Introduction

As part of the New Zealand Sustainability Dashboard Project, a pilot data for 13 environmental variables were gathered from 30 orchards in Bay of Plenty between 2004 and 2010.

The dataset is available from http://www.data.govt.nz

After a quick look at the data and typing some lines of code in R, some interesting insights came along.

Exploration Data Analysis

From the Figure above we can see that the indicators are measured in alternative years. Agricultural pests and Ecosystem Composition are measured together in the same years, separately from Soil measurements. It means that a time series analysis considering all the indicators would not be appropriate. In 2008 there was no data collected.

Let’s analyse the distribution of the attributes by indicators.

Environmental Indicators Visualization

The previous barcharts show the number of attributes for each indicator as well as the distribution and variability of them.

Because in measurements like pH and Nitrogen small variations might indicate a change in the scale, for example, from acidic to neutral pH, the following Table shows the main statistics for each attribute.

## # A tibble: 13 × 6
##             Attribute        Min         Max       Mean          SD
##                <fctr>      <dbl>       <dbl>      <dbl>       <dbl>
## 1                 AMN 37.0000000 128.6666667 60.3148148 13.88146368
## 2         BulkDensity  0.5816667   0.9316667  0.7446667  0.08244099
## 3              Carbon  2.9616667   8.2150000  5.6706463  1.23492131
## 4           intro_all  6.9461970  59.0252342 21.0738714 11.44914460
## 5   intro_granivorous  1.2618158  31.4174207  8.2015098  5.90868942
## 6   intro_insectivore  2.2937433  50.8921632 12.6193776  8.34283140
## 7              native  0.0000000   9.7518746  1.7013120  1.74044054
## 8  native_insectivore  0.0000000   2.3696272  0.6285343  0.58818386
## 9       native_nectar  0.0000000   8.5368602  0.8278273  1.30124982
## 10           Nitrogen  0.2450000   0.7416667  0.4849670  0.11472103
## 11             OlsenP 23.6666667 142.6000000 58.5955556 20.82429449
## 12              pests  0.4928747  26.2636460  6.7786058  5.14588468
## 13                 pH  5.8666667   7.0666667  6.5866667  0.22358167
##        Median
##         <dbl>
## 1  58.0000000
## 2   0.7466667
## 3   5.7075000
## 4  17.8380477
## 5   6.1574431
## 6   9.9641942
## 7   1.0921862
## 8   0.4525208
## 9   0.3355575
## 10  0.4775000
## 11 58.5833333
## 12  5.0833580
## 13  6.6000000

Initial conclusions

In summary, we can conclude that Bulk Density (soil compaction), Nitrogen and pH have the lowest variability in the Soil status. The average pH level 6.58 shows an acidic soil. On the other hand, Among of potential mineralisable nitrogen (AMN) and Phosphorus readily available (OlsenP) show the highest dispersion from the mean in the Soil status.

An examination on the Ecosystem composition shows native insectivores as the attribute with the least variation. In contrast, All introduced birds (intro_all) appear as the attribute with the highest difference from the mean for this indicator.

In regards to Agricultural pests, it includes one attribute with a right skewed distribution.

Further investigation would identify if the dispersion identified in the previous attributes depends on other variables included in the dataset, such as, Orchard management system, years and clusters.