Little Ice Age and contemporary glacier extent in the Western and South-Western Piedmont Alps (North-Western Italy)

Abstract

Glaciers are widely recognized as the best terrestrial indicators of climate change. Nevertheless, changes that have occurred since the end of the Little Ice Age (LIA) are often poorly understood. In Italy, glacier front variations in the last 100 years have been well documented, thanks to the efforts of the Italian Glaciological Committee (Comitato Glaciologico Italiano, CGI), which coordinates annual surveys on all glaciers, since 1927, in order to measure their front positions. However, although front variations are the easiest measurements to take in the field, they only partially describe changes that occur on the entire glacial mass. Moreover, the last national glacier inventory dates back to 1957–1958, which was promoted by the CGI on the occasion of the International Geophysical Year. Authors of the present work herein provide an updated inventory of present-day (2006) glaciers in the Western and South-Western Piedmont Alps (situated in the provinces of Turin and Cuneo). Along with the present-day glacial outlines, the map also includes a reconstruction of the extent of the glaciers during the Little Ice Age, inferred from morphological features such as LIA and post-LIA moraine ridges as well as the aid of historical archive data. Some basic glacier morphometric parameters (surface, length, width, mean slope, highest and lowest altitude, exposure) were computed for the two specified time periods (LIA and 2006). This study was carried out under the framework of the project ‘GlaRiskAlp – Glacial Risks in Western Alps’, project n.° 56 under the 2007–2013 ALCoTra programs.

Keywords:

• Glaciers
• Little Ice Age
• Western Italian Alps
• Glacial retreat

1. Introduction

In the European Alps, glacier extent has dramatically changed since the end of the Little Ice Age (LIA) (around 1865; Painter et al., 2013), with an overall decrease in ice coverage of about 50% (Zemp, Paul, Hoelzle, & Haeberli, 2008). Due to the specific physiographic and geographic settings, glacier shrinkage has been as much as 80% in the Western and South-Western Piedmont Alps, which are the focus of the current study (Fioraso, Bertotto, Lucchesi, Nigrelli, & Chiarle, 2013). The change in glacier extent is one of the most direct and easily recognizable terrestrial indicators of climate change (Houghton et al., 2001). At the same time, the distribution and extent of present-day glaciers are relevant to a number of economic activities, such as tourism (Orlove, 2009), freshwater resources and their management (in particular hydropower production; Kaser, Großhauser, & Marzeion, 2010; Vergara et al., 2007), along with agriculture (Bury et al., 2011). The location, extent and characteristics of recently deglaciated areas also have important consequences on catchment sediment budgets (Koppes & Hallet, 2002; Stott & Mount, 2007), as well as on the initiation of instability processes such as debris and mud flows, which may also pose hazards to areas located much further downstream (Chiarle, Geertsema, Mortara, & Clague, 2011; Holm, Bovis, & Jakob, 2004). In spite of these important implications, accurate maps of current and past glacier outlines are only available for a few areas of the Italian Alps (Bonardi et al., 2012), as the latest, nationwide and uniform cartographic data dates back to 1957–1958 (Comitato Glaciologico Italiano – Consiglio Nazionale delle Ricerche, 1961). In recent years, the GLIMS (Global Land Ice Measurements from Space) project has attempted to overcome the lack or inconsistency of worldwide data on glacier outlines by using images derived from optical satellite instruments, such as ASTER (Raup et al., 2006). However, many Italian glaciers, such as those included in the Western and South-Western Piedmont Alps, cannot easily be detected on satellite images due to their size (58% of glaciers have an area less than 0.1 km²) and because they are often extensively covered by debris.

This contribution aims to partially fill this gap by providing an updated inventory of present-day glaciers of the Western and South-Western Piedmont Alps, and includes a reconstruction of the glacier extent at the peak of the LIA (around 1850).

The investigations were carried out under the framework of the GlaRiskAlp (Glacial Risks in Western Alps) project n.° 56 under the 2007–2013 ALCoTra (Alpi Latine Cooperazione Transfrontiera, Franco-Italian Transborder Alpine Cooperation) program (www.glariskalp.eu), as part of the activities of ‘Action 1A’, devoted to the localization and quantification of glaciated and post-LIA deglaciated terrains as potential sources of geohazards.

2. Geomorphological setting

The Western and South-Western Piedmont Alps are located at the South-Western end of the European Alps, which are part of the Alpine-Himalayan belt extending from South-Western Europe to Asia. This belt developed from the Cretaceous onwards by subduction of a Mesozoic ocean, and the subsequent collision between European and Adriatic continental margins (Dal Piaz, Bistacchi, & Massironi, 2003). The European Alps, about 1300 km long, are characterized by an arched shape, and are generally divided into Western, Central, and Eastern Alps, based on their orography. The Western Alps comprise peaks of different altitude, stretching from the Argentera Massif (Cima Sud dell'Argentera, 3297 m), which is just a few kilometers from the Mediterranean sea, to the Mont Blanc (4810 m). Italian glaciers are located on the Southern flank of the Alps, mainly in the Western and Central Alps, and the Italian Alps tend to have far fewer and smaller glaciers than the Northern and North-Western sides, due to exposure. Moreover, most Italian glaciers are situated at high elevations, but other factors, such as local climatic conditions (Faletto, Prola, Acquaotta, Fratianni, & Terzago, 2013), related to the source and propagation of humid maritime air, also influence their size and distribution (Calmanti, Motta, Turco, & Provenzale, 2007).

In the study area, the minimum altitude of present-day glaciers is reached by the snout of the Capra Glacier (2430 m), while the maximum altitude is attained by the Colle dell'Ape Glacier (3865 m). During the LIA, Italian glaciers reached a smaller extent compared to the other glaciers of the Western Alps, and at the end of the LIA underwent a more pronounced withdrawal. This was due to (i) the low mean altitude and latitude of the glaciated areas (between 44°06′43″N and 45°35′34″N), (ii) the local climatic factors (temperature and amount of precipitation) and (iii) the exposure (Mercalli & Cat Berro, 2005).

Glaciers of the Piedmont Alps have previously been investigated by Sacco (1925, 1927, 1928, 1932) and the Italian Glaciological Committee (Comitato Glaciologico Italiano; CGI, 1961); recently, accurate studies were carried out in the Maritime Alps by Federici, Pappalardo, and Ribolini (2003) and Federici and Pappalardo (2010), and in the Central Graian Alps by Mercalli and Cat Berro (2005).

The glaciated and recently deglaciated areas represented in the map are grouped into five main sectors, from North to South: Central Graian Alps (highest altitude Gran Paradiso, 4061 m), Southern Graian Alps (highest altitude Uia di Ciamarella, 3676 m), Northern Cottian Alps (highest altitude Pierre Menue, 3506 m), Central Cottian Alps (highest altitude Monviso, 3840 m) and Maritime Alps (highest altitude Cima Sud dell'Argentera, 3297 m).

3. Methods: map production

In order to produce the map of the LIA and current glacier extent, data available from different archives, as well as new data were considered.

3.1. Archive data

Data archives examined for this work were:

• glacier front variations documented by the annual surveys of the volunteers of the Italian Glaciological Committee since 1927 (www.glaciologia.it);

• the national glacier inventory promoted by the CGI on the occasion of the International Geophysical Year 1957–1958 (www.glaciologia.it);

• the World Glacier Inventory (WGI) on glacier distribution from the 1980s onwards (www.nsidc.org);

• official topographic maps of the Italian Military Geographical Institute (Istituto Geografico Militare Italiano, IGMI), at a scale of 1:25,000, dating from the end of the nineteenth century and the beginning of the twentieth century;

• historical photos and images from glaciological publications and photographic archives (IGMI, CGI, Italian Alpine Club, etc.; Figure 1).

Figure 1. The Teleccio Glacier under the pyramidal peak of Torre del Gran San Pietro (3692 m), viewed looking North-East from the peak of Gran Carro (2961 m), in a drawing of 1850 by M. A. Relly (left) and in a 2003 photo by D. Cat Berro (right).

The information collected from these archives facilitated identification of the maximum extent of the glaciers at the peak of the LIA, reconstruction of the steps of glacial retreat, and detection of some glaciers not included in the previous inventories. Although data archives have proven to be discontinuous and lacking regarding several aspects, they have been redrawn and represented on the topographic map of the Regione Piemonte (Carta Tecnica Regionale – 1:10,000 scale).

A general limitation of the data derived from these archives relates to the locations of the morphological features (i.e. morainic ridges), which are difficult to map in detail. Furthermore, the 1:10,000 scale topographic map is not congruent with the old IGMI maps, also used as base maps for the CGI inventory. In addition, data derived from historical photos, images and text documents (such as the annual glaciological campaigns carried out by the CGI volunteers) are difficult to interpret, in many cases, for the purpose and the detail of this map. Archive data are therefore important mainly for qualitative comparison of different glacial configurations.

3.2. New data

In order to draw both present-day and LIA glacier outlines (see Main Map), detailed analyses and mapping of glacial landforms related to the glacial expansion during the LIA (moraines and glacial cirques) and of more recent glacial features (post-LIA moraines, current glacial boundaries, supraglacial morainic ridges, supraglacial and proglacial lakes and main crevasses) were carried out. Geomorphological analysis was performed for the entire study area (about 140 km²) through a detailed interpretation of traditional aerial photos and digital orthophotos, available at a variety of scales and for different years.

For this purpose, the following images were considered:

• aerial photographs of the Regione Piemonte, produced in 2001, at a scale of approximately 1:20,000 (CGR-Parma and Rossi-Brescia);

• aerial photographs and digital orthophotos of the Comunità Montana Valle Orco, produced in 2005, at a scale of approximately 1:20,000 (Alifoto-Torino), only available for the Central Graian Alps (A, B, C in the Main Map);

• digital orthophotos, produced in 2006, at a scale of 1:10,000, accessible online on the Portale Cartografico Nazionale (Italian National Cartographic Web Portal, www.pcn.minambiente.it/GN/).

Traditional aerial photographs were analyzed using a classical stereoscope, while digital aerial photographs were analyzed using a digital stereo viewer.

Photo analysis led to the identification and mapping of glacial landforms and morphological features for the subsequent reconstruction of the current and LIA glacier extent; these data were stored in a spatial database.

Values of certain morphometric parameters (surface, length, width, mean slope, highest and lowest altitude, exposure) of the glaciers, for the LIA and 2006 time periods, were computed and associated with the glacial polygons through a geographical information system (GIS) dataset. With reference to surface parameters, only the planimetric values were taken into account because of the coarse-resolution of the digital terrain model (DTM; 50 m grid) and year of production (1991). In addition, topographic data related to the LIA were not available, and any attempt to convert the projected glacial areas into surface area would prove too unreliable. Finally, field surveys were carried out by the authors for the period 2010–2012 in order to check glacier boundaries and front positions, mainly in the most complex areas.

3.2.1. Reconstruction of peak LIA glacier extent

In order to define the maximum glacial expansion at the peak of the LIA, the main morphological features (moraines and glacial cirques) in the deglaciated areas were identified and mapped. Following this, the morainic ridges were interpolated to define lateral and frontal glacier boundaries and to reconstruct the glacier outlines. With the aid of archive data, glacier perimeters were then extrapolated to include areas where the morainic ridges were not formed or preserved.

The reliability of the reconstructed boundaries of LIA glaciers depends on the presence of evident morphological features, their spatial continuity and degree of preservation, sometimes conditioned by the presence of snow and debris cover. The accuracy of the mapping is also a function of the quality of the aerial imagery; the main obstacles to clear observations are the presence of shadows in steep slopes, cloud cover obscuring underlying areas, and the low resolution of some photos. In these cases, the interpretation was quite subjective, and the data acquired by photo analysis where possibly were integrated by means of field surveys. However, a systematic field check of the morphological data was beyond the objectives of the GlaRiskAlp project, given the extent of the study area, along with time constraints. Nevertheless, we tried to overcome any uncertainties related to photo interpretation by comparing different sets of aerial photos, and with the aid of the historical documentation.

3.2.2. Current glacier extent

For mapping the current glacier extent, the most recent available documentation was analyzed, and compared with the glacial boundaries drawn through the photo interpretation of the 2006 orthophotos. Glacial boundaries were also compared to the data of the annual glaciological campaigns of the CGI, and with field surveys carried out by the authors during the period 2010–2012. Other significant morphological features related to present-day glaciers, such as crevasses and lakes, were also included in the map.

A major uncertainty is introduced in the definition of glacier outlines when large portions of ice, especially at the lateral contact with slopes, are masked by snow or debris. Furthermore, dead ice and snowfields cannot always be clearly distinguished from glacier ice by means of photo interpretation. In these cases, we tried to use other morphological elements (such as crevasses and slope profiles), together with the surface appearance of snow cover, to identify glacial boundaries. In these cases, field observations were also useful to precisely locate the present-day glacier fronts.

3.3. The main map

The final document is a synthetic map at a scale of 1:25,000 containing the extent of 96 LIA glaciers and 36 current glaciers, related linear landforms (morainic ridges, crevasses and glacial cirques) and present-day lakes.

A progressive code number, from North to South, has been attributed to each LIA glacier; the same code also refers to the current glacier, if preserved. As some LIA glaciers were fragmented into smaller units, current glacial bodies (66) are identified by the codes of the original LIA glaciers, followed by a letter. This numbering was introduced by the authors for the specific needs of the GlaRiskAlp project: in Table 1, glacier project codes are associated with the national codes (CGI codes) as well as international ones (WGI codes). Nevertheless, in the morphometric analysis, glacial bodies coming from the fragmentation of the same LIA glacier were computed as a single glacier.

Table 1. Morphometric parameters of the LIA and 2006 glaciers. n.°: LIA and 2006 glacier codes. H_max: highest altitude. H_min: lowest altitude. S: surface. N.^o: CGI and WGI codes.

	Little Ice Age				2006					CGI	WGI
Glacier Name	N.°	H max (m)	H min (m)	S (km^2)	N.°	H max (m)	H min (m)	S (km^2)	Glacier Name (CGI)	N.°	N.°
Sengie Orientale Nord	1	3265	2635	0.322	–	–	–	–	Sengie Orientale	82	IT4L01052004
Sengie Orientale Sud	2	3290	2870	0.121	–	–	–	–	Sengie Orientale	82	IT4L01052004
Ciardoney	3	3295	2600	1.551	–	3110	2865	0.608	Ciardoney	81	IT4L01052003
Geri Ovest	4	3010	2550	0.486	–	–	–	–	Geri	85	IT4L01052002
Geri Est	5	2960	2565	0.148	–	–	–		Geri	85	IT4L01052002
Valsoera	6	3380	2800	0.595	6a	3210	3080	0.037	Valsoera	80	IT4L01051036-7
					6b	3170	3010	0.049	–	–	–
Teleccio	7	3460	2670	0.637	7a	3400	3220	0.046	Teleccio	79	IT4L01051035
					7b	3430	3050	0.388	–	–	–
Monte Nero	8	3460	2850	0.341	–	–	–	–	–	–	–
Roccia Viva	9	3645	2585	0.968	–	3440	3010	0.403	Roccia Viva	78	IT4L01051033
Becchi	10	3170	2755	0.289	–	–	–	–	Becchi	84	IT4L01051032
Colle di Noaschetta Est	11	3010	2760	0.060	–	–	–	–	Colle Noaschetta	83	IT4L01051031
Colle di Noaschetta Ovest	12	3150	2905	0.096	–	–	–	–	–	–	–
Losa	13	3210	2985	0.199	–	–	–	–	Losa	77	IT4L01051029
Becca di Gay	14	3595	3225	0.131	–	3550	3430	0.045	Becca di Gay	76	IT4L01051028
Gay	15	3530	2895	0.892	–	3500	3260	0.083	Gay	75	IT4L01051027
Noaschetta	16	3670	2720	3.992	16a	3430	3115	0.355	Noaschetta, Goi	71	IT4L01051023, IT4L01051024-5
					16b	3470	3110	0.454	–	–	–
					16c	3500	3180	0.524	–	–	–
					16d	3430	3240	0.115	–	–	–
Punta Ceresole	17	3690	3525	0.042	–	3670	3540	0.035	Punta Ceresole	74	not evaluated on WGI
Colle dell'Ape	18	3880	3600	0.058	–	3865	3650	0.035	Colle dell'Ape	73	not evaluated on WGI
Ciamousseretto Nord	19	3480	2990	0.531	–	3400	3110	0.321	Ciamousseretto	70	IT4L01051022
Ciamousseretto Sud	20	3190	2915	0.076	–	–	–	–	–	–	–
Breuil Nord	21	3480	2805	0.617	21a	3210	3045	0.161	Breuil	69	IT4L01051020
					21b	3440	3160	0.038	–	–	–
Breuil Sud	22	2930	2620	0.203	–	–	–	–	–	–	–
Porta Nord	23	3110	2705	0.372	–	–	–	–	Porta Nord	68	IT4L01051018
Porta Occidentale	24	3070	2810	0.232	–	–	–	–	Porta Occidentale	67	IT4L01051017
Mare Percia	25	3270	2820	0.391	–	–	–	–	Mare Percia	66	IT4L01051016
Punta Fourà	26	3250	2970	0.417	–	–	–	–	Punta Fourà Meridionale	65	IT4L01051015
Basei	27	3320	2635	0.875	–	3310	2945	0.274	Basei	64	IT4L01051013
Losa	28	3100	2510	0.528	28a	2995	2830	0.104	Losa	62	IT4L01051011
					28b	2990	2950	0.004	–	–	–
Capra	29	3005	2345	0.474	–	2860	2430	0.250	Capra	61	IT4L01051010
Carro Occidentale	30	3320	2485	1.695	30a	3280	2920	0.315	Carro Occidentale	60	IT4L01051009
					30b	3050	2840	0.156	–	–	–
					30c	3040	2760	0.076	–	–	–
					30d	2940	2750	0.056	–	–	–
Carro Orientale	31	3235	2320	0.820	31a	2910	2650	0.179	Carro Orientale	59	IT4L01051006
					31b	3250	3090	0.023	–	–	–
Nel	32	3385	2370	3.068	–	3250	2730	1.427	Nel Occidentale, Nel Centrale	58-57	IT4L01051005
Levannetta	33	3010	2340	0.885	–	2850	2550	0.156	Levannetta	56	IT4L01451004
Col Perduto	34	3440	2535	0.346	–	–	–	–	Col Perduto	55	IT4L01051002
Forno	35	2990	2560	0.492	–	–	–	–	Forno	54	IT4L01051001
Barrouard	36	2710	2420	0.176	–	–	–	–	Barrouard	53	not evaluated on WGI
Levanna Orientale	37	3480	2665	1.128	37a	3440	3130	0.158	Levanna Orientale Sud, Nord	51-52	IT4L01463013, IT4L01463015
					37b	3280	3210	0.009	–	–	–
					37c	3460	3330	0.019	–	–	–
					37d	3230	2910	0.132	–	–	–
Talancia Girard	38	3070	2705	0.092	–	–	–	–	Talancia Girard	50	IT4L01463012
Martellot	39	3100	2340	0.316	39a	2910	2755	0.021	Martellot	49	IT4L01463011
					39b	2850	2600	0.041	–	–	–
Mulinet	40	3210	2320	1.197	40a	3040	2635	0.154	Mulinet Sud, Mulinet Nord	47-48	IT4L01463009, IT4L01463010
					40b	3060	2690	0.312	–	–	–
Bonneval Est	41	3145	3030	0.036	–	–	–	–	–	–	–
Bonneval Ovest	42	3260	2935	0.298	–	3200	3045	0.014	–	–	–
Bonneval Sud	43	3260	2950	0.281	43a	3155	3050	0.056	–	–	–
					43b	3110	3060	0.019	–	–	–
Sea	44	3670	2440	2.711	44a	3260	2740	0.753	Albaron di Sea, Tonini, Sea	44-45-46	IT4L01463001, IT4L01463002, IT4L01463003
					44b	3675	3450	0.040	–	–	–
					44c	3620	3520	0.020	–	–	–
					44d	3400	3200	0.030	–	–	–
					44e	3410	3245	0.147	–	–	–
					44f	3250	3000	0.126	–	–	–
Ciamarella	45	3460	2770	1.179	–	3410	3090	0.659	Ciamarella	43	IT4L01462007
Collerin d'Arnas	46	3480	2495	2.114	46a	3315	3090	0.066	Collerin d'Arnas, Pian Gias	42-41	IT4L01462006, IT4L01462005
					46b	3460	3020	0.209	–	–	–
					46c	3100	3060	0.004	–	–	–
					46d	2950	2810	0.053	–	–	–
Bessanese	47	3220	2565	1.750	47a	3160	2810	0.309	Bessanese	40	IT4L01462004
					47b	2840	2670	0.079	–	–	–
Collarin Ovest	48	3020	2805	0.265	–	–	–	–	–	–	–
Collarin Est	49	2990	2720	0.351	–	–	–	–	–	–	–
Servin	50	2910	2520	0.515	–	–	–	–	Servin	39	IT4L01462001
Croce Rossa Inferiore	51	3085	2700	0.407	–	–	–	–	Croce Rossa	38	IT4L01461010
Croce Rossa Superiore	52	3565	3155	0.117	–	3550	3340	0.070	Croce Rossa	38	IT4L01461010
Altare	53	3210	2860	0.224	–	–	–	–	–	–	–
Pera Ciaval	54	3220	2900	0.568	54a	3170	3075	0.012	Pera Ciaval	37	IT4L01461008
					54b	3060	2995	0.007	–	–	–
Bertà	55	3260	2715	0.994	55a	3200	2975	0.071	Bertà	36	IT4L01461007
					55b	3160	2900	0.024	–	–	–
Rocciamelone	56	3510	2570	1.374	56a	3255	3100	0.060	Rocciamelone	35	IT4L01461002
					56b	3260	3110	0.166	–	–	–
					56c	3275	3050	0.060	–	–	–
					56d	3390	3320	0.018	–	–	–
Savine Orientale	57	3260	2990	0.212	–	–	–	–	Savine Orientale	31	IT4L01471007
Muttet	58	3300	2900	0.510	–	–	–	–	Muttet	30	IT4L01471006
Agnello	59	3290	2700	1.753	–	3200	3010	0.089	Agnello	29	IT4L01471004
Punta Sommeiller Est	60	3180	2990	0.058	–	–	–	–	Punta Sommeiller	28	IT4L01471003
Punta Sommeiller Ovest	61	3315	2900	0.238	–	–	–	–	Punta Sommeiller	28	IT4L01471003
Galambra	62	3180	2610	0.697	–	–	–	–	Galambra	26	IT4L01471002
Vallonetto	63	3170	2890	0.328	–	–	–	–	Vallonetto	25	IT4L01472002
Fourneaux	64	3135	2785	0.559	–	–	–	–	Fourneaux	27	IT4L01472001
Boucher	65	3070	2790	0.257	–	–	–	–	Boucher	24	not evaluated on WGI
Due Dita	66	3010	2440	0.070	–	–	–	–	Due Dita (Visolotto)	23	IT4L01481004
Coolidge Inferiore	67	3070	2350	0.215	–	–	–	–	Coolidge Inferiore	22	IT4L01481003
Coolidge Superiore	68	3720	3030	0.087	–	3280	3140	0.023	Coolidge Superiore	20	IT4L01481003
Viso Nord-Est	69	2850	2470	0.382	–	–	–	–	Viso Nord-Est	21	IT4L01481002
Punta Dante	70	3065	2965	0.041	–	–	–	–	Punta Dante	19	IT4L01482008
Sella	71	3410	3275	0.033	–	–	–	–	Sella	18	IT4L01482007
Viso	72	3450	3020	0.119	–	–	–	–	Viso	17	IT4L01482006
Quarnero	73	3350	3275	0.053	–	–	–	–	Quarnero	16	IT4L01482005
Caprera	74	3040	2705	0.120	–	2930	2805	0.025	Caprera	15	IT4L01482004
Vallanta Inferiore	75	3245	2750	0.268	–	2980	2765	0.057	Cadreghe, Vallanta Basso	12-13	IT4L01482002
Vallanta Superiore	76	3740	3220	0.136	–	3670	3450	0.028	Vallanta Superiore	14	IT4L01482003
Ubac	77	2920	2600	0.068	–	–	–	–	Ubac	10	IT4L01323004
Schiantala	78	2820	2530	0.160	–	–	–	–	Schiantala	9	not evaluated on WGI
La Madre di Dio	79	2745	2175	0.217	–	–	–	–	–	–	–
Cessole	80	2730	2350	0.128	–	–	–	–	Cessole	8	IT4L01312016
Lourousa	81	3200	2390	0.140	–	–	–	–	Lourousa	7	IT4L01312015
Baus	82	2860	2555	0.109	–	–	–	–	–	–	–
Brocan Ovest	83	2900	2625	0.109	–	–	–	–	–	–	–
Brocan Est	84	2760	2350	0.208	–	–	–	–	–	–	–
Agnel Ovest	85	2775	2460	0.054	–	–	–	–	–	–	–
Agnel Est	86	2890	2320	0.146	–	–	–	–	–	–	–
Valletta	87	2650	2500	0.035	–	–	–	–	–	–	–
Gelas Sud	88	2600	2410	0.073	–	–	–	–	–	–	–
Gelas Ovest	89	2980	2480	0.237	–	–	–	–	–	–	–
Ciafraion	90	3000	2620	0.273	–	–	–	–	Ciafraion	5	IT4L01312007
Gelas Est	91	2950	2525	0.503	–	–	–	–	Gelas	6	IT4L01312006
Muraion	92	2820	2450	0.103	–	–	–	–	Muraion	4	not evaluated on WGI
Maledia	93	2810	2600	0.110	–	–	–	–	Maledia	3	IT4L01312004
Peirabroc Ovest	94	2870	2620	0.054	–	–	–	–	–	–	IT4L01312003
Peirabroc Est	95	2750	2310	0.161	–	–	–	–	Peirabroc	2	IT4L01312002
Clapier	96	2870	2475	0.386	–	–	–	–	Clapier	1	IT4L01312001

Current linear landforms, current glaciers and lake polygons, along with LIA glacier polygons were digitized into three separate shapefiles. To compile the Main Map, vector layers were then exported (as .EPS files) into vector graphic editing software. As a topographic support, the digital map (1:50,000 scale) of the Regional Agency for the Protection of the Environment (ARPA Piemonte) (http://webgis.arpa.piemonte.it/geoportale/index.php/it/servizi-wms-wfs.html), published in 2011, was used.

4. Discussion and conclusions

To evaluate the changes in the glaciated areas over the last 150 years, the limits reached by glaciers in the Western and South-Western Piedmont Alps during the LIA, and in 2006, were mapped. Morphological analysis performed with interpretation of aerial photos, combined with field work and analysis of historical maps and images, made it possible to identify and trace the outline of the 96 LIA glaciers, and to carry out a preliminary analysis of their extent.

The glaciers are distributed as follows:

• 36 glaciers (n.° 1–36 in the Main Map) in the Central Graian Alps, including the Orco and the Soana Valleys, with an overall area of 23.12 km²;

• 20 glaciers (n.° 37–56) in the Southern Graian Alps, including the Lanzo Valleys, with an overall area of 15.92 km²;

• 9 glaciers (n.° 57–65) in the Northern Cottian Alps, including the Susa Valley, with an overall area of 4.61 km²;

• 11 glaciers (n.° 66–76) in the Central Cottian Alps, closest to the Monviso Massif (upper Po and Varaita Valleys) covering an area of 1.52 km²;

• 20 glaciers (n.° 77–96) in the Maritime Alps, including the Stura di Demonte and the Gesso Valleys, with an overall area of 3.27 km².

The total area covered by ice during the peak of the LIA amounted to 48.45 km², whereas individual glaciers extended over areas ranging between 0.03 km² (Sella Glacier, n.° 71, in the Monviso Massif) and 3.99 km² (Noaschetta Glacier, n.° 16, in the Gran Paradiso Massif). The minimum altitude reached by the glacier fronts ranged from 2175 m (La Madre di Dio Glacier, n.° 79, in the Maritime Alps) to 3600 m (Colle dell'Ape Glacier, n.° 18, in the Gran Paradiso Massif).

Most of the mapped LIA glaciers were cirque glaciers, sometimes resulting from the coalescence of several smaller cirque glaciers (e.g. Carro Occidentale and Nel Glaciers, n.° 31 and 32; Figure 2). In just a few cases, the volume and length of the glacial masses were sufficient to enable the development of valley glaciers, such as in the case of the Sea and Collerin d'Arnas Glaciers (n.° 44 and 46, in the Lanzo Valleys), with lengths of 2.8 km and 3.8 km, respectively.

Figure 2. The Levanna Orientale Glaciers viewed looking West from Barrouard (2856 m). The upper photo was taken in 1902 by C. Grosso, the lower photo by F. Rogliardo in 2011 (source: CGI archives).

The maximum glacial advance reached at the end of the LIA is almost always evidenced by well-developed ice-marginal moraines, rising up to 40–50 m. Impressive terminal dump moraines (Boulton & Eyles, 1979), up to one kilometer in length, are visible at the edges of the Capra Glacier, in the Orco Valley (Figure 3), and along the left flank of the Bessanese Glacier in the Lanzo Valleys (n.° 47 in the Main Map). Currently, most of the nineteenth century moraines are irregularly colonized by pioneer herbaceous vegetation, mostly on slopes that are not subjected to intense rain-wash. In some terminal moraines, blocks and rock fragments are almost completely devoid of lichens, whereas in other cases an incipient lichen cover (thallus size <8–10 mm) is visible on rock surfaces.

Figure 3. The Capra Glacier viewed looking South-West from the Lago Serrù dam (2283 m). Left in a 1957 photo from CGI archives, right in a 2010 photo by G. Fioraso.

Since the end of the LIA, end-moraines have been affected by intense erosion, also favored by the absence of vegetation cover, which in some cases radically changed the original morphology of such glacial landforms. For instance, on 24 September 1993, after an intense rainstorm, a large portion (800,000 m³) of the frontal moraine of the Mulinet Glacier (n.° 40 in the map) was removed by a huge debris flow that traveled 4.5 km downstream (Mortara, Dutto, & Godone, 1995).

Between 1850 and 2006, the overall number of glaciers in the studied area has decreased by 63.5%. The 36 glaciers still present in 2006 are distributed as follows:

• 18 glaciers in the Central Graian Alps, covering an area of 6.72 km²;

• 13 glaciers in the Southern Graian Alps, with an overall area of 3.83 km²;

• the Agnello Glacier in the Northern Cottian Alps (0.09 km²);

• 4 glaciers in the Monviso Massif (in total 0.13 km²).

In 2006, the total surface area covered by ice in the studied area was 10.77 km², which correspond to a reduction of 77.8% compared with the glacier extent during the LIA. The positive/negative interplay of factors such as altitude, aspect, solar radiation and relief morphology has been critical in the survival/melting of glaciers. In some cases, impressive recession of glaciers also involved their upper edges; for instance, the highest point of the Coolidge Superiore Glacier (n.° 68 in the Monviso Massif), has reduced from 3720 m during the LIA, to a current altitude of 3280 m (Figure 4). Moreover, compared to the nineteenth century, in 2006 some glaciers (17) were subdivided into individual smaller parts (e.g. Sea and Rocciamelone glaciers, n.° 44 and 56).

Figure 4. Comparison of the Coolidge Superiore Glacier extent (yellow dashed line) between around 1925 (left, CGI archives) and 2012 (right, photo by G. Fioraso), viewed looking SSE from Pian del Re (2013 m).

Over the last 150 years, deglaciation was interrupted for short periods by a halt or readvance of glacier fronts (for instance from 1950 to 1980, as a consequence of a slight global cooling; Biancotti & Motta, 2000; Citterio et al., 2007; Zemp et al., 2008). Standstill and/or fluctuations of ice margins, in some cases (e.g. Capra Glacier, n.° 29), are documented by recessional moraines abandoned at the innermost edge of the LIA moraines; these moraines do not exceed 1–2 m in height, and are less continuous compared to those attributed to the LIA.

Software

Map design and database construction were carried out using ESRI ArcGIS 9.3. The final graphical production of the map was performed using Adobe Illustrator CS3.

Acknowledgments

The present study was carried out under the framework of the Alcotra 2007–13 Project n. 56 ‘GlaRiskAlp’. We acknowledge all the partners of the project (Fondazione Montagna Sicura, project leader, ARPA Valle d'Aosta, Université de Savoie – Laboratoire LISTIC, CNRS France – Laboratoires LGGE, GIPSA, EDYTEM), for fruitful discussions.

Special thanks go to Giovanni Mortara, Marco Giardino, Luigi Perotti, Franco Godone and Guido Nigrelli, for their valuable scientific and technical advice during the development of the project. Finally the Authors thank Matteo Spagnolo, Philip Deline and John Abraham for constructive reviews.