Cedar-Zen science!

By Jasper Slingsby, Vegetation Scientist, SAEON Fynbos Node Tweet

Citizen science - "science done by ordinary people, often for or with the help of scientists" - Cambridge dictionary
Zen - "emptying the mind of thoughts and giving attention to only one thing"- Cambridge dictionary

An odd title perhaps, but how best to describe the efforts of Capetonian cartographer Peter Slingsby, who spent some 200 hours using Google Earth imagery to map more than 13 000 precise locations of the Clanwilliam Cedar (Widdringtonia cedarbergensis) for the revised edition of his map Hike the Cederberg (Figure 1)?

Figure 1. An excerpt from the Hike the Cederberg Slingsby Map showing the mapped cedars (green dots)

"THIS is exciting!" said Associate Professor Ed February of the University of Cape Town’s Department of Biological Sciences and long-time cedar researcher. The development of this dataset gives us a near-perfect fine-scale description of the Clanwilliam Cedar’s distribution, allowing estimates of population numbers, providing a detailed baseline that allows monitoring of future change, and allowing inference of fine-scale habitat preferences that could lead to a better understanding of the species’ ecology and causes of its decline.

The sad, mysterious decline of an iconic tree

The Clanwilliam Cedar (Figure 2) is a long-lived tree, endemic to an area of approximately 250 km² in the Cederberg Mountains, and is currently listed by the IUCN as Critically Endangered. It is the namesake of the Cederberg mountain range and was the driver behind the declaration of the Cederberg Wilderness Area in 1973.

Figure 2. Older Widdringtonia cedarbergensis individuals are very charismatic and can be a highlight on hiking trips in the Cederberg

The cedar wood was much sought after by early settlers due to the paucity of good timber in the Cape region and rapidly led to a noticeable decline in population numbers. Felling of live cedars was banned in 1879, making it one of the first protected species in the world, and the sale of dead cedar wood was restricted in the late 19th century and banned in 1967.

Unfortunately, tree numbers have continued to dwindle, and monitoring of permanent plots indicates a >90% decline in mature individuals between 1977 and 2003. While a detrimental fire regime is commonly cited as the dominant cause of mortality, observations of high numbers of dead trees in the absence of fire has researchers suspicious of other drivers such as increased drought stress due to climate change, as witnessed for the Atlas Cedar (Cedrus atlantica) in Algeria and Morocco, or pathogens similar to the Mountain Pine Beetle that is ravaging the pine forests of North Western United States.

Zen and the art of tree mapping

When asked what drove him to spend so much time mapping the trees, cartographer Peter Slingsby replied, "Wouldn’t it just be cool to have them on the map?"

Tree localities were manually mapped from high-resolution imagery available from Google Earth Pro based on canopy colour and shadows (see Figure 3), and, where possible, verified with ground photographs from a publicly contributed archive accessible through Google Earth and a personal collection of 19 000 georeferenced images from research for the Cederberg hiking map. While there may be errors or omissions, this provides a very good indication of the locations of adult trees with live canopies.

Accuracy may vary depending on topography, the similarity of co-occurring species, and the recent occurrence of fire; but error rates for localities are likely to be low.

"...there’s an area near Krakadouw which has not burned and here the process was difficult; I may have selected some incorrect species - affecting perhaps 200 trees. Nevertheless, the cedars are usually a distinctive colour, and their shadows always give them away", states Slingsby on his blog about the Hike the Cederberg map.

Checking the numbers

An afternoon spent enumerating all cedars in a defined area on a small ridge of the Sneeuberg revealed some interesting comparisons. SAEON’s preliminary ground survey revealed 16 large adults with healthy green canopies >4 m², while the Google Earth survey revealed 13.

Figure 3. A Google Earth image of the area ground-truthed by SAEON, showing 13 healthy large adult trees mapped from the satellite imagery. The ground campaign revealed far more trees, but most were much smaller (<1 m2 canopy), unhealthy or dead.

This difference is explained by small clumps of trees being mapped as one individual, because of the difficulties of discerning multiple individuals from above. In addition, we mapped 32 large adults with <1 m² live canopy, 22 sub-adults (0.25 m²> and <1 m²) and 36 seedlings (<0.25 m²), none of which can be discerned from satellite imagery.

While this suggests that actual numbers may be much larger than obtained from the Google Earth survey, this was one of the healthier stands we have encountered and the population structure is likely to vary greatly across the Cederberg, cautioning against extrapolation. Improving the population estimates requires further surveys, stratified across environmental gradients and spanning the species’ full range.

Unfortunately, there is still strong evidence of a dramatic decline, even in this stand. Our ground survey revealed 192 dead trees (i.e. a ratio of two dead to every live tree), made up of 96 adults, 74 sub-adults and 22 seedlings, many still bearing dead leaves.

Recent work by Profs Ed February and Timm Hoffman and other researchers from the Plant Conservation Unit at the University of Cape Town (UCT) using repeat photographs (see Figure 4) suggests that this decline is consistent across the species’ range, but is more dramatic in lower elevation sites with higher temperatures and more frequent fire (White et al., submitted manuscript).

Figure 4. Matched repeat photos, such as these taken in 1977 and 2003, allow researchers to detect trends of environmental change, like this obvious loss of many cedar individuals. Where multiple repeat photos, or photos spanning different time periods, can be assembled this is a very powerful method for examining environmental change. The University of Cape Town recently launched a repeat photography citizen science project for South African landscapes.

Where do cedars call home?

The mapped tree localities allow a rough-and-ready exploration of the cedar’s habitat preferences in relation to topographical features derived from a fine-scale digital elevation model, revealing some interesting aspects of the cedar’s habitat preference (Figures 5 and 6).

Figure 5. Histograms of various topographic variables and count of fires experienced in the period 1944-2015. White histograms show data from a random sample of 100 000 points from the mapped extent shown in Figure 4, illustrating the available habitat conditions in the landscape, while the grey superimposed histograms show the conditions where the cedars actually occur.

Compared to a random sample of the available habitat, the species is more frequent at mid-to-high elevation sites (950 to 1500 m) on gentle to steep slopes with a higher Terrain Ruggedness Index (TRI - a measure of the variation in elevation in the area immediately surrounding a site). It is thought that the cedar prefers rugged terrain since rocky outcrops can provide refuge from fire.

Figure 6. Plots of the four variables explored in Figure 2, illustrating how they vary in space

Exploring C.A.P.E.’s fire history data for the Cederberg (available from http://bgis.sanbi.org/) reveals that the species occurs in sites where intermediate numbers of fires have been recorded over the past 70 years. Its rarity in sites with no or very low fire frequency in this analysis is likely because the mapped extent includes a portion of the arid Tanqua Karoo (top right of Figure 6D), which does not burn, but is too hot and dry for the cedar.

More thorough analyses of this nature, combined with further detailed habitat and climate data, have the potential to provide good insights into the species’ ecology and drivers behind the waning population numbers.

Where to from here?

The Cederberg has come to the fore as a site for SAEON to build on as a focal research area due to its iconic status within the Cape Floristic Region, the existence of many valuable long-term ecological datasets, and the support, willingness and enthusiasm of neighbouring landowners and staff within the conservation authority overseeing the Cederberg Wilderness Area and Matjies Rivier Nature Reserve (CapeNature).

Figure 7. SAEON Fynbos Node staff and CapeNature Field Ranger Nicolaas Hanekom pose next to the recently installed Automatic Weather Station above Engelsmanskloof on Sneeukop. Additional tipping-bucket rain gauges with temperature loggers were installed at Sneeukop hut and on Sneeuberg.

In collaboration with UCT and CapeNature, SAEON have recently begun deploying a network of weather stations and tipping-bucket rain gauges to better understand climatic variation in the reserve (Figure 7), and have been searching for permanently marked vegetation plots set up and surveyed by Hugh Taylor in the period 1986-9 (Figure 8), with the aim of performing a modern survey to explore vegetation change.

Figure 8. Paired photos of Taylor Plot 37 on the De Rif path above Driehoek Guest Farm. Left: Ranger, Isak Koopman, circa 1986; Right: SAEON Postdoc, Glenn Moncrieff, December 2015. So far roughly 40 of the original 120 plots have been tracked down based on site descriptions and matching photographs. Plots are marked with stone cairns with wire tags.

There are also historical demography datasets for a number of charismatic species (e.g. Protea cryophila, Protea nitida, Protea laurifolia), among other monitoring datasets and sites set up by the former South African Forestry Research Institute that have potential to provide long-term records for the area.

This article is dedicated to the late Patrick Lane, former manager of the Cederberg Wilderness Area, whose passion for the Cederberg and its people, the flora, and the Clanwilliam Cedar in particular, will remain an inspiration to all who knew him.