Earthmoving and pond civil works in the Catamarca puna
A lithium pond is, first of all, a million cubic metres of soil moved at 4,100 metres. Tres Quebradas' contractor publishes the complete bill of quantities and there is the market with nobody's estimates: for each 200-hectare stage, 2.1 million m² of crust to break, 662,800 m³ of borrow excavation, 242,300 m³ of berms and 420,500 m³ of floor, in eight months that include winter, with winds of up to 120 km/h and temperatures of −30 °C. And here the diagnosis flips versus the rest of the portfolio: this category does exist in Catamarca —31 construction companies and 12 equipment rental firms in the provincial registry— and there is a joint venture of three Catamarca companies executing 60 kilometres of road above 4,000 metres. The problem is not capability: it is that both pond packages tendered went to the same joint venture from outside, and the only open and proven door is to enter as a partner.
The TAM is activity, not capturable loot — but here the split has a peculiarity: a cubic metre is not imported. Zijin's Chinese chain brought the geomembrane from China and, to move the soil, hired an Argentine joint venture. So what is captive is not foreign trade: it is the contracting structure — what the head of the contract executes with its own fleet and what the operator prefers to buy as equipment rather than contract. Your gap is the items the lead releases (crust breaking, roads, rental with an operator), operating maintenance and the complete package if you enter via a joint venture.
This niche has two rules pushing it and one that decides its margin. Pushing: the local-content commitments written into acts of State: Hombre Muerto Oeste directs 95.34% of the investment amount to suppliers —the highest in the whole RIGI— and Fénix 60%, three times the regime's minimum; and this category is the easiest to attribute against those commitments, because the labour, the equipment and the borrow material are in the province. The key to collecting from that amount is the Re.P.E.M., which requires two years of domicile, half the partners local and 70% of the payroll from Catamarca — and admits the joint venture with a local partner as a shortcut, with two proven precedents. And the margin is decided by Tax Law 5927: the same contract pays 2.5% of Turnover tax if it is classified as construction and up to 4.8% if it is classified as a mining support service, plus Stamp tax of 1.5% or 2% depending on the subsection — and the classification is chosen at registration, not at invoicing. The ones below open in the reforms panel on the home page, with their status and primary source.
enablesRe.P.E.M.: Catamarca's mining local-content rule is a resolution, not a lawIt is the counter that turns the local-content commitment into a contract: with no Re.P.E.M. you do not collect from Hombre Muerto Oeste's 95.34% or Fénix's 60%. Its roots requirements (2 years of domicile, 50% local partners, 70% Catamarca payroll, semi-annual sworn statement) are a barrier for the outsider and an asset for whoever is already established — and here it weighs differently than in the rest of the portfolio, because in this category there are indeed 31 registered Catamarca construction companies that could use it and are not monetizing it in ponds. The joint venture with at least 50% local suppliers is the legal vehicle, with two documented precedents: Huasi Construcciones on Sal de Vida's USD 130 M EPC and the Mogetta-Huasi-Italca joint venture on Section I of RP 43.see the reform →This market does not float on its own: concrete megaprojects drive it. These are the ones moving demand for this niche — each with its investment and status.
Stage 2 of Tres Quebradas: doubles capacity by adding 40,000 t/year of new output (all stages combined target 60,000-80,000 tpa, projected)…
see the project →Lithium brine project in the Salar del Hombre Muerto (western zone), to produce 12,000 t/yr of lithium carbonate equivalent…
see the project →Expansion (Phase 1B) of the historic Fénix lithium-carbonate project in the Salar del Hombre Muerto, operated by Rio Tinto via Minera del Altiplano…
see the project →Who splits the market, where you get in, what pays and what could break it.
already in
split
The two halite pond stages of Tres Quebradas for LIEX S.A. (Zijin): 200 ha executed + 200 ha under way, with a published bill of quantities (4.18 M m² of crust breaking, 662,800 m³ of borrow, 404,300 m³ of berms, 794,500 m³ of floor and base course) at 4,100+ masl and an 8-month schedule including winter. It is the head of the contract: earthmoving IS its scope, not a sub-scope it subcontracts entirely — what it releases are items (crust breaking, roads, equipment rental). And one detail that matters: ITALCA also appears in the Catamarca joint venture on RP 43.
Civil works for the Fénix expansion, phases 1 and 2, status 'Executed' in the Catamarca puna, with the verbatim scope the project declares: site preparation earthmoving, reinforced concrete foundations, brine pipeline and drainage, fire water network, earthing, drains and underground works, roads, pavements and fencing, an equalization pond with geomembrane, earthmoving for the 'Piletas Ponds', a biodigester and a nitrifying bed. Plus Sal de Oro (Salta) and Rincón (Salta, SBDF, under way). It is the most complete contractor in the lithium NOA and the most obvious joint-venture partner — or the toughest rival.
The only documented precedent of a local partner inside a large package, with an amount and with a clause giving priority hiring to residents of Antofagasta de la Sierra under a 70/30 policy. It is the niche's entry route, with a name of its own: Huasi Construcciones S.R.L.
Section I of RP 43: 60.3 km of earthworks and paving at puna elevation, a single bid for more than ARS 33,000 million against an official budget of ARS 24,353,443,962.69 and a 24-month schedule; 15% physical progress as of Feb-2026 and ~20 km paved up to the Cuesta de Paicuqui. It is THE PROOF that there are Catamarca construction companies capable of executing at scale in the puna: this niche's gap is not one of raw technical capability. ALL THREE, with their corporate names: Empresa Constructora Guido Mogetta S.A. (CUIT 30-56874394-2, domiciled in Sumalao, Catamarca), Huasi Construcciones S.R.L. and ITALCA Constructora S.R.L. — all three members of CASEMICA. The works are paid for with the Mining Royalties Trust Fund, so lithium royalties are financing the road that gives access to the lithium.
A vertical integration threat: it sells the INTEGRATED pond package (geological and geotechnical assessment, design, grading, construction, geomembrane installation, containment berms and fill supervision) verif as an offering. If it comes in, it takes earthmoving and liner together and closes the subcontract door from both sides.
The 2023 registry (19 categories, exactly 123 memberships) declares 31 construction companies with 'modular housing, civil works, earthmoving and road maintenance' and 12 vehicle and equipment rental firms —whose declared description is 4x4s, lighting towers, aerial work platforms and chemical toilets, i.e. it does NOT evidence a heavy fleet—. None is named in a pond package. It is at once the entrant's competing base and its pool of joint-venture partners. A 2023 count: verified as a citation, out of date as a measurement of today's supply (the Re.P.E.M. went from 194 in May-2025 to 229 in May-2026).
They do not execute a single m³: they write the geotechnical specification and the QA plan —pond floor elevations, tolerances, compaction densities, tests— and they decide what gets certified and who qualifies. Technical homologation goes through them, not through the provincial counter.
At MARA, 'mine equipment' is 10.1% of the initial capex (USD 281 M) against 'mine contractor' at 21.6% (USD 601 M) verif Asset Overview Jan-2021: the buy-vs-contract decision is explicit and the supplier does not control it. Rio Tinto also buys its own salt harvesting fleet verif 20-F 2025. Every m³ the operator does with its own equipment never reaches the market.
A figure that corrects the analogy with san juan: in the province's largest package, the Chinese operator did NOT bring in a Chinese contractor — it hired an Argentine joint venture (Pietroboni + ITALCA + BMI). It did import the geomembrane (20 trucks via the Paso San Francisco). In earthmoving there is no importable m³: the risk of 'direct import through the operator's chain' exists but did not materialize in the measured case.
From Catamarca, from ANDALGALÁ -the department with 36 registrations in the Re.P.E.M., second after the capital-. CUIT 30-71002365-0, incorporated on March 3, 2007 verif. Declared activities: construction, repair and restoration of transport infrastructure works; wholesale of construction materials; Rental of machinery AND EQUIPMENT without an operator; road freight transport including specialized freight; and manufacture of non-metallic mineral products. It appears in the supplier registry of CAMYEN, the provincial State mining company. It is the exact profile of the equipment rental firm this niche identifies as the local leg of earthmoving. WHAT is not proven: not a single work front was documented for it on the Salar del Hombre Muerto lithium projects; its presence is evidenced in the province and in the category, not on the salar.
Here the easy argument «the category does not exist» does not hold: it does exist, with 31 construction companies in the registry and a Catamarca joint venture moving soil on Route 43, in the middle of the puna. The gap is a different and finer one: the scale per package, the surety bond and the position in the chain. Four doors, from lower to higher capital:
Equipment rental with an operator to whoever is already on site. The only door that does not require winning a tender — and the front of Tres Quebradas' stage 2 is open today, in equipment mobilization and crust breaking. The registry has 12 rental firms, but what they declare is 4x4s, lighting towers and chemical toilets: heavy equipment suitable for 4,100 metres, with an operator and on-site maintenance, is a different product.
Crust breaking as a sub-scope. The perfect entry item: 2.1 million m² per stage (USD 2-4 M), low technology content, zero liner risk because it is executed before the geomembrane exists, it is the first item on the schedule —so it gets invoiced early— and the lead subcontracts it without resistance because it does not move the package's margin.
Internal roads, minor platforms and the maintenance retainer. USD 1-3 M a year, recurring. It is the only thing that survives 2031, and it is structurally local: an access cut off by snow or an eroded shoulder is solved in hours, not by mobilizing a fleet from Rosario. It is the line that turns a job into a client.
A pond package of your own, via a joint venture with a local partner. 50-100 hectares is USD 3-12 M; the complete 200 ha package, USD 13-24 M. The vehicle is the one the rule itself designs —a joint venture with at least 50% local suppliers— and it has two precedents: Huasi Construcciones inside Sal de Vida's USD 130 M EPC, and the Mogetta-Huasi-Italca joint venture in an ARS 33,000 million contract at puna elevation. It is not a legal acrobatic: it is this market's normal route.
~USD 27 M/year (~60% of the TAM): what the extra-provincial lead executes with its own fleet —the largest block and the key difference from geosynthetics, because earthmoving IS the scope of Pietroboni, Milicic and Chediack, not a specialized sub-scope they subcontract entirely—; what the operator internalizes by buying equipment instead of contracting (at MARA 'mine equipment' is 10.1% of capex against 21.6% for 'mine contractor'; Rio Tinto also buys its harvesting fleet); and the technical control the global EPCM firms capture (Worley/Ausenco/Spark do not move a m³ but they write the geotechnical specification and the QA plan). What this niche does NOT have as captive, and it is worth saying: a m³ is not imported — Zijin's Chinese chain brought the geomembrane but hired an Argentine joint venture to move the soil. Here the captive share is about contracting structure, not foreign trade.
~USD 8-18 M/year addressable by a local/domestic entrant (midpoint ~USD 13 M/year): the items the lead does release —crust breaking (USD 2.1-4.2 M per stage), equipment rental with an operator, transport of equipment and borrow material, mobilization, internal roads, drainage and minor civil works—, plus a complete pond package via a joint venture with a local partner (the Huasi precedent; local high-altitude capability proven on RP 43), plus almost the entire perpetual core of road and civil maintenance in operation, which is local by nature. How the local content commitments play: HMW's 95.34% and Fénix's 60% ENLARGE the SAM because this category is the largest line and the easiest to attribute as a local purchase, but they are commitments on the investment amount, with no quota by category: mistaking them for guaranteed demand in this category is the classic error.
USD 1.5-5 M/year for ONE entrant within 2-3 years (midpoint ~USD 3 M/year). Year 1: equipment rental with an operator on an open front plus a crust breaking sub-scope (USD 0.5-3 M). Year 2: internal roads and minor platforms plus a maintenance retainer at 1-2 operations (USD 1-3 M/year recurring — this is what survives 2031). Year 3: a 50-100 ha pond package via a joint venture (USD 3-12 M per contract). The complete 200 ha package (USD 13-24 M) is not won in year one: it is won with certified m³ and with a surety bond secured.
When you get paid, and what blocks it
effect
It is the niche with the most direct employment per dollar billed in the whole Catamarca portfolio, and it arrives FIRST: earthmoving is the first job on the schedule — before the plant, before the liner, before production.Headcount per package, calculated on the verified bill of quantities: 200 ha in 8 months is 6,600 m³/day of pay items and 10,500 m²/day of crust breaking, i.e. 2-4 load-and-haul trains (a 30-40 t excavator plus 3-4 trucks or articulated haulers), one crust breaking front (2-3 bulldozers with rippers), one compaction front (motor grader, roller, water truck) and services (fuel, workshop, survey); with 8-14 people per front and shift, a 7×7 regime with a double crew and two shifts ⇒ 100-180 direct jobs per package of 200 ha estim our own, and extrapolated to the pipeline (1.5-2.5 simultaneous package-equivalents) ⇒ 200-400 construction jobs sustained through the window estim. A contrast that gives coherence: Sal de Vida declares ~900 jobs at the construction peak with more than 70% local verif and 3Q declares 565 direct with 80% from Catamarca prob.The trades, by name: excavator, bulldozer and motor grader operator (the best paid on the front: the platform's finish is played out in centimetres), front loader, compaction roller, articulated or dump truck and water truck operator; heavy equipment and field hydraulics mechanic (the scarcest trade and the one that commands most); equipment repair welder; lowboy driver for equipment transport; survey technician with GPS, total station and 3D models —the one who decides whether the pond floor supports the liner properly—; soils laboratory technician (density, Proctor, moisture); earthmoving foreman; flagger and quantity tallyman.Why it is the right door for this province: unlike geomembrane welding or the accredited laboratory, the heavy equipment trade ALREADY EXISTS in the Catamarca valley —31 construction companies and 12 rental firms in the registry— and the Mogetta-Huasi-Italca joint venture is moving soil on RP 43 right now; what is missing is not the trade, it is the altitude qualification, the volume discipline and elevation certification, and that is taught in weeks or months. The training rail exists and is measured: +43 industrial technical training courses with +600 attendees in Antofagasta de la Sierra verif and more than 70% of the participants in the 2017-2023 apprenticeship contracts hired permanently prob.For those with a degree, not just a trade: surveying and elevation control, soil and compaction testing, quantity take-off and progress certification, planning with digital terrain models, and safety management for high-altitude work —where there is a stated regulatory vacuum: there is no legislation setting fitness criteria above 3,500 masl—; it is technician and civil engineer work and today much of it is signed off by a foreign EPCM.Local linkages, the widest in the chain: every front drags fuel (3 service stations in the registry), spare parts (7), OTR tyres and a hydraulics workshop (with no category of its own), transport of equipment and material (4 members: the bottleneck and the focus of the 2026 conflict), aggregates and borrow pits, catering and camp (2), health and ambulances (6) and insurance and surety bonds (1). A 200 ha package feeds ten local categories at once, and that is why it is the category that buys the most social license per peso spent.What does not add up, without make-up: (i) construction employment is temporary by design —8-month campaigns tied to the 2026-2031 window— and what remains is 20-50 permanent jobs across the whole province between road and civil maintenance in operation estim; (ii) the fleet is capital-intensive, so if the pipeline slips the equipment bought is paid for all the same; (iii) and the environmental counterweight is the most direct of all the niches: this is the category that breaks the salar's crust —2.1 M m² per stage— and it is exactly the intervention challenged by the Atacameños del Altiplano community and the Río Los Patos litigation, with a spill already fined on the Chaschuil river. Whoever enters this niche cannot treat social license as a public relations cost: it is their number 1 operating risk.
calculate it
Concentration High per package, low by category, and with an asymmetry that defines the whole play.In ponds: 2 of 2 documented packages to the same extra-provincial lead joint venture, with a single buyer (LIEX/Zijin).In plant civil works: one contractor per operator and every one of them from outside (Milicic at Fénix, the Pecom-Chediack-Huasi joint venture at Sal de Vida) — there is no visible competition per package, there is award by track record and relationship.In the category as such, concentration is low: 31 construction companies and 12 rental firms in the CAPPROMIN registry and 229 active suppliers in the Re.P.E.M. (May-2026), but with 75 of 194 concentrated in the capital and only 12 in Antofagasta de la Sierra, the department of the three Hombre Muerto projects: the supply exists where the work is not.The asymmetry: the concentration is not one of technical capability —the Mogetta-Huasi-Italca joint venture is executing 60.3 km of earthworks and paving in the middle of the puna for more than ARS 33,000 million, and that proves it— but of position in the contracting chain. The big packages are awarded to whoever has already done them, and the only open, documented door is the joint venture with a local partner (Huasi, twice). Conclusion: the market is not unserved and the local supply is not incapable — it is displaced from the head of the contract.
The obvious name —«the mine»— is the wrong door in three of the four entry routes. The money flows through three different doors and confusing them means quoting to the party that does not buy:
At Tres Quebradas both stages were contracted by LIEX S.A. (Zijin) with the joint venture Lemiro Pablo Pietroboni + ITALCA + BMI: 200 hectares executed and 200 under way, with the bill of quantities published in the project sheet. A figure that corrects the easy analogy: the Chinese operator did not bring a Chinese contractor — it hired an Argentine joint venture.
Whoever wins the package is an earthmover and earthmoving is its scope: it does not subcontract it whole, it releases items. That is where the local firm comes in. The precedent of how it is done at a larger scale has a name: Huasi Construcciones S.R.L. came in as the Catamarca partner of the Pecom + Chediack joint venture on the USD 130 M EPC for Sal de Vida's plant.
At Fénix it was done by Milicic S.A. (Rosario) in phases 1 and 2 of the expansion, on the salar: its published scope runs from site preparation earthmoving to concrete foundations, the brine pipeline, the roads and pavements and the earthmoving for the ponds. The subsequent maintenance —access roads cut off by snow, pond edges, platforms— is local by nature: it is answered in hours.
It is not «what breaks it»: it is the dashboard for entering at the right moment. In earthmoving the warning arrives months in advance, because the pond is quantified in a permit before being tendered and it is the first job on the schedule.
Every environmental permit and every RIGI resolution publishes hectares of pond before the works are tendered: it is the niche's leading indicator, and earthmoving is the first item contracted. There is also a cross-reading worth taking in full: Hombre Muerto Oeste has the highest local-content commitment in the whole RIGI —95.34% of the investment amount to suppliers— and it is precisely the one with 250 hectares of pond still to build.
Official Gazette (RIGI resolutions) + permits from Catamarca's Mining Ministry + the pond contractors' project data sheets, by event ↗Sal de Vida Stage 2 is 5.98 M m³ = 41% of the volume ceiling, and it is a 'very preliminary' Class 4 estimate for years 3-40 of the mine plan, not sanctioned. It is the largest block and the most fragile: if the price does not cooperate, the TAM ceiling collapses on its own. The floor of the range already assumes it is not done.
3Q Phase 2 has RIGI approval from the Committee on Jul 14, 2026 but NO published resolution as of Jul 25, 2026: it is not a done deal, and it is 2.65-5.30 M m³. HMW Phase 2 (250 ha) depends on Galan's financing, which raised USD 20 M to start 4 ktpa. And Fénix 1B closes construction in Nov-2026 with a minimum investment deadline of Dec 1, 2026: that front switches off by calendar, not by decision.
It is the local equivalent of 'direct import': a m³ is not imported, but the lead contractor can execute everything with its own fleet and the operator can buy equipment instead of contracting (at MARA, 10.1% of capex in own equipment against 21.6% in mine contractor). It is the most likely killer of the SAM and it is not solved with price: it is solved by being inside the joint venture.
The 50/50 split of the disputed Hombre Muerto strip is voided once Congress settles the boundary prob. If the strip falls on Salta's side, work counted today as Catamarca's changes jurisdiction — and with it the Re.P.E.M. and the 2 years of roots the entrant paid for. It is a DOMICILE risk, not a demand risk.
Here it bites harder than in any other niche: crust breaking IS the physical intervention on the salar that is being challenged — 210 ha of crust broken per package. The Mar-2024 injunction over the Río Los Patos sub-basin was lifted 5-2 in 2026 with the case open before the IACHR prob, and the Zijin-Liex brine spill into the Chaschuil river (Mar 25, 2026), with a Ministry fine, has already happened.
It is the politically hottest category precisely because there are local bidders: in Jun-Jul 2026 CAPROMITI denounced that Zijin-Liex hired a carrier from Córdoba and threatened road blockades in Fiambalá, with the Catamarca Transport Chamber and the Fiambalá Economic Chamber demanding compliance with 70% local purchasing. An extra-provincial entrant with no local partner does not just buy a commercial risk: it buys a blockade risk.
Past the peak, demand falls to road and civil maintenance in operation: USD 2-10 M/year for the whole province. A company built on construction has to amortize its fleet in 4-5 years. The perpetual core exists and is more solid than in geosynthetics —the road is maintained every year, snow cuts the access every winter— but it is small: 20-50 permanent jobs across the whole province.
Fénix is selective adsorption, Kachi is ion exchange, 3Q is hybrid and Rio Tinto states 'Design One, Build Many': every new tonne through DLE is pond that does not get built. A FAVOURABLE NUANCE for this niche: DLE still needs plant platforms, foundations, roads and wellfields, so the site civil works component is DLE-resistant — the pond component is not.
If MARA is sanctioned, it competes for the SAME base of 31 construction companies and 4 carriers: its initial capex allocates USD 601 M to 'mine contractor' and USD 341 M to Agua Rica infrastructure verif PFS(B), a multiple of the entire lithium pipeline. It is demand upside and cost risk at the same time (scarcity of fleet, drivers and operators, wage pressure). A WARNING: this is our inference about competition for the same supplier base, not a measurement — MARA is pre-FID, with no approved exploitation EIA, with estimated operational entry in 2032.
RP 43 shows it in numbers: an official budget of ARS 24,353 M against a single bid of more than ARS 33,000 M (+35%). An 8-month fixed-price contract in the puna, with a weather window that can close —the storm of Jul 21-23, 2026 left RP 43 impassable, recorded -27 °C at the Tres Quebradas camp and forced a helicopter to be brought from Buenos Aires— is exactly where an entrant with no financial cushion goes bust.
The TAM is built from four variables you can watch from outside: how many hectares of pond get built, how many cubic metres and square metres each hectare demands, how much each unit costs at 4,100 metres and over how many years it is executed. Change one and the number is recalculated.
A robustness check by two independent routes. Top-down: Catamarca's portfolio under construction moves on the order of USD 1,500-2,400 M over the window, and earthmoving together with site civil works runs 12-20% of a lithium project's capex ⇒ USD 36-96 M/year, an interval our range contains with a more conservative floor. By pace: the only documented package did 200 hectares in 8 months including winter, so the midpoint equals ~1.5 packages running simultaneously — consistent with three or four open fronts and not with a single one. There is also a reading that can rarely be made: this niche and the geosynthetics one are the two halves of the same pond package, and it is worth looking at them together. It goes as context and not as a third confirmation, for method transparency: both rest on the same Sal de Vida capex line, so they back each other in a circle. The figure stands on the two routes above, which share no inputs.
The number rests on a few variables. The formula shows how it moves when each one changes; and each variable carries its freshness seal — how often it is worth revisiting. estim
Every figure is checked against its source before we publish it. Here we show what backs it — and where the verified data ends and our estimate begins.
This is the niche with the best-measured demand in the province, and not because we calculate it: the contractor itself publishes the bill of quantities —cubic metres of excavation, of berms and of floor, and square metres of crust breaking— for the two Tres Quebradas pond stages, and we reopened those sheets to confirm every number. What does not exist is a public price for earthmoving on a high-altitude mining project, so the USD per cubic metre is an explicit assumption and we cross-check it by three routes that share no inputs: the pond capex declared in Sal de Vida's technical report, a real puna road contract in the same province (Section I of Route 43, with a published official budget) and the productivity physics of a fleet at 4,100 metres. All three give the same band. The annual total is our own estimate and we say so.
7 sources · 3 official or agencies · 4 of high reliability · each data point links to its source.
Get on board the takeoff
This week’s updates: the map of earthmoving and pond civil works in the Catamarca puna and the niches opening up, related courses and new provinces as they launch. Free.