News Article | November 15, 2016
GENEVA, SWITZERLAND--(Marketwired - Nov. 15, 2016) - Gunvor SA and Gunvor Belgium N.V., both subsidiaries of Gunvor Group, have closed a US $725 million borrowing base facility ("Facility"), which will provide working capital for Gunvor's refining activities in the Amsterdam-Rotterdam-Antwerp (ARA) region. The Facility was launched at US $700 million and was oversubscribed at US $805 million, before being scaled back to US $725 million. Gunvor's operations in the ARA consist of Gunvor Petroleum Antwerp, Gunvor Petroleum Rotterdam (GPR) and the Maasvlakte Olie Terminal (MOT) - the latter two of which were added to the Facility this year. Gunvor acquired GPR and its share in MOT in February. ING acted as Coordinator and is the Facility- and Security Agent. In total, 13 banks participated in the Facility, which has a 364-day tenure whereby: ING Bank N.V. was Bookrunning Mandated Lead Arranger, and Rabobank was Senior Mandated Lead Arranger and Co-Arranger. Société Générale and DBS Bank LTD, London Branch were Mandated Lead Arrangers. Raiffeisen Bank International AG, Credit Suisse AG, Credit Agricole (Suisse) S.A, Nedbank Ltd London Branch, ABN Amro Bank N.V, Unicredit Bank AG, Sumitomo Mitsui Trust Bank, Mizuho Bank Ltd all served as Lead Arrangers. Bank of Tokyo Mitsubishi UFJ, Ltd was Arranger. Gunvor Group is one of the world's largest independent commodities trading houses by turnover, creating logistics solutions that safely and efficiently move physical energy and bulk materials from where they are sourced and stored to where they are demanded most. With strategic investments in industrial infrastructure-refineries, pipelines, storage, terminals, mining and upstream-Gunvor further generates sustainable value across the global supply chain for its customers. More information can be found at GunvorGroup.com or @Gunvor.
News Article | November 10, 2016
That the government is now at last being forced to do more to reduce the dangerous levels of nitrogen dioxide air pollution is welcome news (Court defeat for government on air pollution, 3 November). More than a year on from the “dieselgate” revelations, ministers should have been in no doubt about the dishonest and illegal methods used by some manufacturers to cheat emissions tests. Not only has the government failed to update its pollution modelling based on realistic emissions figures, it has also done nothing to support the 1.2 million UK diesel vehicle owners caught up in this scandal. If anything, the government seems to be going out of its way to protect manufacturers. Cars fitted with the defeat device software are still able to pass MOT emissions tests, so the Driver and Vehicle Standards Agency seems to have turned a blind eye to this. Compare this with the situation in the US, where the Environmental Protection Agency has successfully prosecuted VW and arranged for diesel owners to be compensated. In the UK, it seems, Defra does nothing. Successful prosecution here would send a clear message to all manufacturers to be honest with their emissions and fuel economy figures. In addition, the introduction of on-the-road testing, rather than the static lab testing presently used, would provide more reliable figures. Alan Beamish Thirlby, North Yorkshire • Your editorial (9 November) is right to call on government to address the UK’s dangerous problem of air pollution by reprioritising greener transport, but failed to mention in its proposed solutions the most environmentally friendly and cost effective modes of all: cycling and walking. Cycling’s bang for its buck outstrips every other mode of transport, with conservative Department for Transport estimates showing that for every pound spent, the economy takes five back, particularly in terms of health and reduced congestion. We still await the government’s promised Cycling and Walking Investment Strategy which currently allocates just 72p per person for cycling and walking in 2021. Our level of investment in motorways and A-roads at the same time will be £84 a head. If government wants to make a real difference to cleaning our air, then let’s rebalance this spend, and give everyone the opportunity to choose cycling for shorter every day journeys. David Murray Head of campaigns, Cycling UK • George Monbiot (Cars don’t just choke our children. They tear a hole in our communities, 9 November) is right that air pollution is a public health emergency. Yet our local council still proposes to build hundreds of homes on the wrong side of our town for commuter travel, thus putting up to a thousand extra cars on the streets of a medieval town. This is despite declaring an Air Quality Management Area across the whole town and despite its own highways assessments that predict an average 350% increase in traffic queues at peak times. It seems that no health emergency is severe enough to stop relentless development in the face of all the evidence. When the revised plan goes to the Planning Inspectorate we will see how seriously the government takes this issue. Richard Gilyead Saffron Walden, Essex • George Monbiot is right to draw attention to the need to reduce pollution from traffic. However, what he may not realise is that this is being used as a smokescreen to distract attention from the much more serious issue of indoor air pollution. According to the World Health Organisation and many other scientific bodies, hazardous chemicals found in indoor air are at least five times more damaging to health than external air pollution. The government has told its committee on the medical effects of air pollutants to ignore internal air and concentrate on external. This means that the occupants of increasingly airtight buildings are at risk of asthma, respiratory problems and cancer from breathing in dangerous chemicals emitted inside buildings. Dr Tom Woolley Crossgar, County Down • George Monbiot underestimates the damage to our communities from the motor car. Where roads are safe and children can play out in them as they have done for countless generations, then parents talk to each other and speak of “keeping an eye out for each other’s children”. The damage to feelings of community means that the neighbourhood feels full of strangers rather than friends and acquaintances. This explains the exponential rise in fear of “stranger danger” which has no basis in evidence. Children’s loss of freedom to play outside results in lack of exercise and consequential obesity. Rob Wheway Children’s Play Advisory Service • Once again George Monbiot tells us what we need to – but often would rather not – hear. The car has given as many problems as solutions for mankind. And as regards the freedom it brings us, it comes in inverse proportion to that of other people, both locally and globally. Barry O’Donovan London • When John Prescott as transport minister imposed the same duty on biodiesel as on ordinary diesel he stopped the development of the biodiesel industry. If you want to improve air quality in London and our other major cities quickly and effectively, cut the duty on biodiesel . Coach operators road hauliers and private motorists would all start using biodiesel again. Biodiesel does not contain the same harmful particulates one finds in ordinary diesel. If one stands on Highgate Hill and looks out across London one can see a cloud of smog over the city as if London was sitting in a bowl. Londoners are breathing that air. When I visit London after a day or so, if I wipe my nose, I find soot on my handkerchief. Nigel Boddy Darlington • George Monbiot’s article on cars and pollution never mentioned the important role of free bus passes . My wife and I use our passes when ever we can , rather than our own transport. We have a small campervan that is our only transport. I believe that bus passes are an important way to cut back on car use, I always feel that we are making a sound evironmental choice. When visiting cities we use park and rides, or else purchase cheap rail tickets on line. We are both concerned about climate change, always looking for ways in which we can make a difference. Every little counts is my belief. Brian Woods Weymouth, Dorset
News Article | October 28, 2016
ProRehab Physical Therapy (PRPT) is pleased to announce the addition of Hand Therapists Mary Cecil, PT, CHT, and Toby Cross, MOT, OTR/L to their staff. Both will provide specialized hand therapy to patients with injuries to or dysfunction of the hand, wrist, elbow, or shoulder from the group’s Middletown and Highlands locations. This is important because early and convenient access to the skilled services of a Certified Hand Therapist (CHT) improve outcomes and length of treatment for patients. The CHT specialty designates a therapist with a minimum of five years of clinical practice, including over 4,000 contact hours in designated hand therapy practice. Once these two criteria have been achieved candidates are eligible to sit for national board certification. Successful completion of this process denotes the highest level of competency in the hand therapy field. Cecil received her CHT designation in 1992, and Cross will be eligible to sit for her exam in November of 2016. Hand therapy at PRPT is recommended for patients who have suffered any number of injuries to their hand, wrist, elbow, or shoulder, including: Cecil and Cross provide specialized hand therapy to patients using interventions such as therapeutic exercise, sensory reeducation, manual therapy, and soft tissue mobilization, with great emphasis placed on home programs and patient education. Cecil and Cross are also able to provide custom orthotic fabrication and fitting of appropriate prefabricated orthotics for a variety of surgical and nonsurgical conditions. For more information about hand therapy or to book an appointment with one of our expert certified hand therapists, please visit http://www.prorehablou.com. About ProRehab Physical Therapy: Founded in 1999, ProRehab is a locally owned private physical therapy practice with over 15 years of experience as one of the region’s preeminent providers of outpatient physical therapy. Our therapists are charged with providing long-lasting outcomes without surgery, medication, or diagnostic imaging to our patients in the Louisville, Evansville, southwest Indiana and western Kentucky areas. ProRehab Louisville’s clinicians have a wide range of clinical expertise in manual physical therapy, spine care and rehabilitation, general orthopaedic conditions, sports specific injuries, workplace injuries and much more. 100% of our expert practitioners are either board certified or in the process of completing advanced residency, fellowship, or clinical mentoring programs, allowing them to provide the highest standard of care to our patients. Our friendly, caring therapists take their patients’ care personally and are always on standby to offer support. Attuned to the latest evidence-based clinical, service, and compassion research, we combine Clinical Excellence and AmaZing! Customer Service with empathy, compassion, and positivity and incorporate them into our mission of keeping our patients healthy and out of the costly, depersonalized healthcare system. For more information, find ProRehab Louisville at http://www.ProRehabLou.com, on Facebook, or @ProRehabLou on Twitter and Instagram.
News Article | November 28, 2016
Black men are three times more likely to develop prostate cancer than other demographics, yet black men are consistently underrepresented in research studies, say researchers from King's College London in a new paper published in ecancermedicalscience. The problem? Healthcare research conducted in one demographic often doesn't translate to results for other demographics, and recruitment of ethnic minorities often fails to meet a balance. For example, in 2015, researchers realised that the HPV vaccines being given in the United States of America were less effective in African American women. The vaccines were developed for strains of the virus more commonly found in white women, on whom the clinical trials had been based. Now, researchers warn that black men may not benefit from current research studies on prostate cancer. "It's important that future research be generalizable," says study author Dr Mieke Van Hemelrijck of King's College London, London, UK. She explains that if black men don't participate in research studies, their unique needs and experiences will be missing from the conclusions. To find out why this community is less likely to participate in research recruitment, the researchers surveyed the opinions and attitudes of sixteen black men undergoing treatment for prostate cancer in London. "We uncovered a number of barriers to the recruitment of black men with prostate cancer into research studies," says Dr Van Hemelrijck. These include mistrust of researchers and their motives, lack of understanding about the benefits and consequences of participating, and even a feeling in immigrant communities that traditional medicines and coping methods are more reliable. So how can research studies on prostate cancer become more inclusive for black men? The researchers suggest that community-driven engagement may be the way forward, as well as dissemination of reliable information. Hearing from black male survivors of prostate cancer may improve screening uptake as well as participation in research. "There was mention of a mechanic who offered a free MOT to every black man who had received a prostate check at their GP surgery," says lead author Dr Charlotte Toms of King's College London, London, UK. "When asked "do you think it helps the fact that he was a black man, who had been through it himself? And then he was then trying to engage people to come in?" An individual responded, "yes, yes. That helped.""
News Article | December 14, 2016
The rogue practice of removing vital pollution filters from the exhausts of diesel vehicles has suffered a blow with the Advertising Standards Authority (ASA) for the first time banning an advert for the service. However the number of filterless cars on UK roads, pumping out high levels of toxic particles, remains unknown and air pollution campaigners say the government must investigate and then crack down on the shady practice. Particulate pollution causes almost 29,000 early deaths each year, costing £15bn in health costs. Diesel particulate filters (DPFs) have been compulsory in new diesel cars since 2009 but can become clogged and cause breakdowns, particularly for city drivers. Replacing DPFs usually costs more than a £1,000 and, as a result, many garages across the country offer to remove the filter completely for a few hundred pounds. They exploit a legal loophole which means removing a DPF is not an offence, only driving the car afterwards on a public road. Without the filter, particulate emissions are likely to soar fivefold. Friends of the Earth complained to the ASA about an online advert from a garage in Bristol which offered to remove DPFs. The advert said: “The only MOT regulation regarding the DPF is a simple visual inspection. As long as the DPF still appears to be fitted, the vehicle will pass the MOT visual inspection. Therefore we only remove the internal core, leaving the outer casing in place. The vehicle will appear to have a DPF fitted and will appear unmodified.” A qualification was included at the bottom of the advert which stated “*Our DPF Removal service is sold for off-road use only”. However, the ASA concluded on Wednesday that this “was not sufficiently prominent to counter the overall impression that vehicles which had their DPF removed could be used on public roads” and banned the advert as misleading. Any future adverts must make “immediately clear with sufficient prominence that it is illegal to drive such vehicles on a public road”, the watchdog said. In 2014, ministers said the “unacceptable” practice of removing the filters was “clearly detrimental to people’s health” and introduced a visual check into the MOT. The Guardian revealed in May this had caught 1,188 vehicles. But with some garages promising “invisible” DPF removal to beat the MOT test, as in the case of the banned advert, the true number of filterless vehicles is likely to be much higher. “The more we uncover about diesel vehicles, the dodgier they seem,” said Aaron Kiely, at Friends of the Earth. “Removal of pollution filters – which are designed to reduce dangerous emissions and protect people’s health – should clearly be illegal. The Department for Transport must also investigate the full scale of this problem and clamp down on those who are putting people’s lives at risk.” A spokesman for the Department for Transport (DfT) said: “The government is firmly committed to improving the UK’s air quality and cutting harmful emissions. Changes to the MOT test were implemented in 2014 to ensure that particulate filters are fitted where necessary and if not then the vehicle would fail the test.” “Alternative methods of detecting the removal or effectiveness of particulate filters are under development and we are examining their suitability for use in an MOT,” said DfT spokesman: “Further research is required to ensure potential methods accurately determine a pass or fail.” Almost all diesel cars have also been revealed as emitting far higher levels of nitrogen dioxide than official limits. NO2 levels are illegally high in most of the UK’s air quality zones and calls are rising to ban dirty diesel vehicles from British city centres, as Paris, Madrid, Athens and Mexico City have already pledged. In May, the Guardian also revealed that Google, eBay and Gumtree had refused to ban adverts offering to remove DPFs.
News Article | October 28, 2016
Unlike a new vehicle, fresh from the factory and unloaded off the transporter complete with a manufacturer’s warranty, a used commercial vehicle will usually have travelled tens of thousands of miles humping, carrying and grafting for its previous owners. So regardless of the type of vehicle you are considering purchasing, you can increase your chance of netting a reliable workmate through taking note of the below considerations. Inspect the Bodywork Arguably the most important element to check for when purchasing any used vehicle, is the presence of rust. Although most obvious on wings, doors and bumpers – ensure you also check the wheel arches and at the corners of the windscreen. It’s important to also check not only the vehicle, but the vehicle’s body. This may include tipping bodies, cages and traffic management bodies. During this inspection, you should also consider the following: • Are the panels consistent? • Is there any recorded evidence of an accident? • Does the colour match across all panels? • Is the exhaust in good condition? • Are the tyres legal? • Does the condition of the vehicle match up with the vehicles recorded mileage? Inspect the Suspension Walk around the vehicle and confirm that it is standing level. Press down on each corner of the vehicle, if the shock absorbers are in place then the vehicle will bounce back without a problem. If you have any reservations about the performance of the suspension, view the previous MOT and take the vehicle on a thorough test drive. Inspect the Engine You don’t have to be a mechanic to carry out some simple checks which will confirm the overall condition of the engine and its associated parts. Carry out the following checks: • Search for oil leaks around the engine block • Look for worn, damaged and frayed belts and hoses • Check the oil levels, look for a smooth consistency. Gritty, lumpy or foamy oil could signal underlying problems • When the vehicle is cold, inspect the water levels. A tinge of anti-freeze is fine but keep an eye out for oily traces and low water levels. Learn the History Failing to check three key elements of your new vehicles history is an unnecessary risk, which may cost you in both the long and short term. The three main elements you need to consider are: Clocking In layman’s terms, a vehicle has been clocked if it has had its displayed mileage turned back to increase both its saleability and value. It’s estimated that almost 600,000 vehicles in the UK have been ‘clocked back’ the below information gives you some tips on how to spot this increasingly common problem. • Check MOT Certificates and Service History – The correct mileage is recorded on every single MOT, so review these and ensure there are no odd gaps or periods when the mileage is lower than the previous year. If there aren’t any old MOT’s you can check the details online using the V5c. It is also well worth matching the recorded MOT mileage to the vehicle’s service history documents. • Commercial vehicle interiors are generally built pretty tough, so look out for clear indicators that the mileage and usage may be higher than is claimed. Focus on worn out interior items including steering wheels, seats, pedals and switches. With regards to the exterior, listen out for clunks and bangs which may indicate prematurely worn out bearings and parts. • Test drive some similar models. Through test driving similar models, you can get a real sense of how the vehicle should both drive and perform. Outstanding Finance It could be that the vehicle you are considering purchasing has outstanding finance on it, making it the legal property of the finance company. If this is the case, you can easily come to an agreement with the finance company, and still end up owning the vehicle. Do not however, rely on the seller’s word that they will settle the outstanding balance. Insurance Write Off A history check may reveal that that the vehicle you are viewing has been written off or categorised as a total loss by an insurance company due to damage or theft. Although this doesn’t necessarily mean you should walk away from the vehicle, tell your insurance company about the classification, it won’t affect the yearly premium you pay, however it may affect what your vehicle is worth in the event of a pay-out. Note that you may have some difficulty in re selling a written off vehicle, any potential buyer will share the same reservations you had at the point of purchase. It is also possible that the vehicle will no longer be marketable as part exchange at dealerships. Remember to have a full mechanical inspection carried out to ensure that the repairs have been carried out to a good standard. Never pay full book price for a written off vehicle. CPD Bodies Ltd of Stockton on Tees, North East England, are leaders in Commercial Vehicle Bodybuilding and repairs, specialising in hand built premium quality Truck and Van Bodies. Based in Stockton on Tees in the North East of England, we are accessible to all areas of the UK, offering a pickup and delivery option for our commercial bodybuilder services in dedicated towns and cities across the country. We pride ourselves on fast turnaround times and quality of workmanship regarding traffic management and repairs alike.
News Article | December 5, 2016
It has been one year since brother Ali and Iman Samizadeh launched Carcodeal.com at Web Summit Dublin and to celebrate they are heading to TechCrunch Carcodeal is an online platform that connects over 200 used car dealers with sellers to provide instant online quotes. It has been one year since brother Ali and Iman Samizadeh launched Carcodeal.com at Web Summit Dublin and to celebrate they are heading to TechCrunch on 6th December. Sellers simply enter their town or postcode, registration number and mileage, along with how far they wish to travel, to receive instant bids valid for 72 hours. No fees for sellers. No need to provide personal details to receive quotes. No need to post pictures or additional information. Sellers can review dealers’ profiles, cash and part-exchange offers instantly. There is no fee for the seller to pay and currently the site is free to use for dealers too. Dealers will pay a membership in the future and can currently set their parameters to automate their bidding and second-hand car buying preferences. Dr. Iman Samizadeh, Carcodeal co-founder and CEO, said: “Carcodeal is here to completely overhaul and radically improve both the private seller and car trader user-experience. It not only cuts out the middlemen, but raises the car-selling bar by firmly moving away from hidden fees and often unfairly-low vehicle valuations. And it’s both equitable and fair for all users. We estimate an optimistic gross margin over the industry average. We will be successful because of the unique idea and the excellent team we have assembled, as well as the drive and determination of the owners. Our future sits within dominating used car market sales by providing a service to both vehicle traders and private car sellers who are not internet savvy, or want to move from traditional methods of selling their vehicles online. Carcodeal.com will be expanded to motorbikes and commercial vehicles trades in the coming years, and providing service to other high demand markets such as the US, Germany and the Netherlands.” When dealers sign up they gain access to an online control panel to bid on specific types of makes and models of used cars. Dealers can tailor their bids further by age, MOT status, Service history and mileage. The dealers receive email and text message notifications when the system automatically bids for a car to contact the seller. Owned and run by brothers Iman and Ali Samizadeh, Iman has a PhD in Computational Intelligence and Ali is the marketing search engine geek with a degree in CGI. For further information please contact [email protected].
News Article | March 2, 2017
The use of self-employed drivers and pressure to make on-time deliveries are risks to road safety, warn cycling and road safety campaigners. The growth in online shopping and home delivery helped push van traffic in the UK to a new peak in 2016. However, vehicles most used by online delivery drivers, those under 3.5 tonnes, are not subject to the operator licence regulations that apply to larger vehicles, and anyone with a standard driving licence can drive them. Giving evidence to MPs on Tuesday, Siwan Hayward, deputy director of enforcement and on-street operations at Transport for London, admitted that for anyone who drives non HGVs for work, there are not sufficient standards for vehicle safety and driving hours. Campaigners say reported issues involving the delivery companies FedEx and CitySprint illustrate their road safety concerns. Last August, a FedEx driver was implicated in a road traffic incident with a cyclist. After the cyclist contacted the company, a FedEx director accidentally forwarded an internal email in which staff decided not to answer any of his safety procedure questions because he wasn’t considered “high engagement” enough. The safety incident was the second in a week. It came after a FedEx driver was filmed driving on the wrong side of the road, apparently asleep at the wheel. In its corporate brochure FedEx states its support for the road safety initiative Safe Kids Worldwide, yet in the email conversation staff discuss how to tackle what Trevor Hoyle, senior vice president of northern Europe operations, describes as “social media incidents”, rather than safety breaches. The company did not respond to the cyclist’s specific questions, including what driver training was provided. On request, it sent a statement to the Guardian saying the driver had been dismissed following an internal investigation, which identified him as a subcontractor engaged by FedEx. “Road safety is a critical public concern which FedEx takes extremely seriously,” it added in the statement. “FedEx has zero tolerance for such unacceptable conduct and we expect anyone who works for the company to comply fully with all traffic laws and regulations to ensure our roadways are safe.” Duncan Dollimore, Cycling UK’s road safety and legal campaigns officer, said the concern was not about staff being disciplined but whether they were trained to help avoid future safety breaches. “Just dismissing such concerns and hoping they will simply ‘go away’ is to disregard corporate responsibility,” he said. When asked what it was doing, beyond disciplining staff, to avoid a repeat of the two incidents, FedEx directed the Guardian to a page on its website, which referred to “proactive, safety-focused workplace education” and investment in safety equipment and safety standards. Dollimore believes employers who operate and manage vans should adopt an approach similar to heavy goods vehicles operators, who are subject to operator licence regulations, as well as voluntary schemes like FORS (Fleet Operator Recognition Scheme). One self-employed CitySprint driver, who spoke to the Guardian anonymously, says he doesn’t make enough to pay for regular servicing and maintenance of his vehicle, while the CitySprint livery on his van means he can’t get work for other companies. A company is not obliged to train drivers classed as self-employed and, if it does, it risks reclassification by HMRC as an employer, and liability for holiday and sickness pay. “The thing about safety, it’s not about your ability to handle a vehicle, not if we are talking about a van,” said David Davies, executive director of the charity PACTS, the Parliamentary Advisory Council for Transport Safety. “It’s more about your state of mind, the pressure you are under, your professional attitude and those are things that are quite difficult to train; they are part of the company culture. I think it’s more important that companies actively manage the safety of their drivers,” he added. Jay Parmar of the British Vehicle Rental and Leasing Association, which leases and maintains around one in six vans on UK roads, says many operators use the MOT as a diagnostic tool, do the minimum to pass and undertake little to no maintenance for the next year even though a working van can travel 20,000 miles in that time. CitySprint says it offers automated MOT checks for fleet vehicles, online and at 41 service centres. If completed online by their self-employed couriers, at least one of these each year must be completed in a service centre. It also says it runs service vehicle checks every three months, examining roadworthiness and fuel efficiency. Campaigners argue the Health and Safety Executive (HSE) needs to start investigating and prosecuting cases of employer negligence. But the HSE denies it has any responsibility for incidents on public roads and says the police or Vehicle and Operator Services Agency, since renamed the Driver and Vehicle Standards Agency (DVSA), are responsible. Davies said: “[Police] won’t think: was this guy tired, was he under unrealistic pressures to make delivery slots? If the police refer it to the HSE, they might investigate but they are very unlikely to investigate in a proactive way.” Dollimore says the DVSA deals mainly with tachograph offences. There is, he says, a gaping hole in tackling corporate road safety failures. CitySprint said all its couriers are self-employed, can work as little or often as they like, and are free to turn down work. It says it pays some of the best rates in the industry and tries to maximise driver earnings using smart technology. Sign up to be a Guardian Sustainable Business member and get more stories like this direct to your inbox every week. You can also follow us on Twitter.
News Article | November 18, 2016
DETROIT, Nov. 18, 2016 /PRNewswire/ -- Wayne S. Brown, CEO and President of Michigan Opera Theatre, today announced the appointment of Maestro Stephen Lord as Principal Conductor, a newly created position at MOT. The appointment is effective immediately and will continue through the 2018-...
News Article | January 27, 2016
The incident probe laser power is always increased and decreased adiabatically. The duration of the measurement pulses is 200 μs—the shortest possible time while avoiding significant ringing in the homodyne signal. Typical incident probe powers are of the order of 10 pW (intra-cavity photon numbers: 90 photons per pW on resonance). Since the homodyne signal is time dependent, we apply a time-dependent weighting function in the analysis of the time traces to extract the cavity frequency shift. This method gives the same signal and photon shot noise values irrespective of pulse shapes and duration as long as the pulse area is conserved and the intra-cavity power adiabatically follows the incident power. Quantitatively, given a time-dependent homodyne signal s(t), the cavity shift is , where D is the frequency discriminator (Hz/V) at the peak signal level, and ρ(t) is the temporal shape of the signal normalized such that its peak is at 1. ρ(t) and D are determined experimentally in the absence of atoms by detuning the probe from the exact cavity resonance by a small amount (well within cavity linewidth), and recording the homodyne signal. The procedure above extracts the correct cavity shift as long as the homodyne signal is in the linear regime. We apply an additional correction factor to properly measure the shifts which are outside this linear regime. This factor is calibrated by setting the cavity–probe detuning to a known value and noting the discrepancy between the known shift and the inferred shift. This is important when establishing CSS noise levels. At large atom numbers the cavity linewidth broadens owing to atomic scattering, causing the squeezing saturation. The atom-number-dependent linewidth is κ = κ + κ , with κ the empty cavity linewidth and κ the additional scattering contribution. The fractional change is . We incorporate this signal degrading effect into the cavity shift analysis. The change in the overall signal shape due to broadening is negligible for our parameter range. We experimentally verify that we indeed obtain κ /κ ≈ 0.30 at 5 × 105 atoms. To do this, we jump the cavity frequency by a known amount (smaller than κ) between consecutive measurements and observe the signal reduction in comparison to the empty cavity case. This measurement takes 10 min; it would have taken 11 h to reach the same precision had we not employed squeezed states. The homodyne detection setup (Fig. 1c) is seeded with 780 nm light obtained by frequency doubling the 1,560 nm lattice light, which is frequency stabilized to the main cavity following an intermediate stabilization step involving a scrubbing cavity (to narrow down the linewidth of the laser). Thus, the 780 nm light is already stable in the short-term with respect to the cavity (80 mHz/ level in the 0.2–4 kHz band—about 8 Hz (r.m.s.) stability). In the long term, thermal drifts in the cavity mirror coatings cause variations in the individual cavity lengths seen by the 780 nm and 1,560 nm light. During experimental cycles of 1 s, the probe frequency can drift by 100 Hz (r.m.s.); we correct the drift at the end of every cycle with an auxiliary empty cavity measurement. Using a 200 μW local oscillator (shifted by 80 MHz from input) on path A of the homodyne system, the balanced detectors operate photon shot-noise limited from 10 Hz up to 5 MHz. Two 10 nW spectral components on path B (offset by 78 MHz and 82 MHz from input) travel to the cavity and promptly reflect back from the first mirror to give a heterodyne beat-note signal (2 MHz) at the detectors. This signal is used to stabilize the interferometer path lengths via feedback onto the AOM on path A. Stabilization covers the DC to 15 kHz frequency band, thus removing the influence of optical phase noise for the squeezing measurements. Path B also contains the probe (80 MHz offset from input), interfering with the local oscillator to form the homodyne signal after returning from the cavity. The overall detection efficiency limiting the achievable squeezing is ε = 0.16. The breakdown is as follows: a factor of 0.50, since we collect light only from one cavity mirror; 0.57 due to loss in cavity mirrors; 0.80 backwards fibre coupling efficiency; 0.85, loss in isolator and other optical elements; 0.85, interferometer mode matching efficiency. Multiplying these factors results in the stated overall efficiency. Our ability to estimate the centroid of the cavity resonance frequency improves with the square root of the number of photons contained in a 200 μs measurement pulse, and saturates around 15 Hz owing to laser frequency instability. Although we have uniform atom–cavity coupling, we still have residual inhomogeneities due to thermal motion. Figure 4b shows the noise between back-to-back measurements for varying measurement separations: noise is smallest when the separation is half-integer multiples of the 2.2 ms transverse-oscillation period. We model this behaviour to make an estimate of retrievable squeezing with fluorescence imaging. The observed effect can be explained with the statistical fluctuations in atomic positions from one experimental cycle to the next. A classical trajectory analysis on the atoms in a harmonic trap (transverse motion) suffices to predict the noise in two consecutive cavity measurements as a function of measurement separation time. Given their initial phase space points during the first measurement, the location of the atoms during the second measurement, and hence the cavity frequency difference, can be deterministically predicted. However, each time the experiment is repeated, the atoms will start at a different configuration in phase space giving rise to a slightly different cavity frequency difference. We calculate this additional noise assuming a thermal distribution for the atoms. The resulting prediction is , where ω is the transverse oscillation angular frequency and α = 2σ /w with σ the transverse size of the atomic cloud and w the probe beam waist (α2 = 0.076). The solid lines in Fig. 4b are the addition of this function in quadrature with the baseline without any free parameters. A similar analysis could be used to predict the additional noise that will arise when the first measurement is done using the cavity and the second one using fluorescence imaging. The second measurement is insensitive to atomic positions; hence the noise comes only from the statistical fluctuations in atomic positions during the first measurement. The estimate for the additional noise is , which is 17 dB below CSS noise for our experiment. In particular, an 18.5 dB squeezed state would read 14.6 dB with fluorescence imaging. For transitions from the clock states with π-polarized light, the single atom cooperativity is C = 4g2/κ Γ = 0.78, with atom–cavity coupling g = 2π × 96.7 kHz, empty cavity decay rate κ = 2π × 8.0 kHz, and atomic decay rate Γ = 2π × 6.06 MHz. The value for g2 is an average over the position distribution of the atoms inside the lattice due to the finite temperature. The 25 μK atoms are distributed over about 1,000 lattice sites. The r.m.s. atomic cloud size inside each 520-μK-deep lattice site is 17 μm in the transverse direction and 37 nm in the axial direction. The trap frequencies in the corresponding directions are 460 Hz and 205 kHz. We vary the atom number in the experiment by changing the initial magneto-optical trap (MOT) loading time. Strictly speaking, the residual inhomogeneity in atom–cavity coupling due to thermal motion requires one to define effective atom numbers N for the purpose of identifying the CSS noise level. This method was adopted in refs 19, 20, 21 and 24, owing to lack of homogeneity in couplings, resulting in N ≈ 0.66N , with N denoting the real atom number. In this work we achieve (see next section for details). For practical purposes, we do not differentiate between real and effective atoms. We would like to measure the collective observable , with N the total number of atoms, and the z-component of the spin operator for atom n. However, because of the residual inhomogeneity in the atom–cavity coupling we measure a slightly different collective observable that is a weighted sum over . If all atoms coupled identically to the cavity with the per-spin-flip cavity frequency shift for atom n given by δ(n) = δ , the total shift would be . However, owing to the small fractional deviations in the coupling constants for the atoms n, the total shift is given by: Here Z is a normalization constant, and δ = δ Z is the effective cavity shift per spin flip. To decide on a normalization, we utilize two properties of : its maximum and its projection noise with uncorrelated atoms . Here 〈•〉 indicates an ensemble average. We choose Z such that the statistical condition satisfied by J is also satisfied by . This leads to , and thus: Consequently, one can think of the non-uniformly coupled system of N atoms as a uniformly coupled system of effective atoms in conjunction with an effective cavity shift per spin-flip. Here δ is the cavity shift for an atom on the cavity axis averaged over the distribution along the tightly trapped longitudinal direction. This gives δ = 0.83δ = 5.5 Hz, where δ is the cavity shift for an atom localized at a peak of the probe mode profile. Using the cavity, we detect J only in a very narrow range. For observing J in its full range of ±N/2 we use fluorescence imaging. We release the atoms from the lattice, and for state selectivity, we apply a laser beam resonant with the F = 2 to F′ = 3 transition to momentarily push the F = 2 (that is, ) atoms. This spatially separates the and states, permitting the simultaneous detection of the number of atoms in both states. We image the fluorescence from the two clouds for 2 ms on a CCD camera using the cooling and re-pumping lights of the MOT. This is how the observed Rabi and Ramsey fringes are mapped out and hence how the coherence is measured. The pre-squeezing procedure is a supplement to the J measurement protocol, increasing robustness and efficiency. However we note that it does not alter the nature of J measurements—we obtain the same final squeezing results in the absence of pre-squeezing if we post-select the runs with the first measurement outcome lying within the linear regime of the homodyne signal. When approaching the saturated regime of squeezing, discussed in the main text, the r.m.s. cavity shifts due to CSS noise approach the linewidth of the cavity. This quantum noise prevents us from preparing initial J distributions that fall purely within the linear region of the homodyne signal using microwave rotations alone. We therefore resort to atom–cavity nonlinearities to deterministically pre-squeeze the state by a sufficient amount such that the distribution fits within the linear regime. The nonlinearity employed is a interaction causing one-axis twisting, similar to the one used in ref. 20. For an initial state near the J axis, the procedure dynamically compresses the distribution on the Bloch sphere in the z direction while expanding it in the y direction. The technical noise on the initial J preparation is also suppressed. The pre-squeezing occurs after the composite π/2 pulse brings the state to the equator. We send 100 nW of light at 6.25κ detuning from the bare cavity resonance (generating interaction) and simultaneously turn on a 400 μs microwave π/12 pulse (generating rotations around the J axis). With the combined action, we observe up to 7 dB unconditional J squeezing while retaining 99% coherence. The cavity measurements, while projecting the atomic ensemble onto a state with reduced J noise, also act back onto the conjugate observable J and increase its noise. In the ideal case a measurement would preserve the area of the uncertainty ellipse, that is, the reduction and the increase in the noises of J and J respectively would be through the same factor. However owing to photon losses, inefficiencies in extracting the information in the read-out, and the additional spin-flip noise in J , the balance is expected to be broken. Experimentally, the variance of J scales linearly with measurement strength, reaching 39 dB above CSS noise at π measurement strength accompanying the quoted 20.1(3) dB squeezing in J . This corresponds to a factor of 8.8 increase in the uncertainty ellipse area. The observed level of anti-squeezing is within 2 dB of the expectations. The maximum attainable spin noise reduction can be found by considering the individual effects of the measurement noise and the spin-flip noise. We will specify these quantities as functions of the (experimentally accessible) differential phase shift accumulated on the clock states. Here, n (t) is the intra-cavity photon number. We will also use the cooperativity, . The number of scattered photons is m = ϕ (Γ/ω ). The hyperfine splitting enters because the atom–cavity detuning is set to Δ = ω /2. Using the branching ratios for 87Rb, it can be shown that only 1/6 of the scattering events will give rise to a spin-flip, that is, a change of the hyperfine state. These spin-flips will give rise to a random walk on J with a variance of . This is the spin-flip noise; it grows with atom number and probe power. To examine the measurement noise, we analyse the information imprinted on the light transmitted from the cavity. The total decay rate of the cavity is κ = 2κ + κ + κ , where κ is due to mirror out-coupling, and κ is due to optical losses in the mirrors. The term due to atomic scattering can be expressed as κ /κ = NC(Γ/ω )2, where κ = 2κ + κ . Around zero cavity–probe detuning, the number of photons transmitted through the cavity is , where ε = 2κ /κ is the cavity efficiency incorporating the optical losses. As the atoms shift the cavity frequency by , the phase shift on the light upon transmission is . Given the quantum phase noise for a coherent state, the noise equivalent J resolution is . For a symmetric cavity, equal amounts of information leak out from each mirror. Thus, including the information gained from the reflection would improve the resolution by . Lastly, we also include the effect of photon losses on the way to the detectors, and bundle all efficiency factors into the quantity ε which is further discussed in the ‘Atom/cavity parameters’ section above. The final expression for the measurement noise is ; it decreases with increasing probe power. As a first approximation, the total noise in the estimation of J can be found by adding the contribution due to the two sources: . An optimal ϕ (that is, measurement strength) minimizes this expression, at which point . Assuming negligible coherence loss, as is the case experimentally, we arrive at an optimal metrological enhancement . At first, the achievable enhancement increases with atom number, attaining a maximum of at . This saturation effect can be traced back to cavity linewidth broadening from atomic absorption. For ε = 1, the maximum achievable enhancement is around 28 dB. Exact numerical agreement should not be expected between the naive model presented here and the experiment, since the latter is more complicated. We use the functional forms derived here to fit to the data in Fig. 3. In particular, for Fig. 3a we use , and for Fig. 3b we use . Here, α, β and γ are the fit parameters. Measurements of the collective spin operators for an ensemble of N two-level atoms can be used to quantify the amount of entanglement in the ensemble27, 28 without further reference to the specific nature of the states. We will follow the analysis of ref. 28, where the J variance and the mean-square Bloch vector length places the measured states on a plane with boundaries corresponding to different entanglement depths. In our experiment, because of the residual inhomogeneities in atom–cavity coupling, we measure the collective observable , where Z is a normalization constant and ε is the small fractional deviation in coupling for atom n (see ‘Coupling inhomogeneity’ section). This implies that we cannot directly utilize the measured spin noise values for the purposes of calculating entanglement depths. However, even without a direct measurement of J itself, its maximum possible variance can be inferred, as was argued in the Methods section ‘Inhomogeneity analysis for free-space release’. There, we found that there would be an additive noise of 17 dB below CSS noise if we tried to read out the prepared states via fluorescence imaging, which we consider to be a true J measurement. Therefore, we infer J on the basis of this analysis. Unlike the cavity-based measurements, the Bloch vector length measurements, which are done via fluorescence imaging, can directly be used for calculating entanglement depths. In Extended Data Fig. 1, we plot the inferred J variances for the 5 × 105 atom data set using the experimentally established noise after the first measurement ( of the values in Fig. 3a). The point with the largest metrological gain (π measurement strength) gives an entanglement depth of 330(15) atoms, while the largest entanglement depth is 680(35) (0.5π measurement strength) atoms. This exemplifies that entanglement depth is in itself not a direct predictor for metrological improvement. The additional noise in our model in inferring J originates from shot-to-shot randomization of atomic positions. In the absence of this randomization, we expect the discrepancy between the variances of and J to be less. Thus, the quoted entanglement depths should be taken as lower bounds. For reference, had we not taken into account the coupling inhomogeneity we would have found the largest entanglement depth to be 1,605(30) atoms. No statistical methods were used to predetermine sample size.