Soil nutrient supply is one of several environmental and biotic drivers of planted forest productivity and testing the soil and tree foliage can inform nutrient management decisions. For many forest companies and research organisations the collection and analysis of a large number of soil samples for chemical properties has previously been considered expensive if not cost-prohibitive. Consequently, site nutrition status may not be measured or, as is the case with many forestry managers, efforts are directed towards foliar sampling to assess tree health and vigour. Diffuse reflectance infrared spectroscopy has emerged as a quick, cost-effective, and non-destructive method for the analysis of many soil properties. Because of the advances in diffuse reflectance spectroscopy, we see potential advantages to the forest industry and research organisations by utilising this method as a cost-effective means of analysing forest soil and foliage nutrient status. In this study we build a New Zealand planted-forest specific mid-infrared spectral training library for soil and foliage chemical properties. Nationally applicable partial least squares regression models for soil and foliage chemical properties were built from the respective libraries and were applied to an independent national data set to test their robustness. The soil properties with good (R2 >0.8) and fair (R2 0.5–0.8) predictions on the independent data set were soil pH, total carbon, total nitrogen, total phosphorus, available aluminium, available calcium, CEC (cation exchange capacity), base saturation, and the following elemental totals: aluminium, calcium, iron, potassium, magnesium, nickel, and zinc. The foliage properties with fair predictions (R2 0.5–0.8) were total calcium, total potassium and total phosphorus. Total nitrogen had an R2 of 0.47 which was a result of a narrow range in nitrogen values within the test set. The root mean squared error (RMSE) for foliage nitrogen was acceptable. Most of the key planted forest soil and tree foliage nutrient properties were able to be predicted using MIR (mid infrared) spectroscopy. However, some soil available macro-nutrients and all measured soil available micro-nutrients and foliage micro-nutrients predictions were unreliable (R2 <0.5). The high throughput of samples and efficiency of analysis when using MIR spectroscopy will result in cost savings whereby forest companies and research organisations can afford to routinely analyse soil and foliar samples for their nutrient status that will then benefit the industry in their quest for productive and sustainable planted forests.