0.05
C	0.03	−0.11	−0.07	−0.03	−0.01	−0.03	−0.03	0.01	0.03	0.01
D	−0.52	−0.04	0.39	0.34	0.20	0.41	0.27	0.29	0.14	0.11
E	−0.56	−0.04	0.43	0.36	0.23	0.44	0.28	0.32	0.15	0.11
	PA
A	−0.63	−0.37	0.49	0.44	0.28	0.51	0.30	0.42	0.32	0.22
C	−0.01	−0.08	−0.06	0.00	0.01	0.01	−0.01	0.04	0.05	0.03
D	0.10	−0.07	−0.37	−0.16	−0.06	−0.16	−0.09	−0.10	0.01	−0.01
E	0.00	0.06	0.05	0.01	−0.01	0.00	0.01	−0.02	−0.04	−0.02
F	−0.07	0.00	0.16	0.09	0.04	0.09	0.05	0.07	0.01	0.01
	RC
A	−0.05	0.07	−0.02	0.04	0.01	0.05	0.01	−0.01	0.01	−0.01
C	0.19	0.06	−0.21	−0.13	−0.09	−0.15	−0.10	−0.14	−0.04	−0.04
D	0.18	−0.03	−0.14	−0.12	−0.07	−0.15	−0.10	−0.11	−0.03	−0.03
E	−0.13	0.00	0.11	0.07	0.05	0.09	0.07	0.08	0.01	0.02
F	−0.88	0.02	0.29	0.29	0.19	0.39	0.28	0.28	0.16	0.13
	RM
A	0.31	0.17	−0.43	−0.35	−0.20	−0.39	−0.19	−0.38	−0.18	−0.13
C	−0.09	−0.13	0.05	0.06	0.05	0.07	0.03	−0.04	0.09	0.04
D	−0.03	−0.03	0.03	0.02	0.02	0.03	0.01	0.04	0.01	0.01
E	−0.46	−0.19	0.31	0.27	0.18	0.33	0.22	0.31	0.15	0.13
	TO
A	0.54	0.16	−0.42	−0.38	−0.24	−0.46	−0.29	−0.14	−0.27	−0.18
B	0.61	0.22	−0.55	−0.49	−0.29	−0.58	−0.35	−0.57	−0.30	−0.24
C	0.33	0.10	−0.13	−0.16	−0.10	−0.21	−0.16	−0.15	−0.14	−0.12
D	−0.47	−0.15	0.16	0.15	0.11	0.20	0.12	0.13	0.11	0.08
E	−0.96	−0.21	0.11	0.17	0.12	0.27	0.24	0.16	0.21	0.19
	VE
A	−0.48	−0.16	0.05	0.06	0.21	0.12	0.47	0.64	−0.03	0.42
B	−0.06	0.00	−0.15	−0.13	0.02	−0.12	0.10	0.22	−0.06	0.09
C	0.11	0.03	−0.01	−0.01	−0.05	−0.02	−0.15	−0.48	0.01	−0.14
D	0.08	0.02	−0.02	−0.09	−0.04	−0.10	−0.09	−0.23	−0.01	−0.08
E	−0.17	−0.06	0.24	0.42	0.10	0.42	0.11	0.18	0.12	0.10

. The results show how the individual specialization of Italian MRs presents specific characteristics for each metropolitan region. First of all, it should be noted that not all the territories of the MRs belong to all SNAI areas. The case of Milan, for instance, is emblematic, as it has no internal areas (D, E, and F) but only central areas (A, B, and C), and among all the MRs, only Genoa has all the different SNAI areas in its territory. This does not necessarily seem to depend on the size of the MRs, but rather on the territorial conformation and the travel times within the cities. Milan, for instance, is not the smallest metropolitan region in terms of surface area, yet it is the only one that lacks inland areas. This is largely due to its dense service infrastructure and extensive transport network, enhancing accessibility across the region. Its flat terrain has been one of the key factors enabling the development of this well-connected infrastructure system. This particular situation brings the aggregate PSI for central areas (Table 8) of Milan to zero; there is no real relative specialization of the central areas, but rather a complete absence of internal areas (which do not present a value in Table 8). Looking more closely at both the prevailing land use (Table A1) and the values of the ESs generated (Table 4), we can see that the Milan area is mainly composed of impermeable and agricultural surfaces, making it one of the MRs with the lowest supply of regulating ESs.

The trade-off already highlighted between specialization in the agricultural ES of food and fodder production and the other ESs, seems to be also confirmed at the level of each single MR, but the expertise of the central areas in food and fodder production, although representing the general trend of the MR, is not true for all the MRs. For example, Venice and Palermo show an average under-specialization in food and fodder production. The most evident trade-offs between the specialization in food and fodder and the other ESs (calculated as the distance between the average SI of the food and fodder production ESs and the average SI of the other ESs) are present in the central areas of Turin, predominantly vocated in food and fodder production, and in the inner areas of Bologna, with an important vocation for the other ESs (especially air purification and wood supply). In general, these two cities have the most pronounced SI values (both negative and positive), followed by Naples and Rome. The other cities show less pronounced SI values, but this also depends on the fact that in the case of Table 8 shown here, the values are grouped by a weighted average of the central and inner areas. The disaggregated results (see Table A4) show more evident specializations per single SNAI area. For example, area E in Bari clearly shows the aforementioned trade-off between food and fodder production and the other ESs, area B in Milan shows extreme under-specialization for some forest-related ESs, and area A in Catania is one of the areas with the most negative SI values for the supply of ESs other than food and fodder. As mentioned above, the SI values are strictly dependent on the specific LULC configuration of individual MRs. These results can therefore provide a basis for reflecting on the territorial policies adopted so far in individual MRs, and they can be the starting point for new, more conscious policies.

3.5. Comparison of the ES Supply Specialization of SNAI Areas in MRs and SNAI Areas Nationwide

Comparing the ES supply SI values of the MRs’ SNAIs (Table 7) and that of the SNAI at the national level (

Table 9. ES supply SI [−1, +1] of national SNAI areas (Note: Red indicates the minimum values; green indicates the maximum values).

SNAI	Food	Fodder	Wood	Mushrooms	Climate Regulation	Air Purification	Groundwater Recharge	Water Purification	Erosion Protection	Flood Mitigation
A	0.31	0.08	−0.34	−0.28	−0.16	−0.31	−0.18	−0.18	−0.12	−0.10
B	0.25	0.06	−0.19	−0.20	−0.11	−0.24	−0.13	−0.07	−0.12	−0.07
C	0.20	0.06	−0.15	−0.15	−0.09	−0.18	−0.11	−0.13	−0.09	−0.06
D	0.16	0.03	−0.17	−0.17	−0.11	−0.20	−0.13	−0.15	−0.11	−0.09
E	−0.06	−0.03	0.06	0.04	0.03	0.05	0.03	0.05	0.02	0.02
F	−0.40	−0.08	0.14	0.18	0.11	0.24	0.16	0.12	0.16	0.10

and

Table 10. Difference between the ES supply specialization index of SNAI areas of MRs (aggregated) and national SNAI areas (Note: Purple indicates the minimum values; light blue indicates the maximum values).

SNAI	Food	Fodder	Wood	Mushrooms	Climate Regulation	Air Purification	Groundwater Recharge	Water Purification	Erosion Protection	Flood Mitigation
A	−0.14	0.00	0.11	0.10	0.07	0.11	0.12	0.40	0.02	0.08
B	0.06	0.06	−0.10	−0.09	−0.04	−0.10	−0.04	−0.12	−0.05	−0.03
C	−0.08	−0.03	0.06	0.06	0.04	0.07	0.04	0.02	0.04	0.02
D	−0.23	−0.08	0.24	0.22	0.14	0.26	0.16	0.19	0.14	0.11
E	−0.10	0.00	0.04	0.05	0.03	0.07	0.05	−0.03	0.04	0.03
F	0.09	0.00	0.06	0.01	0.00	−0.01	−0.03	−0.04	−0.04	−0.04

below), there is generally an overlap in the patterns found, both in the pattern in which food and fodder production predominates in the central areas, compared to the inland areas, and the trade-off between specialization in food and fodder supply and the other ESs. There are, however, some differences, as follows:

• The SNAI D area of the MRs is less specialized in food and fodder supply than the national SNAI, but it is more specialized in other ESs. Thus, it appears that, nationally, the SNAI D areas are more agricultural than those of the MRs.
• The specialization values of the national SNAI for the food supply are generally more pronounced than the SNAI of the MRs. The SNAI areas of the MRs are thus less suited for food production, compared to the national territory. This is udnerstandably due to the fact that MRs probably have more areas devoted to (anthropogenic) services and infrastructure than the national average, which is more devoted to agriculture.
• The ES of water purification does not have the same important specialization in the national A pole areas compared to the Apole areas of the MRs.

4. Discussion

The expansion of MRs and the related socio-economic and environmental processes make it necessary to equip territories with strategic and planning capabilities to adapt decision making and operational processes to changing contexts. In Italy, Law 56/2014 assigns several functions to MRs, including the drafting of the Metropolitan Strategic Plan, which aims to improve governance. Providing a cognitive framework of ESs and including them in the Strategic Plan can be useful in ensuring the maintenance and supply of flows of goods and services essential for human well-being, at both local and global scales [8,50]. In this work, a combination of methodologies (benefit transfer, specialization indices, and mapping) was adopted with the objectives of quantifying the provision of ESs in metropolitan areas and exploring the specialization of metropolitan SNAI regions in the provision of ESs. The results show that, in general, ES specialization in metropolitan SNAI regions is largely dependent on the prevailing land use and land cover. Indeed, land use and land cover generate specific ES bundles [51]. Knowledge of the specific ES bundles generated by land use types is of paramount importance, as it allows us to predict and potentially plan the type of ES needed. The results also show how the core areas are characterized by mainly artificial and agricultural land uses, while the inner areas, further away from the poles, are more natural, with the predominance of forests as one moves away from the core. This configuration is reflected in the supply of ESs, with provisioning services (especially food and fodder) presenting high values in central areas and low values in inland areas, and vice versa, and regulating services presenting lower values in central areas and higher values in inland areas. This pattern is recurring in most MRs, with a clear trade-off between the two types of services. Our results show that polo areas use land at multiple rates compared to other areas. This is due to the greater land consumption caused by urban specialization, which raises the obvious problems of unsustainability in the current model (Table 6). Some critical issues we highlight concern hydrogeological instability, as well as air purification, as indicated by the lower values of the specialization of regulation ESs in the core areas. In urban areas, population growth generates a greater demand for ESs, and urbanization processes are responsible for altering their status. This creates a clear dependence of central areas on peripheral areas for the provision of various ESs, particularly regulation Ess (Table 8). Demographic changes affect both supply and demand to some extent. On the one hand, the growth in population concentration in urban areas has an impact on the greater demand for ecosystem services (food, water, but also climate regulation, and regulation services, in general), on the other hand, the lower human presence in rural areas favors renaturalization processes that induce changes in the supply of services (less production and more adjustment). This dependence manifests itself differently in MRs due to the different socio-economic dynamics of the areas (Table 8). As Figure 4 shows, a general model emerges in the patterns of specialization between urban and rural areas for the provision of ecosystem services. Nevertheless, it is also true that each metropolitan area evolves according to unique and specific trajectories that depend on territorial, economic, and demographic factors. This double evolutionary track requires careful multiscalar governance that includes national and local policies. The National Strategy for inland areas is—at least theoretically—a good example; however, it is a national cohesion policy that can be altered through local projects adapted to specific conditions. According to the Economics of Ecosystems and Biodiversity (TEEB) [52], agricultural and agri-food production is closely linked to ESs. Environmental degradation can negatively affect ESs and, consequently, food security. In our study, food production is more prevalent in central areas compared to inland areas. In these areas, socio-economic factors have influenced the depopulation of these territories, resulting in the abandonment of agricultural practices [19]. According to Von Thünen’s theorem then, the core areas (SNAI A and B) show a strong specialization in supply services (e.g., food and fodder) related to intensive agricultural land use. This follows the logic of Von Thünen, where intensive agricultural production was located close to markets to minimize transport costs. In contrast, the peripheral and ultra-peripheral areas (SNAI E and F) are characterized by a greater supply of regulation services, such as air purification and flood mitigation. This is also consistent with the predominant presence of forests and natural soils, which recall the more extensive zoning of the Von Thünen model. In general, the SNAI A pole areas can therefore be seen as the city, according to Von Thünen’s model, and these areas require the most ESs, being the most densely populated and urbanized. In this context, it becomes important to link MRs with inner areas to enhance efficiency in the use (demand) and supply of goods and services. Is the ES supply specialization found functional? At a general level, the utility of different ESs can be defined by the relative demand for specific ESs [53,54]. Assuming there is a generally higher demand for ESs in hub areas (particularly ESs with local benefits), the prevailing specialization of core areas in the production of food and fodder and their under-specialization in other ESs (particularly regulation ESs) such as air purification, flood mitigation, and water purification would suggest a potentially unmet need for some of the most useful ESs in urban areas. This unmet need can potentially translate into possible health damage and reduced resilience to extreme events with the associated economic costs. It is also true that to state what has just been argued more robustly, it would be necessary to compare the specialization of ES supply with the specialization of ES demand [55]. Moreover, in this work, the scale of the investigation focuses on SNAI areas, whereas a potential mismatch between ES supply and demand could be more effectively assessed at lower scales. Furthermore, the aggregated results (all MRs) of the SI the ESs offer, while certainly relevant, do not show extreme values of specialization or sub-specialization, considering that the index can vary between −1 and +1 and that the results range between −0.3 and +0.3. The range of the SI index, however, becomes wider when analyzing individual MRs with specific individual characteristics. Our work, compared to other studies in the literature [34,56,57], applied a SI based on the economic value generated by the flow of benefits. Furthermore, we tested this index by investigating the national context and focusing on MRs and SNAI areas. The results, interpretable as a measure of territorial performance, represent a tool supporting the comparison between different planning scenarios, both in terms of environmental protection issues and socio-economic development strategies. An example of an action that can be taken is to incorporate the topic of natural capital and ESs into spatial planning policies, in line with the European Commission’s guidance document on integrating ecosystems and their services into decision making (SWD (2019) 305 final). It is also important to support urban planning policies to enhance the provision of urban ESs and make cities more sustainable. Investing in nature-based solutions (NBS), such as urban forestation, would support air purification and heat island mitigation, which are essential ESs for the community. Finally, it is also desirable to update spatial planning and programming tools with an ecosystem approach and to ensure interoperability between the various tools. To address the limitations of our research, we propose investigating the effects of socio-economic dynamics on the supply of ES bundles by applying the SI on the scale of time. Moreover, to further support the planning process, another challenge to our research will be to investigate the specialization of the ES supply at the level of protected natural areas, where most of the benefit flows that satisfy urban demand originate. This approach can support the implementation of the System of Environmental-Economic Accounting (SEEA) [58] and the achievement of the Sustainable Development Goals (SDGs) outlined in the 2030 Agenda. In conclusion, we think it is also important to understand these dynamics in MRs in the light of the legislation that assigns planning and programming powers to these areas. This raises the issue of specific governance for ESs, including through economic remuneration for those who conserve and manage natural capital, ensuring the generation of ESs. This is even more important in the current climate crisis that threatens the resilience of both natural capital and the societies that benefit from it.

5. Conclusions

The results reported in this study highlight how the social and economic dynamics that have led, from both an administrative and political point of view, to the identification of specific territorial typologies—in Italy, the SNAI areas, but similar characterizations are also present at a European and global level—that can be read in terms of specialization on ES supply and their aggregates. Through the application of the SI, we offer a key to understanding the spatial differentiation of providing ESs regarding land use. Understanding this relationship between central and peripheral areas and between urban and rural areas is fundamental for territorial planning policies that look both at the management of ecosystems that more extensively provide ESs, and at social and economic policies aimed at maintaining or increasing the well-being of the population that is increasingly concentrated in urban areas across the planet. To better understand these dynamics, the study focuses on MRs in light of the regulations that attribute planning and programming powers to these areas. In this sense, the different (inverse) specializations among ES should lead to specific governance [59]. The results of our research therefore provide a tool for mapping data and identifying areas that require timely intervention for better planning and management. In addition, our economic estimation of ESs highlights areas of surplus and deficit and identifies sustainable limits of natural resource consumption [19]. These results depend largely on the specific land use and coverage of these areas. The data show that there is a functional specialization of MRs that can be interpreted in light of Von Thunen’s theorem, as follows: the most peripheral areas are specialized in silvo-pastoral systems; the intermediate areas in agricultural systems; and the central and urban areas, in addition to the obvious settlement function, in water ecosystems but also agricultural ones. Therefore, ecosystem services are more prevalent in the former, while more accentuated production is seen in the latter. Consequently, this phenomenon often presents a ’nuanced’ specialization, i.e., not particularly accentuated, especially when compared to the national average. This phenomenon could be due to the effect of sprawling, which in territories not covered by MRs could be more attenuated. It should also be noted that there are significant differences between different MRs, which may be due to both their geographical conformation (e.g., altimetry or the presence of river basins) but also due to different economic models and historically consolidated urban–rural networks. In any case, the analyses carried out also raise the question of urban–rural relations in terms of demand for services, concentrated in urban areas, and supply, in which the more peripheral areas are specialized. The question therefore arises as to whether and how the implementation of innovative governance models, based on the remuneration of ecosystem services (e.g., Payments for Ecosystem Services, PES), can be a useful tool not only for maintaining the well-being of the population but also for preserving the specificities of the territory, passing them on to future generations and countering the phenomenon of depopulation. In other words, these models could contribute to ensuring the quality and attractiveness of life in rural contexts [60]. A concrete example of actionable governance measures could be the implementation of a PES system for the ecosystem service of air purification. For example, an annual payment could be established, whereby metropolitan areas (buyers) pay rural areas (suppliers) based on the amount of air pollution purified or reduced. This payment could be expressed as a sum per ton of CO₂ absorbed or based on the reduction in specific pollutants (e.g., PM₁₀) through agricultural or forestry practices in rural areas. Finally, it should be considered that ES supply, besides depending on local territorial characteristics, is strongly correlated with changes in European agricultural policy and how this is, in turn, linked to global geopolitical and environmental phenomena. These relationships bring with them an additional challenge in terms of governance.