2.1.4.3 Data quality assessment

Since the quality of streamflow data depends on a number of factors including the rating curve, data gaps, methods of flow measurements and measurement errors, it cannot be meaningfully described using a single descriptor. There are also year to year variations.

The gauged data were also examined for any anomaly in the record. It was found that quite a few of the non-zero data points are repeated for many (more than 3 days) consecutive days as though the gauge is stuck at a height. For model calibration these were treated as missing data.

The daily streamflow data were examined and categorised into six generic data quality classes based on the collecting agencies’ data quality coding (Viney et al., 2011) (Table 8). The six categories are good, fair, poor, unverified, non-conforming and missing. The categories are defined as follows:

The streamflow data flagged as good, fair, poor and unverified were used while the flow data flagged as non-conforming were excluded in the model calibration. The non-conforming and missing streamflow data are both labelled in the dataset as –9999.

Table 8 Quality codes for the NSW gauges used in the Namoi subregion

Data: NSW Office of Water (Dataset 1)

Table 9 shows the proportion of data falling into each data quality category for streamflow, including data gaps, for all 37 stations for the Namoi subregion. Apart from four stations, the data quality for all the stations is labelled as predominately poor, unverified or missing. Nearly 75% of these stations have poor data as a result of a mostly uncertain rating curve (code 140 – ‘current rating – may be subject to change’), the remaining unverified data are simply not quality coded. This may result in large (not quantified here) uncertainty in the streamflow data leading to uncertainty in the modelling results.

Table 9 Percentage of data under each data quality category for streamflow data for the Namoi subregion using data from 1983 to 2012

Data: NSW Office of Water (Dataset 1, Dataset 2, Dataset 3)