Coal Properties & Paleoenvironments

The study focuses on New Zealand bituminous coal, primarily located on the West Coast of the South Island, within Westland and Northwest Nelson. The main coal measures investigated are the Paparoa Coal Measures (CM), spanning the uppermost Cretaceous to lowermost Tertiary (Haumurian-Teurian), and the Brunner CM, largely from the Eocene epoch. The Paparoa and Brunner Coal Measures are superimposed in the Greymouth and Pike River Coalfields; however, elsewhere, the Paparoa CM are absent, with the Brunner CM resting on Middle Cretaceous terrestrial sediments or older granite and greywacke rocks. Laird's 1968 research proposed an extensional (rift) basin system for Middle Cretaceous terrestrial sequences (Pororari Group), a model relevant to the Paparoa CM. This current research highlights the significant influence of syndepositional faulting on the highly variable paleogeography, sedimentary facies, and coal seam characteristics of the Paparoa CM. The Paparoa Tectonic Zone is notable for being flanked by lower-rank Brunner Coal Measures, ranging from lignite to high-volatile bituminous C.

Prior research, particularly Gage's 1952 work, subdivided the Paparoa CM into seven formations with alternating fluvial and lacustrine characteristics, highlighting lateral thickness and lithofacies changes. Detailed geological maps (1 inch = 0.25 miles) were created, offering valuable structural information and seam distribution details for coal exploration and mine development. Earlier interpretations of the Paparoa Coal Measures at Greymouth are expanded upon, linking them to the Pike River Coalfield. The Pike River Paparoa Coal Measures are considered to represent more proximal sedimentation in a shared, narrow fault-bounded basin. This research proposes that syndepositional differential subsidence localized major fluvial channels, creating conditions more favorable for thick, clean peat accumulation than uniform basin subsidence would allow. The study aims to understand the influence of paleoenvironmental factors on peat accumulation and coal characteristics, using an integrated approach combining coal measure sedimentology and coal type analysis.

Exploration at Pike River Coalfield by Otter Minerals (later Pike River Coal Company), starting around 1979/80, significantly contributed to the research. Although initially focused on the Brunner CM, substantial mapping, sampling, and analysis of both Paparoa and Brunner coals occurred. Close collaboration with the company facilitated studies of lateral sedimentary facies changes at Pike River (Newman, J. and Newman, N.A., 1981; Newman, J., 1982a), supporting tectono-sedimentary models for Greymouth. These studies involved extensive analysis of the petrology of Pike River coals, including vitrinite reflectance and maceral composition. The research suggests that the Pike River Coalfield represents a more proximal sedimentation environment within a narrow, fault-bounded basin shared with the Greymouth Coalfield. The study hypothesizes that syndepositional differential subsidence was crucial in localizing major fluvial channels to areas of rapid subsidence, which led to the accumulation of thicker, cleaner peat deposits than would have been expected under uniform subsidence.

The study area's physiography is characterized by elevated, rugged terrain (over 1000m) due to the Kaikoura Orogeny (Late Miocene to Recent), which uplifted the Paparoa and Papahaua Ranges. Highest rank coals (high volatile bituminous A to semi-anthracite) are typically found in these high-country areas. While coking coal likely existed continuously between Greymouth and Buller Coalfields initially, post-uplift erosion has left significant remnants only at Pike River. The Paparoa Tectonic Zone is flanked by lower-rank Brunner CM coals (lignite to high-volatile bituminous C). Laird (1968) suggested an extensional basin system for the Middle Cretaceous Pororari Group, a model relevant to the Paparoa CM. This research provides evidence that syndepositional, likely normal, faulting significantly influenced the highly variable paleogeography, sedimentary facies, and coal seam characteristics of the Paparoa CM. The basin's structural controls, oriented northwest-southeast in early Paparoa sedimentation, differed from the north-northeast trending margin that dominated later sedimentation. The study challenges previous interpretations, particularly Gage's (1952) isopachs, suggesting a more gradual termination of basin subsidence. The study doesn't support post-depositional denudation as a major factor in shaping the basin's development.

The Greymouth Coalfield is described as having thick coal measures, exhibiting diverse sedimentary facies and complex stratigraphy. Well-developed seams of low-ash, low-sulfur bituminous coal are present at several horizons. The complex depositional history of the coal measures was expected to have resulted in a variety of paleoenvironments and diversity in seam characteristics, including geometry, mineral matter content, and coal type. The core research aim was to understand how paleoenvironmental factors influenced peat accumulation and the resulting coal character. An integrated study of coal measure sedimentology and coal type was chosen as the best approach to achieve this aim. This involved analyzing the relationships between lithological characteristics (member thickness, texture, composition, and coal abundance), seam thickness and coal quality, and coal type and its analytical properties.

The Paparoa Coal Measures at Greymouth are detailed, showing a complex stratigraphy and diverse sedimentary facies. Gage's 1952 work divided the Paparoa CM into seven formations of alternating fluvial and lacustrine character, recognizing significant lateral changes in thickness and lithofacies. Extensive mapping, using data from drillholes and mines, resulted in detailed geological maps (1 inch = 0.25 miles), individually mapping each formation. This provided structural information and seam distribution details crucial for coal exploration. The study re-examined existing data and models relating coal measure sedimentation to basin subsidence (Newman, J., 1981), using this foundation to further investigate the paleoenvironmental influences on coal formation within the Greymouth Coalfield. The research noted the presence of a southwestern 'basement high' area which experienced the least subsidence during Rewanui Member accumulation, and which is notable for abundant thick coal seams, indicating extensive, long-lived swamps. This suggests that restricted competition from fluvial sedimentation favored peat accumulation in this region.

The research utilized various data sources, including geophysical and lithological logs from drillholes (e.g., drillholes 620, 625, 627, 630), to understand the basin's development and sediment distribution. The study highlights the inconsistencies in data from different sources and across time, especially regarding terminology used by earlier and later workers (e.g., fine and coarse sand being termed silt and grit, respectively). This inconsistency presented challenges in identifying subtle facies changes. Analysis of geophysical logs, particularly gamma logs, revealed an unusually high proportion of mudstone (>60%) in some drillholes (e.g., drillholes 620, 625, 627), which was difficult to explain using simple tectonic and paleogeographic models. The comparison of Drillhole 630 (with a continuous core) to nearby drillholes (e.g., 625, 620) highlighted that occasional sandstones had anomalously high gamma traces, indistinguishable from mudstone, suggesting limitations in log interpretation. The study emphasizes the importance of carefully considering data reliability, especially when using older logs with limited sample availability.

The Rewanui Member is a key coal-bearing unit, thicker and more extensive than underlying members. Its sediments overlap the Waiomo Member onto basement in the southwest and south, thinning toward basin margins. The northwest is dominated by coarse greywacke conglomerates, well-exposed at Twelve Mile Beach. The basin center contains quartzofeldspathic sandstones, with coal seams thickest near the top and bottom of the member. Paleoflow analysis, although limited, suggests southward flow in the Seven Mile Creek area, based on planar foresets and trough axis orientations in current-bedded sandstone. The progressive increase in conglomerate abundance and clast size towards the northwest indicates a northwest-to-southeast paleo-slope and source. The significant lateral variations in thickness of the Rewanui Member, and subsequent members (Goldlight and Dunollie), are attributed to differential subsidence during basin development, rather than solely differential compaction. The presence of rounded coal pebbles in basal Rewanui beds at Twelve Mile Beach and Ten Mile Creek supports the interpretation of contemporaneous erosion and redeposition of pre-existing peat.

The Pike River Coalfield is presented as a key area for studying coal formation, described as representing more proximal sedimentation within a narrow, fault-bounded basin. This basin is shared with the Greymouth Coalfield. The research emphasizes the role of syndepositional differential subsidence in localizing major fluvial channels to zones of most rapid subsidence. This complex paleogeography, resulting from sedimentation in a tectonically active basin, is argued to have been locally more favorable for the accumulation of thick, clean peat than a uniformly subsiding basin. The commencement of exploration activity at Pike River Coalfield by Otter Minerals (later Pike River Coal Company) in 1979/80 is noted as a significant development, providing substantial data through mapping, sampling, and analysis of both Paparoa and Brunner coals. Close cooperation with the company greatly enhanced the research, particularly in studies of lateral sedimentary facies changes which supported previously postulated tectono-sedimentary models for Greymouth.

Detailed petrological studies of Pike River coals were conducted, focusing on vitrinite reflectance and maceral composition. The study notes that Pike River coals exhibit slight anisotropy, with Ro max measured for analysis. Only relatively thick telocollinite (>50µm) was used for reflectance measurements; the inclusion of desmocollinite and thin telocollinite would likely have resulted in significantly lower Ro max values. Reflectance values were often diffusely distributed across seams, though differences between certain seams were distinct. The research found that while some Paparoa seams showed Ro max >0.90%, most had substantially lower reflectance than previously reported by Robertson Research. The presence of Paparoa seams with low Ro max, similar to Brunner coal, underlying M4 seams with much higher Ro max indicates that reflectance variation cannot be solely explained by varying burial depth. This challenges previous interpretations by Robertson Research.

The study addresses the causes of anomalously low vitrinite reflectance observed in some Pike River coals. While alginite can depress vitrinite reflectance, its absence in Paparoa coals at Pike River makes this explanation unlikely. Although liptodetrinite (degraded exinitic debris) could be present in the Brunner coal, it was deemed unlikely to be abundant. Another possibility, the coalification of unusually lipid-rich plant tissues, was considered, but this was also found to not be dominant. The fact that the reflectance of several telocollinite varieties appeared to follow interseam reflectance variations suggested that telocollinite reflectances are not directly dependent on the primary composition of precursor plant tissues. The conclusion drawn is that very poor peat drainage, resulting in oxygen deficiency, was the primary cause of the low vitrinite reflectance observed in Brunner and Paparoa M3 coals.

The study compares the paleoenvironmental conditions of Brunner and Paparoa swamps at Pike River. Brunner swamps are hypothesized to have been relatively warm and brackish, promoting high bacterial activity in the peat. Combined with oxygen deficiency, this would have produced extremely perhydrous vitrinites. The vegetation in these swamps was considered to be predominantly a hydrogen-rich “reed swamp” type, further contributing to the low reflectance. In contrast, the intermontane Paparoa swamps were likely relatively cool and acidic. Low reflectance in basal Paparoa coals at Pike River is attributed primarily to oxygen deficiency due to very wet conditions and potentially a cellulose-rich and/or lipid-rich flora. The influence of underlying alkaline volcanics on swamp chemistry through calcium enrichment was deemed insignificant because common indicators of such enrichment (high organic sulfur, syngenetic pyrite, extreme degradation of plant tissues, abundance of carbonates) were absent from these coals. Many Paparoa coals exhibiting perhydrous properties are therefore regarded as having abnormally low reflectance due to oxygen deficiency rather than other factors.

A key finding of the research is the linear inverse relationship observed between coal yield and vitrinite reflectance in isorank samples (samples of equal thermal maturity). This relationship is attributed to variability in peat oxygenation during coal formation. The research carefully defines coal rank as thermal maturity, emphasizing the difficulty in ensuring coals are of the same rank unless they are from a common seam intersection (serial samples). In cases where samples aren't from the same seam, relative rank can often be estimated, for example by assuming coal rank decreases upwards in a section where fault repetition is absent. Understanding this relationship between yield and reflectance is crucial for interpreting the paleoenvironmental conditions influencing coal formation and for making accurate predictions about coal properties.

The study utilized vitrinite reflectance measurements as a primary analytical technique. Pike River coals, exhibiting slight anisotropy, had their Ro max (maximum reflectance) measured. The methodology emphasizes the selective use of only relatively thick telocollinite (>50µm) for reflectance measurement to ensure accurate results. The researchers acknowledge that including desmocollinite and thin telocollinite would have likely resulted in significantly lower Ro max values for most samples. The variation in reflectance values across seams is noted, with many seams showing a diffuse distribution of values and weakly defined peaks, although clear differences existed between certain seams. The detailed approach to data collection and analysis is essential for drawing reliable conclusions about the relationship between vitrinite reflectance and other coal properties.

The research addresses the causes of anomalously low vitrinite reflectance in certain samples. While alginite in coal can depress reflectance, its absence in Paparoa coals at Pike River eliminates this factor. Although liptodetrinite (degraded exinitic debris) could theoretically be present in the Brunner coal, its abundance is deemed unlikely. The possibility of unusually lipid-rich plant tissues influencing vitrinite reflectance is discussed but is not considered the primary factor. The observation that the reflectance of several telocollinite varieties follows the interseam reflectance variations further suggests that telocollinite reflectance is not primarily determined by the original composition of precursor plant tissues. The study concludes that poor peat drainage, leading to oxygen deficiency, was the primary cause of low vitrinite reflectance in Brunner and Paparoa M3 coals. The implications of this finding are considered in relation to the broader paleoenvironmental context of coal swamp development.

The Rapahoe Sector is characterized by relatively uniform coal types based on traditional maceral analysis of Upper Rewanui coals: approximately 80% vitrinite, with inertinite and exinite ranging from 3% to 10%. However, studies of correlatives in the Pike River Coalfield reveal that maceral textures, sub-varieties, and associations are crucial for differentiating coal types and understanding their genesis. Two coals with similar maceral proportions can be petrographically and chemically distinct. The available ACIRL maceral analyses are considered of limited use for differentiating coal types in the Rapahoe Sector, particularly lacking values for suberinite (resin-impregnated cells). The use of ACIRL and CRA analyses, with noted discrepancies in volatile matter values (up to 6% lower from ACIRL prior to 1984), highlights the importance of considering analytical methodology and potential inconsistencies when interpreting data from different sources. The varying analytical approaches emphasize the need for detailed petrographic examination to fully characterize coal types and their associated paleoenvironmental significance.

Significant challenges are presented in correlating drillholes reliably within the Rapahoe Sector due to stratigraphic variability. The existing drillhole spacing is too wide for dependable correlation. The mineral matter content of seams is influenced by multiple source areas and complex drainage patterns, leading to significant variations in geophysical profiles over short distances. These variations reduce the usefulness of geophysical logs for correlation. The difficulty in extrapolating from mined seams to drillholes, even over short distances (400m), despite some seams extending over 1500m (e.g., C seam at Strongman Mine), underscores the complexities of stratigraphic interpretation. Thorburn's (1981) Strongman isopachs illustrate rapid changes in seam thickness, attributed to streams draining through the swamp and causing contemporaneous sedimentation. These challenges necessitate a more detailed approach to understanding the paleoenvironmental context of coal accumulation.

The study proposes a revised model for Upper Rewanui peat accumulation in the Strongman Mine area. This replaces the previous lake-margin model with a model focusing on a meandering river flowing down the basin axis, east of the Rapahoe Sector. The existence of a major axial fluvial system carrying granitic sediments southwards during Rewanui sedimentation is discussed, along with its influence on peat accumulation. The transition from a braided fluvial regime (middle Rewanui) to a meandering regime (upper Rewanui), associated with a rise in base level, is linked to the onset of thick peat accumulation. This meandering river, with associated levees, is hypothesized to have caused drainage ponding in the northwest, where sedimentation rates had declined due to lowered source area relief, promoting swamp development. Peat accumulation was constrained by persistent sedimentation in the northwestern alluvial fan-toe region and by overbank sedimentation near the meandering river. Small streams fed by the alluvial fan occasionally interrupted peat accumulation within the swamp. The model suggests peat accumulation ended when westward migration of the meandering river led to burial of the peat by overbank mudstones and siltstones.

Coal samples were obtained from both outcrop and drillholes, primarily from CRA (presumably a coal analysis company). Most outcrop samples showed moderate to severe weathering, requiring cautious interpretation of data. Analyses were primarily limited to proximate analysis, sulfur content, calorific value, and Crucible Swelling Index (CSI) for weathered samples. A 6-hole drilling program in the eastern dip-slope region of the coalfield in early 1983 yielded cores of fresh coal. The Pike River Coal Company initially subdivided these cores into plies (10cm to 1m thick) for ash, sulfur, and CSI determination. Subsequently, plies were amalgamated into composites (2-4 per drillhole) for more comprehensive proximate, ultimate, and ash constituent analyses and fluidity tests. The differing analytical approaches and sample conditions highlight the challenges inherent in coal analysis and the importance of considering sample quality and preparation.

The study analyzes the properties of both the main seam and a rider seam present in many drillholes (Drillholes 2, 4, & 6). The rider seam properties sometimes differed from the main seam. In Drillholes 1 and 3 the rider seam was very thin and dirty and not sampled. Drillhole 2's rider seam is characterized by low exinite and vitrodetrinite content and bimodal reflectance (0.98% and 0.60%, averaging 0.78%). A portion resembled Webb/Baynes coal from the Buller Coalfield. Drillhole 6's rider seam also had relatively low exinite and vitrodetrinite compared to the main seam, showing unimodal reflectance (0.66%). The rider seam in Drillhole 4 was generally similar to the main seam. Drillhole 5's rider seam was not recovered but geophysical logs suggest it was moderately clean. These variations demonstrate significant heterogeneity within the coal seams and emphasize the importance of detailed analysis to understand the factors influencing coal quality.

The study investigated sulfur enrichment within the coal seams, specifically in the Brunner seam at Pike River. A typical sulfur gradient is observed, possibly resulting from both secondary enrichment through permeation from overlying sediments and a progressive increase in marine influence during swamp development. The relatively impermeable muddy parting and rider seam above the main seam likely inhibited secondary sulfur enrichment, consistent with relatively low whole-seam sulfur values in several drillholes (Drillholes 1, 2, 3, 4, and 6). The abrupt sulfur enrichment observed in uppermost plies in most drillholes probably represents syndepositional or immediately post-depositional enrichment under brackish conditions, related to the development of a lagoon that interrupted peat accumulation. The intense burrowing of roof coal by marine organisms suggests exposure of the peat to seawater during this period of limited sediment supply, potentially due to subdued source area relief. This points to the importance of considering both diagenetic and depositional factors when assessing coal seam properties.

Carbonization properties, specifically fluidity, were assessed using drillhole samples. Fluidity ranged from 45,000 to over 50,000 ddm (dial divisions per minute) for low-to-moderate ash coals; high-ash coals weren't tested. The upper limit of the CRA equipment (50,000 ddm) prevented precise measurement of fluidity in some samples, making trend analysis difficult. Both fluidity and the temperature range from softening to resolidification are provided. Fluidity is found to generally increase from bottom to top of the main seam in all drillholes. The study emphasizes that while carbonization behavior depends on coal rank, it is also strongly influenced by coal type. For West Coast coals, vitrinite chemistry is considered the most important variable, showing substantial variation between different coals of the same rank, as indicated by variations in volatile matter, hydrogen, vitrinite reflectance, and fluidity. These variations are not directly correlated with overall maceral group proportions and are linked primarily to variations in vitrinite chemistry itself. The challenges encountered in fluidity measurement highlight the limitations of analytical techniques and the complexities of coal characterization.