Revisiting the upper Visean mud mounds from Derbyshire (UK): the role of brachiopods in their growth

Several brachiopod-rich mud mounds occur in the upper Visean (Brigantian) of the Derbyshire Carbonate Platform succession in UK. The re-evaluation of the lithofacies architecture of a Derbyshire mud mound complex, developed in an intraplatform middle-ramp environment, led to the recognition of three lithofacies associations: (a) a 10 m thick basal unit of automicrite boundstone with siliceous sponge spicules and brachiopod–bryozoan packstone to wackestone beds; (b) a 10 m thick, 250 m wide, lens-shaped, convex-up massive core of clotted peloidal micrite and fenestellid bryozoan boundstone with sponge spicules; (c) inclined brachiopod–bryozoan–crinoid packstone flank beds. In the mud mound complex core, most of the carbonate mud with clotted peloidal and structureless micrite fabric is the result of biologically induced and influenced in-situ precipitation processes (automicrite). Brachiopods are not, as previously thought, limited to storm-scoured “pockets” in the mud mound complex core but are abundant and diverse in all lithofacies and lived on the irregular mud mound complex surface concentrating in depressions sustained by automicrite boundstone and the growth of bryozoans and sponges. The upper Visean Derbyshire mud mounds are, thus, representatives of a newly defined fenestellid bryozoan–brachiopod–siliceous sponge mud mound category, occurring in various middle–upper Visean Western European sites, a sub-type of the fenestellid bryozoan–crinoid–brachiopod Type 3 buildups of Bridges et al. (1995). These mud mounds, and other types of brachiopod-rich buildups, developed in carbonate platform settings between fair-weather and storm wave base, in dysphotic environments with dispersed food resources during the Visean. Brachiopod mud mound colonisation was favoured by moderate water depth, availability of food resources, and diverse substrates.

The microbial-metazoan community responsible for the growth of mud mounds was increasingly common from the late Frasnian (Webb 1996(Webb , 2002 and became widespread after the extinction of metazoan reef frame-builders (e.g. stromatoporoids, rugose and tabulate corals) at the Frasnian-Famennian boundary (Kellwasser Extinction Event; e.g. Wright and Faulkner 1990;Buggisch 1991) and at the end of the Devonian (Hangenberg Event; e.g. Yao et al. 2016. Carbonate mud precipitation associated with microbial mats and other organic substrates in marine and non-marine present-day and ancient settings can take place through different pathways: (a) mediated by the metabolic activity of microbial communities (biologically induced mineralisation sensu Dupraz et al. 2009), or (b) result from the interaction of CaCO 3 -supersaturated water with organic substrates (biologically influenced mineralisation sensu Dupraz et al. 2009), e.g. decaying organic tissues such as biofilm extracellular polymeric substances (EPS) in microbial mats or siliceous sponges (organomineralisation sensu Trichet and Défarge 1995, Reitner et al. 1995a, 1995b. These processes produce characteristic textures in the carbonate mud precipitates: sub-millimetric peloids with diffuse margins, commonly surrounded by a microsparite mosaic, defined as clotted peloidal micrite (Pickard 1996), or structureless micrite, gravity-defying, sustaining millimetre-size primary cavities, termed leiolite (Braga et al. 1995). In Mississippian mud mounds, most carbonate mud is interpreted as precipitated through microbially mediated in situ processes (Lees and Miller 1995;Monty 1995;Pickard 1996) and/or passive organomineralisation of sponge soft tissues (Reitner et al. 1995a, b), based on the presence of such textures in the carbonate mud (Pickard 1996). Collectively, in situ carbonate mud precipitates related to processes of biologically induced and influenced mineralisation are referred to as "automicrite" (Wolf 1965;Reitner et al. 1995a). Mud-grade carbonate sediment, either produced by carbonate grain abrasion and bioerosion, or by reworking of in situ precipitates, was also common in some mud mounds, and labelled as allomicrite (Wolf 1965;Lees and Miller 1985;Devuyst and Lees 2001).
These mud mounds consist of complexes resulting from the accretion of individual mounds (sensu Gutteridge 1995;Aretz and Chevalier 2007), up to 400 m in thickness and some kilometres in width (Lees and Miller 1995;Somerville 2003). Waulsortian mud mounds show skeletal assemblages varying vertically in the complexes.  and Miller (1985, 1995) proposed a bathymetric succession of four "Waulsortian phases" based on the skeletal assemblages, from deeper to shallower (phases A-D). Also, Waulsortian mud mounds are characterised by multicomponent muds ("polymuds"; Miller 1985, 1995;Devuyst and Lees 2001), highly structured mud deposits showing successive geopetal fillings of different mud generations in primary cavities.
Other mud mound types developed in middle to outerramp settings in the Mississippian include crinoid-fenestellid bryozoan mud mounds with green shale in the core in the Tournaisian of Kentucky (Ausich and Meyer 1990;Meyer et al. 1995; Type 2 buildups in the classification of Bridges et al. 1995), and skeletal-rich mud mounds with a diverse bryozoan community, dasycladacean algae, and rugose corals in the Visean of northwest Ireland (e.g. Schwarzacher 1961;Kelly and Somerville 1992) and Algeria (Madi et al. 1996; Waulsortian-type or Type 1A mounds in the classification scheme of Somerville 2003). Trepostome bryozoan-microthrombolite mud mounds commonly with serpulid and problematic worm tubes (Type 5 buildups of Bridges et al. 1995) occurring in the Maritime Provinces of Canada (e.g. Dix and James 1989;Boehner 1989;Boehner et al. 1989), developed in lower-upper Visean deep basinal to shallow-water high-energy hydrothermal-seep-related settings (von Bitter et al. 1990(von Bitter et al. , 1992Boehner et al. 1989;Webb 2002).
One of the ranges was the Derbyshire High, between the Edale Basin to the north and the Widmerpool Gulf to the south (Fig. 1). The Derbyshire High is interpreted as comprising two basement tilted blocks, the Eyam Block to the north and the Woo Dale Block to the south (Smith et al. 1985), with a fault terrace in-between bounded by the Bakewell Fault and the Cronkston-Bonsall Fault (Fig. 2a;Gutteridge 1989a).
In the late Tournaisian, the marine transgression on the Derbyshire High (Waters et al. 2009), situated at that time at subtropical latitudes (Piper et al. 1991;Woods and Lee 2018), led to the initiation of carbonate deposition on the irregular basement substrate: Woo Dale Limestone Formation (Fig. 2b;Aitkenhead and Chisholm 1982). In the late Visean, the area evolved into a flat-topped carbonate platform, characterised by peritidal deposition at estimated water depths of 0-25 m (Bridges 1982;Walkden 1987;Manifold et al. 2020), with skeletal grainstone shoals on the margins and sponge-microbial reefs at the break-in-slope to the north and west (Wolfenden 1958;Stevenson and Gaunt 1971;Broadhurst and Simpson 1973;Harwood 2005;Manifold et al. 2021): Bee Low Limestone Formation ( Fig. 2b; Aitkenhead and Chisholm 1982).
Reactivation of extensional tectonics in the latest Visean, during deposition of the Monsal Dale Limestone Formation (Aitkenhead and Chisholm 1982), led to the differentiation on the Derbyshire Carbonate Platform of a flat-topped platform interior region to the west and south and intraplatform basins to the east (Fig. 2a) and Chisholm (1982), Gutteridge (1987), Aitkenhead et al. (2002). b Stratigraphic column of the study area with regional UK chronostratigraphy, based on data provided by Waters et al. (2009Waters et al. ( , 2011 Page 5 of 39 9 flat-topped platform interior was characterised by the deposition of shallowing-upward 0.5-5 m thick cyclothems (Stevenson and Gaunt 1971;Cox et al. 1977;Aitkenhead et al. 1985;Gutteridge 1987). The restricted intraplatform basins in the east were characterised by low-angle ramp margins and the deposition of various units in normal marine, above and below fair-weather wave base, poorly oxygenated environments, to hypersaline and brackish ponds (Butcher and Ford 1973;Gutteridge 1989a). The north-western margin of the platform in the area of Castleton and to the south of Buxton evolved into a bypass slope dominated by storm-influenced skeletal grainstone shoals (Gawthorpe and Gutteridge 1990;Harwood 2005), whereas the southern margin was a low-angle distally steepened ramp with tidal bars (Gutteridge 2003;Harwood 2005). Sponge-microbial reefs growth halted on the break-in-slope while mud mounds developed on the outer platform (Harwood 2005). Basaltic lava, hyaloclastite, and tuff occur intercalated in the Monsal Dale Limestone Fm (Aitkenhead and Chisholm 1982).
Near the top of the Monsal Dale Limestone Fm, immediately beneath the regional subaerial exposure surface marking the boundary with the overlying Eyam Limestone Formation (Adams 1980;Aitkenhead and Chisholm 1982;Gutteridge 1991) mud mounds ("knoll-reefs" or "reef-knolls" in the older literature; e.g. Smith et al. 1967;Stevenson and Gaunt 1971;Aitkenhead et al. 1985), referable to Bridges et al. (1995) Type 3 buildups developed in the platform interior, margins, and intraplatform ramps (Fig. 2a;Biggins 1969;Orme 1971;Timms 1978;Gutteridge 1995). Mud mounds frequently accreted to form mud mound complexes whose depositional style was affected by the availability of accommodation (Gutteridge 1995): laterally accreted mud mound complexes developed above fair-weather wave base and are hundreds of metres wide, up to 20 m thick; vertically accreted mud mound complexes resulted from the vertical growth of one or a few lens-shaped mud mounds, attaining a thickness up to 50 m (Gutteridge 1995). The mud mound complex at Ricklow Quarry, the focus of this study, is within the laterally accreted group and was interpreted as developed in an intraplatform inner-ramp setting according to Gutteridge (1995), but a middle-ramp setting below fairweather wave base was proposed by Nolan et al. (2017). Few mud mounds also occur in intraplatform inner-ramp units of the Eyam Limestone Fm (Gutteridge 1995).
Platform growth was terminated by southerly progradation of fluvio-deltaic systems in the early Serpukhovian (Aitkenhead et al. 1985;Guion and Fielding 1988).
The Monsal Dale and Eyam Limestone formations have been traditionally attributed to the Brigantian substage (Ramsbottom in Smith et al. 1967;Mitchell in Stevenson and Gaunt 1971;Fig. 2b). Several studies suggest a correlation of the middle-upper Brigantian of England with the Serpukhovian of the Moscow Basin, mainly based on foraminiferal biozonations (e.g. Sevastopulo and Barham 2014;Vachard et al. 2016;Somerville 2014, 2021;Cózar et al. 2019a). However, this study follows Waters et al. (2009Waters et al. ( , 2011, Aretz et al. (2020), and Lucas et al. (2022) that proposed correlating the upper boundary of the Brigantian with the lower boundary of the Serpukhovian. No biostratigraphic marker of the Serpukhovian occur in the Monsal Dale Limestone Fm of Derbyshire (Nolan et al. 2017;Smith et al. 2017;Carniti et al. 2022).

Materials and methods
The studied mud mound complex crops out to the west of Ricklow Quarry (53° 11′ 30'' N, 01° 45′ 17'' W), near the village of Monyash, Derbyshire, UK (Figs. 2a,3a). The outcrop is crossed by the Ricklow Dale to the north and Lathkill Dale to the south, resulting in natural sections of its succession and surrounding units (Fig. 3b).
Outcrop investigation, measurement of stratigraphic logs, and petrographic analysis of 57 thin sections with transmitted polarised-light microscope led to the recognition of fifteen lithofacies types. The lithofacies were grouped in lithofacies associations based on depositional texture (cf. Dunham 1962;Embry and Klovan 1971), origin of the carbonate mud (e.g. detrital allomicrite vs. in situ precipitated automicrite), composition of skeletal and non-skeletal grains, bedding character, sedimentary structures, and stratigraphic position. The semi-quantitative proportion of grains in thin sections was determined based on visual estimates and comparison charts for limestone by Baccelle and Bosellini (1965). Spatial distribution of the distinguished lithofacies was determined through lateral tracing between measured stratigraphic logs and outcrops and was reported on a geodatabase built with ArcGIS (Esri®) software based on 1:1,250 topographic maps by Ordnance Survey, containing data about outcrop position, lithofacies classification, strike and dip of beds, major structural features and sample location. The geodatabase was used to extract a lithofacies map of Ricklow Quarry (Fig. 3b). Fossil brachiopod assemblages were collected from the mud mound complex lithofacies. Details about palaeontological sampling and the taxonomy are provided in Carniti et al. (2022). The articulation ratio was calculated based on data from Carniti et al. (2022) as the percentage ratio of articulated specimens to the total number of specimens in each lithofacies association. Bias due to possible uneven distribution of articulated specimens was lowered by considering all 9 Page 6 of 39 brachiopods from different assemblages collected in the same lithofacies association. Cathodoluminescence analysis was performed on thin sections with a luminoscope CITL Cambridge Image Technology Limited, Cambridge, UK (model MK 5-2 operating system at 10-14 kV with a beam current between 200 and 600 µA, and vacuum gauge 50-70 mTorr) at the University of Milan.

Lithofacies character and architecture
The sedimentary succession investigated at Ricklow Quarry (Figs. 3,4,5) consists of a massive convex-upward lensshaped unit (about 10 m thick) extending laterally for nearly 250 m in an E-W cross section and adjacent to decimetre-thick inclined beds (dips up to 44° with respect  vey (1978). For legend refer to Fig. 2. b Lithofacies map (outcrops only) of the area of the mud mound complex outcrops to the west of Ricklow Quarry, result of this study. The location of the Ricklow Log studied by Nolan et al. (2017) is reported. L Lithofacies, LA Lithofacies Association to present-day-horizontal), transitional into a horizontally bedded succession (Figs. 4,5). The massive unit represents the mud mound complex core adjacent to its flank beds. Fifteen lithofacies, L1-L15, described in detail in Table 1, were distinguished as pre-mud mound complex (L1), basal unit of the mud mound complex (L2-L4), mud mound complex core (L6-L9), inclined flank beds (L10, L11), lateral to off-mud mound complex strata (L5, L12), and strata overlying the mud mound complex (L13-L15). The lithofacies association (LA1-LA3) characteristics are summarised in Table 2.
The association of L2 and L3 beds with L4 lenses builds the basal tabular unit of the mud mound complex, which is considered as Lithofacies Association 1 (LA1; Fig. 3b; Table 2). Brachiopods are the dominant macrofossils in LA1: specimens are both articulated and not articulated (articulation ratio 37.8%), commonly not in life position. Some productides occur in life position, with articulated spines, at the base of beds (Fig. 4c). A brachiopod faunal list for unit LA1 is given in Table 3. LA1 unit is adjacent to 8-40 cm thick horizontal massive beds of crinoidal grainstone/rudstone with automicrite patches (L5), with sparse bryozoans, brachiopods, and automicrite intraclasts (Figs. 3b,. The LA1 passes upwards to a 10 m thick association of massive lens-shaped units with a thickness of 1-2 m and a diameter of 4-5 m; the units laterally and vertically accrete to form decametre-scale lens-shaped convex-up mud mounds (Figs. 3b,5a,b). Individual mud mounds coalesced to form the mud mound complex; however, in some areas, the buildup appears as an undifferentiated massive object and the composite nature is not recognisable. The lensshaped units consist mainly of millimetre-to decimetre-size patches of clotted peloidal micrite boundstone (L7; 60% rock volume), chaotically mixed with minor fenestellid bryozoan boundstone/cementstone (L6; 10% rock volume), with skeletal grainstone/rudstone filling the primary voids in-between the patches (L8; 30% rock volume).
Also, in the mud mound complex core there are decimetre-to metre-size volumes of poorly defined, laterally discontinuous, 5-15 cm thick sub-horizontal beds (Fig. 5c,d) formed by automicrite boundstone with patches of packstone/rudstone forming up to 30-40% rock volume (L9). Skeletal content is higher than in L7 (Fig. 10e, f). Irregular volumes filled by a bryozoan fragment grainstone similar to L8 are more abundant then in L7 (20% rock volume).
Dissolution vugs, rhizolith, alveolar texture, Fe oxide red colour staining of grains and cement occur at the top of the mud mound complex (Figs. 5, 12a, b). Crinoidal grainstone/rudstone beds (L13), 5-40 cm thick, with sparse brachiopods and rare angular intraclasts of clotted peloidal micrite boundstone (Fig. 12c, d), onlap the flank of the mud mound complex (Fig. 3b) and are characterised by the common occurrence of pendant calcite cements (Fig. 12c).  showing crinoid ossicles and brachiopod shell fragments (below) with intergranular porosity filled by syntaxial and blocky calcite cements. e Photomicrograph of Lithofacies L14: molluscan wackestone to packstone. Note bivalve shells and foraminifers isoriented within the allomicrite matrix. f Outcrop photograph of Lithofacies L15: gigantoproductins packstone/rudstone L13 is overlain by 50 cm thick beds of molluscan wackestone to packstone (L14), with disarticulated and fragmented common bivalves, sparse sponge spicules, and brachiopods (Fig. 12e). Above, there are two metrescale beds of packstone/rudstone (L15), dominated by shells in life position of several species of the genus Gigantoproductus, alongside sparse crinoids and peloids (Fig. 12f). Also L14 and L15 onlap the mud mound complex (Figs. 3b, 5a). Figure 13 shows a summary of the lithofacies architecture and depositional model of the mud mound complex of Ricklow Quarry and the units underlying and overlying the buildup. Also, it provides a lithofacies correlation scheme between the mud mound complex and the log to the east of it described by Nolan et al. (2017, Ricklow Log; Fig. 3b).

Interpretation of depositional environments
Peloidal-skeletal packstone to grainstone (L1) was deposited before the onset of the mud mound complex in a moderate-energy, subtidal, open-marine inner-ramp setting, as suggested by the abundance of peloids, bioclasts, and endothyrid foraminifers (Gallagher 1988;Della Porta et al. 2005). Dark-coloured lithoclasts are interpreted as "black pebbles": pedogenised material reworked in the transgressive strata following subaerial exposures (Flügel 2004).
The overlying mud mound complex basal LA1 (L2, L3, L4) represents the initiation of a microbial-metazoan (bryozoans, siliceous sponges, brachiopods) community on the sea-floor, leading to the production of automicrite, in a moderate-energy middle-ramp environment. This depositional setting reconstruction is based on the absence of sedimentary structure evidence of wave action and a faunal assemblage with bryozoans, crinoids, and few trochospiral (Tetrataxis) and encrusting (Tuberitina) foraminifer morphologies, associated with Gallagher and Somerville's (2003) Biofacies 1: marine low-energy setting with a depth of more than 20 m, below fair-weather wave base. Also, siliceous sponges, whose presence is evidenced by common calcite-replaced spicules and nodules of early lithified automicrite embedding isoriented sponge spicules, representing calcified sponge bodies ( Fig. 6d; cf. tuberoids, Fritz 1958, Warnke 1995, are indicative of subtidal low-energy environments (Wilkinson and Trott 1985;Brunton and Dixon 1994;Cózar et al. 2019a). Hydrodynamic energy was, however, sufficient to sustain the suspension feeders and rework carbonate mud, resulting in the abundance of allomicrite in LA1, and to prevent vertical aggradation of automicrite boundstone beds. A moderate-energy setting is further supported by the low articulation ratio of brachiopods. Carniti et al. (2022) proposed moderate-energy, mesotrophic environment for LA1, based on the dominance in terms of biovolume of both large-sized spiriferides (specimens of species of Latibrachythyris up to 50 mm wide) and large-sized productides (e.g. Echinoconchus punctatus Sowerby, 1822) characterised by specimens up to 51 mm wide) in the Ricklow mud mound complex. Spiriferides had a complex feeding mechanism consisting of a spirolophe lophophore. According to Pérez-Huerta and Sheldon (2006), they were able to generate strong unidirectional inhalant currents allowing them to attain a large size in environments with high food supply (e.g. Angiolini 2007). The feeding mechanism of productides consisted of cilia attached to the dorsal valve interior and they were, thus, able to generate Fig. 13 Lithofacies scheme and depositional model for the mud mound complex and off-mud mound complex area. For a description of lithofacies L1-L15 see Table 1; for a description of lithofacies associations LA1-LA3 see Table 2; for a description of lithofacies in the Ricklow Log refer to Nolan et al. (2017) multi-directional currents and take food resources from a larger area around the shell, reaching a large size mostly in environments with few and dispersed food resources (Pérez-Huerta and Sheldon 2006).
Crinoidal grainstone/rudstone beds (L5) adjacent to LA1 were deposited in the same moderate-energy environment in areas dominated by crinoid communities, as indicated by the low amount of allomicrite and isorientation of unfragmented crinoid ossicles and stems.
The overlying mud mound complex core (LA2: L6, L7, L8, L9) and flank beds (LA3: L10, L11) were deposited in a similar subtidal middle-ramp environment, but with lower-energy than LA1. Decrease of hydrodynamic energy between the deposition of LA1 and that of LA2 and LA3 is supported by the presence in LA2 of metre-scale lensshaped automicrite units (L6, L7, L8), different from the tabular units of LA1. Lens-shaped units formed through lateral and vertical accretion decametre-scale lens-shaped mud mounds, associated to form a mud mound complex with some metre-relief over the sea-floor. Buildup relief during deposition is supported by the dip of the flank beds with respect to present-day-horizontal, up to 44°; as the regional dip is near to horizontal (Aitkenhead et al. 1985), flank beds dip is roughly close to the depositional dip with respect to the horizontal sea-floor. Onlap of the mud mound complex by the overlying beds (cf. Della Porta et al. 2002) strengthens this interpretation.
Decrease in hydrodynamic energy between the deposition of LA1 and LA2 is also supported by less allomicrite in LA2 than in the basal unit LA1. Furthermore, Carniti et al. (2022) suggested a decrease in hydrodynamic energy and thus food distribution during deposition of LA2 with respect to LA1, based on the dominance of productide brachiopods in terms of biovolume.
The flank beds are adjacent to crinoidal packstone/rudstone beds (L12). L12 shows the same texture, composition, and stratigraphic position of the peloidal intraclastic packstone beds with crinoids, bryozoans, ostracods, and brachiopods described in the basal 2 m thick part of the Ricklow Log by Nolan et al. (2017). Calcareous algae are absent from the mud mound complex core and little reworked debris is found in L12 (Nolan et al. 2017), indicating a dysphotic setting.
The mud mound complex in Ricklow, thus, developed in a middle-ramp, dysphotic environment with dispersed food resources, with low to moderate hydrodynamic energy level.
Crinoidal grainstone/rudstone (L13) beds onlapping the mud mound complex were deposited in a high-energy inner-ramp setting (Fig. 13). Good sorting, abundance of endolith microborings, and absence of allomicrite corroborate this interpretation. A cross-bedded 0.5 m thick lens of the same lithofacies occur in the Ricklow Log (Nolan et al. 2017) overlying the middle-ramp crinoidal packstone beds (L12; Fig. 13), and was interpreted by Nolan et al. (2017) as a subaqueous dune. Relative sea-level fall led to the demise of the mud mound complex growth; the partially to fully lithified buildup then provided the automicrite intraclasts in L13 (Fig. 13).
Widespread occurrence of pendant cements and other diagenetic features interpreted as due to vadose and phreatic meteoric diagenesis at the top of the mud mound complex core (LA2), in L13 both near the mud mound complex and in the Ricklow Log (Nolan et al. 2017), and minor occurrence of these features into LA1 and the lower portion of LA2 (Table 1), suggest that relative sea-level fall led to subaerial exposure of the study area after deposition of L13 (Figs. 5, 13).
After subaerial exposure, the molluscan wackestone to packstone (L14) was deposited in restricted lagoon environments with low-energy and variable salinity in-between the mud mound complexes of the intraplatform ramp (cf. Della Porta et al. 2002). This is indicated by the abundance of allomicrite and molluscs and the absence of stenohaline echinoderms and bryozoans (Nolan et al. 2017). After a stratigraphic gap, beds of the gigantoproductins packstone/ rudstone (L15) occur onlapping the mud mound complex. These were deposited in a low-energy, middle-ramp environment, as indicated by the absence of wave reworking and the presence of allomicrite. Both L14 and L15 are described in the Ricklow Log (Nolan et al. 2017;Angiolini et al. 2019) separated by a middle-ramp skeletal wackestone to packstone with brachiopods, ostracods, and crinoids ( Fig. 13; Nolan et al. 2017).

Initiation of the mud mound complex
The Ricklow Quarry mud mound complex deposition initiated during accommodation creation. This is a common feature of several Palaeozoic mud mounds (cf. Pickard 1992;Keupp et al. 1993;Aretz et al. 2012;Cózar et al. 2019a;Zhou and Pratt 2019;. Rapid accommodation creation decreases sedimentation rate, thus resulting in firmground formation: these two conditions are suitable for sea-floor colonisation by a suspension feeder benthic community consisting of bryozoans, brachiopods, and siliceous sponges. The metazoan community was probably associated with microbial mats, whose presence can be inferred from the mud textures indicative of microbially mediated in-situ precipitation (cf. Lees and Miller 1995;Monty 1995;Pratt 1995;Pickard 1996). Microbial communities were probably not forming widespread mats covering the sea-floor: otherwise, the presence of widespread microbial mats would have resulted in laminated automicrite deposits with few skeletal remains, e.g. stromatolite, as seen in modern hypersaline lakes (e.g. Dupraz et al. 2004). No evidence of stromatolite was found in any lithofacies association forming the Ricklow Quarry mud mound complex. It is likely that competition for space with the rich sessile metazoan community colonising the sea-floor and limited grazing activity of gastropods and trilobites, whose rare remains occur in thin sections, prevented widespread microbial mats formation. Instead, siliceous sponges likely hosted abundant and diverse microbial communities in the mesohyl (cf. Brunton and Dixon 1994), as is the case for modern demosponges (Vacelet and Donadey 1977;Wilkinson 1978aWilkinson , 1978bWilkinson et al. 1984;Reitner 1993;Tout et al. 2017). Also, decaying sponge tissues, as well as other metazoan tissues in the sediment, provided suitable organic substrates for microbial mats, leading to the initiation of microbially mediated mineralisation processes (cf. McConnaughey 1989;Frankel and Bazylinski 2003;Weiner and Dove 2003;Guido et al. 2013Guido et al. , 2022Lee and Riding 2020) and/or passive organomineralisation of sponge soft tissues (cf. Reitner 1993;Reitner et al. 1995aReitner et al. , 1995bWarnke 1995;Neuweiler and Burdige 2005;Neuweiler et al. 2007;Shen and Neuweiler 2018).
These processes led to the deposition of the tabular automicrite units of LA1, and later to the lens-shaped units of LA2. Bryozoans and brachiopods were not forming a skeletal framework, but their skeletons associated with the mineralised sponge bodies and automicrite formed an irregular stiff surface with up-right growth forms and depressions.
Mud mounds developed in various localities in the Derbyshire Carbonate Platform at the same stratigraphical level as the Ricklow Quarry mud mound complex, at the top of the Monsal Dale Limestone Fm (Gutteridge 1995). A literature review of the lithofacies distribution in the upper Monsal Dale Limestone Fm (e.g. Aitkenhead et al. 1985;Gutteridge 1989a;Harwood 2005) showed that the mud mounds developed aligned along the north-eastern margin of the platform, near the southern platform margin and in intraplatform inner-outer-ramp settings (Fig. 14).
The event of creation of accommodation leading to the initiation of the Ricklow mud mound complex might have been the trigger for growth of all mud mounds at the top of the Monsal Dale Limestone Fm, either related to upper Visean reactivation of extensional tectonics to the north of the Wales-Brabant High in northern England (Gawthorpe et al. 1989), and/or to eustatic sea-level rise. Mississippian sedimentation was governed by global sea-level changes, resulting in third-order sequences which might be synchronous worldwide (e.g. Ross and Ross 1985;Herbig 2016). Third-order sequence boundaries occurring in the lower and middle Brigantian (upper Visean) of Germany, Belgium and southwestern England (Herbig 2016 and references therein), might be coeval with the boundaries between D5b, D6a, and D6b transgressive-regressive cycles of Ramsbottom (1973Ramsbottom ( , 1977 in northern England. These cycle boundaries seem coeval with the upper boundaries of the Bee Low Limestone Fm and the Monsal Dale Limestone Fm, corresponding to regional subaerial exposure surfaces. The latest Visean was also a time of high-frequency fourth and fifth order glacioeustatic fluctuations (Horbury 1989;Smith and Read 2000;Wright and Vanstone 2001), with an inferred amplitude of 40-100 m , followed by the initiation of the Serpukhovian-middle Bashkirian glacial period (C 2 ) of the Late Palaeozoic Ice Age (Isbell et al. 2003;Fielding et al. 2008).
More data on the lithofacies underlying the mud mounds in the Monsal Dale Limestone Fm are necessary to clarify the role of glacioeustacy and local tectonics on the initiation of Derbyshire upper Visean mud mounds.

Mud mound complex growth: development of relief
A marked change in automicrite texture is observed between the basal LA1 unit and the LA2 mud mound complex core. While in LA1 automicrite beds (L3) have few primary cavities and microsparite content, automicrite in LA2 (L7, L9) is microsparite-rich and forms millimetre to decimetre-size patches with abundant radiaxial fibrous calcite cement and skeletal grains in-between (L8).
Greater accommodation and lower hydrodynamic energy during deposition of LA2 in comparison with LA1 allowed the mud mound complex core to grow both horizontally and vertically and to develop a convex-upward structure sustained by automicrite associated with bryozoans, brachiopods, and siliceous sponges. Automicrite precipitation led to early lithification of the mud mound complex core and then further vertical growth, resulting in metre-scale relief development over the sea-floor. Topographic relief over the sea-floor allowed water pumping by currents into the system of primary cavities in the mud mound complex core leading to the precipitation of radiaxial fibrous calcite cement (cf. Kendall 1985).
Bryozoan dominated areas resulted in deposition of the fenestellid bryozoan boundstone/cementstone lithofacies (L6); scarcity of automicrite in L6 is likely linked to scarcity of siliceous sponges. Other sponge-poor areas were the ones dominated by fenestellid bryozoans and brachiopods: absence of sponges led to little automicrite formation, leading to no vertical accretion, mud reworking and bryozoan fragmentation. These areas resulted in the deposition of the poorly bedded automicrite boundstone with skeletal packstone/rudstone patches (L9). . Reconstruction based upon regional data provided by Smith et al. (1967), Stevenson and Gaunt (1971), and Aitkenhead et al. (1985), integrated with data on the northern platform margin by Gawthorpe and Gutteridge (1990) and Harwood (2005), on the southern margin by Harwood (2005), on the Ashford Basin by Gutteridge (1987Gutteridge ( , 1989a, and on the area around Monyash by Gutteridge (1990Gutteridge ( , 1991Gutteridge ( , 1995 and Nolan et al. (2017). Data about the Asbian (upper Visean) margin of the Derbyshire Carbonate Platform from Harwood (2005) and Manifold et al. (2021). In the Monyash area, crinoidal grainstone shoals developed in innerramp settings overlie the mud mounds in the uppermost Monsal Dale Limestone Fm (Nolan et al. 2017) and in the Eyam Limestone Fm (Gutteridge 1989b;Nolan et al. 2017); they never occur underneath the mud mounds. The white box indicates the study area of Ricklow Quarry. FWWB Fair-weather wave base ▸

Brachiopod distribution
This study demonstrates that brachiopods are common skeletal components in all lithofacies of the mud mound complex at Ricklow, even in the automicrite-rich ones (L7, L9). The centimetre-to decimetre-size lenses of L8 with a richer brachiopod content in the mud mound complex core (LA2) might correspond to those interpreted by Gutteridge (1990) as "storm-scoured pockets" colonised by brachiopods. Gutteridge (1990) based this interpretation on the sharp edges of the pockets truncating structures in the mud mound complex core, concentration of pockets towards the top of the mud mound complex core, and marked asymmetry of their bases. However, no evidence of storm scouring associated with the brachiopod-rich lenses was detected both at the outcrop and thin section scale. Lenses of brachiopod concentrations recognised in this study only rarely show sharp bases, do not display the pocket morphology described by Gutteridge (1990) and are homogeneously distributed in the mud mound complex core.
A revised interpretation based on the data provided in this study is that the brachiopod-rich lenses represent in fact centimetre-to decimetre-size depressions developed in the irregular mud mound complex core surface and colonised by brachiopods and fenestellid bryozoans. Precipitation of marine fibrous calcite cement resulted in the deposition of the skeletal grainstone/rudstone lithofacies (L8) characterising the lenses. The presence of subrounded automicrite intraclasts and the transitional boundary with the surrounding automicrite lithofacies (L7) suggests that L8 is contemporaneous with the mud mound complex core formation. Also, brachiopod assemblages in the lenses, which represent in situ life assemblages (sensu Brenchley and Harper 1998;Carniti et al. 2022) are the same as those occurring in association with the automicrite (L7, L9). This interpretation might apply also to other macrofaunal "pockets" in Mississippian mud mounds, e.g. the Tournaisian Waulsortian mud mounds in the UK (Lees 1964) andUSA (De Keyser 1978), and the upper Visean (Asbian) bioherms on the Derbyshire Carbonate Platform margin (Parkinson 1952).
The substrate variety in the mud mound complex sustained the high brachiopod diversity of the fauna (45 species, 36 genera, 7 orders; Table 3; Carniti et al. 2022). The mud mound complex surface likely was the site for both hard (bryozoan fronds, siliceous sponges, brachiopod shells) and soft (allomicrite) substrates, allowing colonisation of both pedicle-attached rhynchonellides, spiriferides, and terebratulides and seminfaunal concavo-convex productides. The surface depressions were likely the site of allomicrite concentration at their bases. This resulted in patchy soft substrates suitable for the colonisation by seminfaunal concavo-convex productides, whose shells were later colonised by pedicleattached rhynchonellides, spiriferides, and terebratulides (taphonomic feedback).
Even though the role of brachiopods was limited to that of dwellers (sensu Fagerstrom 1988) in the mud mound complex, at Ricklow they were not limited to "pocket-like" concentrations, but were widespread. Brachiopods were a dominant skeletal component alongside fenestellid bryozoans and siliceous sponges, with which they were likely competing for space and resources (Carniti et al. 2022). The environment of the Ricklow mud mound complex had dispersed food resources (Carniti et al. 2022): the presence of abundant decaying organisms (siliceous sponges) and microbial mats likely resulted in more organic material in suspension on the mud mound surface with respect to off-mud mound complex, brachiopod-poor areas. This, in addition to scarcity of crinoids (which can access resources higher than brachiopods in the water column) and diverse hard and soft substrates enhanced brachiopod colonisation of the mud mound complex. Brachiopod shell abundance in turn provided hard substrates sustaining pedicle-attached brachiopods, bryozoans, sponges, and microbial mats (taphonomic feedback), contributing to and influencing mud mound growth and stability.

Comparison with other brachiopod-rich buildups
The mud mound complex at Ricklow Quarry differs from the Tournaisian Waulsortian mud mounds of Western Europe and USA (e.g. Somerville et al. 1992;Lees and Miller 1995) as it lacks any structured distribution of different mud generations (polymuds) and has a less extensive cavity system. Also, the Ricklow mud mound complex is dominated by clotted peloidal micrite, forming roughly 40% of the core volume, while Waulsortian mud mounds seem to be commonly dominated by "non-grumous micrite" (leiolite and allomicrite; Lees and Miller 1985;Devuyst and Lees 2001). The Ricklow mud mound complex also differs in skeletal assemblage: brachiopods in Waulsortian mud mounds are diverse (e.g. de Koninck 1887;Fraipont 1908;Demanet 1923Demanet , 1958Mottequin and Simon 2017;Mottequin 2021), but few in all phases Miller 1985, 1995), apparently always being less important than fenestellid bryozoans and crinoids.
The Ricklow mud mound complex markedly differs from Visean mud mounds consisting of encrusting trepostome bryozoans, serpulid, problematic worm tubes, and abundant brachiopods into automicrite ("microthrombolite" Bridges et al. 1995) classified as Type 5 buildups by Bridges et al. (1995) in having a much more diversified brachiopod fauna and no syndepositional mineralisation in the core. These mud mounds occur in eastern Canada (e.g. James 1987, 1989;Boehner 1989) and were developed in hydrothermal-seep vents related settings (von Bitter et al. 1990(von Bitter et al. , 1992 at depths ranging from greater than 100 m to highenergy shallow waters (Webb 2002).
The Ricklow mud mound complex also differs from brachiopod-rich buildups with a framework built by calcimicrobes and fenestellid bryozoans occurring in the Lives Formation of north-eastern Belgium (Lauwers 1992;Chevalier and Aretz 2005), which developed in an inner-ramp setting, in lacking any calcimicrobe framework and having a more diverse skeletal composition and brachiopod fauna.
The Ricklow mud mound complex is similar to metrescale brachiopod-bryozoan-ostracod mud mounds with micropeloidal mud textures developed on a carbonate ramp to the south of the Askrigg Block, northern Yorkshire, in the Holkerian-lower Asbian (middle-upper Visean), and in the flank facies of the Cracoean reefs developed on the same site in the upper Asbian (upper Visean) on the margin of a flat-topped carbonate platform (Mundy 2000;Waters et al. 2017). Also, the Ricklow mud mound complex is similar to the "bank facies" dominating the core of Cracoean reefs in Yorkshire (Mundy 1994) and Derbyshire (Harwood 2005), which consists of wackestone/floatstone with clotted micritic matrix hosting abundant brachiopods and corals alongside oncoids and dasycladacean algae. The Cracoean reefs show minor development in some areas (Waters et al. 2017) of a "framework facies" consisting of encrusting calcimicrobes, encrusting bryozoans, lithistid sponges, corals, shellattached brachiopods, bivalves, and crinoids (Mundy 1994). Other buildups similar to Cracoean reefs occur along carbonate platform margins in the upper Visean of north-western Ireland (Somerville et al. 2009), Belgium (Muchez and Peeters 1987;Mottequin and Poty 2022), Turkey (Denayer and Aretz 2012), eastern Australia (Webb 1999), and in the upper Serpukhovian of Arkansas (Webb 1987), all of which host abundant and diverse brachiopod faunas.
Conversely, the Ricklow mud mound complex markedly differs from calcimicrobe-coral-lithistid sponge reefs in the Mississippian of Queensland, Australia Webb 2005, 2008), microbial-bryozoan-coral reefs in the 9 Page 32 of 39 upper Visean of Wales (Aretz and Herbig 2003), and microbial-sponge-bryozoan-coral reefs in the upper Visean of north-eastern Morocco (Aretz and Herbig 2008) in lacking any framework built by calcimicrobes, thrombolite/stromatolite (sensu Riding 2000) and/or corals, in having developed in deeper dysphotic settings and in having an abundant fauna of brachiopods.
Upper Visean Derbyshire mud mounds including the complex at Ricklow Quarry were considered as Type 3 crinoid-brachiopod-fenestellid bryozoan buildups by Bridges et al. (1995). Data provided in this study show that crinoids are not common in the mud mound complex core at Ricklow, while brachiopods and siliceous sponges are abundant and widespread in every lithofacies in the complex. A distinct sub-type of Type 3 buildups should here be defined to include the Ricklow mud mound complex, and likely other upper Visean Derbyshire mud mounds, characterised by abundant fenestellid bryozoans, brachiopods, and siliceous sponge spicules, few crinoids in the core, abundant clotted peloidal mud textures and no calcimicrobes. Other mud mounds referable to this newly defined sub-type developed around or below fair-weather wave base in the upper Visean (Brigantian) of the Midland Valley of Scotland (Shiells and Penn 1971;Brown 1977;Jameson 1980Jameson , 1987, and the upper Visean-Serpukhovian of the southern Montagne Noire, France (Poty et al. 2002;Vachard et al. 2017;Cózar et al. 2019a). Also referable to this sub-type they could be the brachiopod-bryozoan-ostracod mud mounds in the middle-upper Visean of Yorkshire (Waters et al. 2017). Conversely, other mud mounds considered as Type 3 buildups by Bridges et al. (1995) from the Visean of Ireland (Kelly and Somerville 1992), Scotland (Pickard 1992), and USA (Brown and Dodd 1990) differ from the Ricklow mud mound complex newly defined sub-type in having a different skeletal assemblage with few brachiopods, and usually abundant crinoids in the core.
In conclusion, brachiopods are a common skeletal constituent of Mississippian reefs, becoming increasingly important in middle-upper Visean buildups developed below fair-weather wave base in carbonate platform settings. Widespread development of buildups in platform settings in the middle-late Visean  coincided, at least in Western Europe and North America (southern margin of Laurussia), with the evolution of Tournaisian ramps into flattopped carbonate platforms (Gawthorpe 1986;Gawthorpe et al. 1989). Brachiopods are a dominant constituent of at least three types of Visean buildups in shallow platform settings: (1) calcimicrobe-bryozoan reefs (Chevalier and Aretz 2005), (2) Cracoean reefs (e.g. Mundy 1994), (3) fenestellid bryozoan-brachiopod-siliceous sponge mud mounds with common clotted peloidal micrite, a newly defined sub-type of Type 3 of Bridges et al. (1995), of which the Ricklow mud mound complex is a representative. Water depth control on the abundance of brachiopods in Mississippian buildups is supported by the scarcity of brachiopods in middle-upper Visean mud mounds developed in middle-outer-ramp and basin settings such as in north-western Ireland (Kelly and Somerville 1992), the Béchar Basin in Algeria (Madi et al. 1996), and the eastern anti-Atlas in Morocco (Wendt et al. 2001). Also, the study by Cózar et al. (2019a) on the ecological distribution of skeletal grains in Serpukhovian buildups from the southern Montagne Noire in France, showed that brachiopod abundance increases with shallowing settings.
Substrate also played a role in controlling brachiopod colonisation of Mississippian carbonate buildups, as these were providing depressions in their irregular surface, muddy facies as well as firm, early cemented substrates more suitable for successful colonisation by brachiopods with respect to packstone to grainstone mobile substrates in the surrounding platform interior settings (cf. Webb 2001, for coral mud mound colonisation).
Occurrence of microbial mats and decaying siliceous sponge tissues in the buildups (cf. Brunton and Dixon 1994) likely provided also more abundant food resources than in the surrounding areas. In upper Visean Cracoean reefs and fenestellid bryozoan-brachiopod-siliceous sponge mud mounds productide brachiopods seem to dominate the fauna (Webb 1987;Brunton and Mundy 1988;Brunton and Tilsley 1991;Mottequin and Poty 2022;Carniti et al. 2022). This might be indicative of dispersed food resources (Pérez-Huerta and Sheldon 2006), a condition that markedly differs from the eutrophic settings related to upwelling currents where Waulsortian mud mounds developed in the late Tournaisian Miller 1985, 1995).
Hence, it was probably a combination of relative shallow depths between fair-weather and storm wave base, moderate-energy, diverse substrate types, and distribution of food resources that favoured the brachiopod colonisation and in turn the structuring of brachiopod-rich buildups in the middle-upper Visean.

Conclusions
Lithofacies and brachiopod faunal distribution analysis of the upper Visean (Brigantian) mud mound complex of Ricklow Quarry, developed on the Derbyshire Carbonate Platform at the top of the Monsal Dale Limestone Fm led to a reinterpretation of its lithofacies architecture and depositional model. The Ricklow mud mound complex originated during a period of creation of accommodation in an intraplatform middle-ramp dysphotic environment with dispersed food resources. The mud mound complex growth was interrupted by a subaerial exposure due to a relative sea-level fall.
Brachiopods are not, as previously reported, limited to storm-scoured isolated pockets (Gutteridge 1990(Gutteridge , 1995, and are widespread in every lithofacies of the mud mound complex of Ricklow. Brachiopod abundance provided hard substrates for other sessile organisms and contributed to mud mound growth. Brachiopod-richer lenses do exist in the mud mound complex core and are reinterpreted as the result of brachiopod colonisation of depressions in the irregular sea floor surface of the mud mound complex core. The Ricklow mud mound complex represents a newly defined middle-upper Visean sub-type of Bridges et al. (1995) Type 3 buildups: fenestellid bryozoan-brachiopod-siliceous sponge mud mounds. These mud mounds developed in Western European platform settings below fair-weather wave base, in environments with dispersed food resources.
This study demonstrates the important role of brachiopods in Mississippian buildups, and the importance of characterising their distribution and diversity to enhance understanding of Mississippian reef diversity, evolution, and depositional environments.
Acknowledgements The authors are grateful to Martina Bruno, who contributed to the fieldwork through her master thesis project, Curzio Malinverno and Matteo Pegoraro (University of Milan) for preparing thin sections for petrographic analysis, the British Geological Survey staff for hosting APC in their facilities and for sample shipping to Italy, and Natural England for study permits for the Ricklow Quarry site, part of the Lathkill Dale SSSI. Markus Aretz and an anonymous reviewer are thanked for their valuable comments, which have led to improvements in the manuscript.
Funding Open access funding provided by Università degli Studi di Milano within the CRUI-CARE Agreement. Financial support was provided by PhD grant from University of Milan and MUR Italian Ministry of University.

Data availability
The materials used for this study are housed in the collection of the Museum of Palaeontology of the Department of Earth Sciences "A. Desio", Università degli Studi di Milano, Milan, Italy, and are labelled MPUM12295.

Conflict of interest
There are no conflicts of interest and data are available on request to the corresponding author.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.