Remotely monitoring ecosystem respiration from various grasslands along a large-scale east–west transect across northern China

Table 3 Regression analysis of 8-day R_e and the associated MODIS-derived products across the four grassland ecosystems using the stepwise multiple regression method

Grassland type	Period	Regression models					Training set		Test set
Grassland type	Period	Constant	EVI	LAI	LSWI	LST	R²	RMSE	R²	RMSE
Temperate meadow steppe	Growing	1.088	4.078	–	3.743	–	0.57	0.72	0.77	0.45
	Dormant	0.064	2.362	–	0.268	0.014	0.61	0.13	0.77	0.45
	Entire year	− 0.619	8.487	–	–	0.018	0.80	0.51	0.71	0.63
Typical steppe	Growing	0.561	–	–	6.058	0.096	0.72	0.60	0.88	0.41
	Dormant	0.044	3.463	–	0.659	0.021	0.41	0.26	0.88	0.41
	Entire year	− 0.083	5.633	–	1.419	0.040	0.84	0.49	0.81	0.50
Desert steppe	Growing	0.213	–	0.758	–	–	0.39	0.32	0.50	0.21
	Dormant	0.122	2.035	–	–	− 0.003	0.11	0.17	0.50	0.21
	Entire year	0.190	–	0.775	–	–	0.43	0.25	0.45	0.27
Alpine meadow	Growing	− 0.321	7.198	− 0.208	2.854	–	0.73	0.98	0.85	0.67
	Dormant	− 0.264	3.826	0.897	0.804	0.019	0.65	0.20	0.85	0.67
	Entire year	− 0.520	6.947	–	1.318	0.020	0.81	0.67	0.83	0.69

Time-series EVI, LAI, LSWI and LST data are derived from the corresponding MODIS products. All data are at an 8-day interval. The training set and test set of these flux sites are instructed in Table 1. When using the stepwise multiple regression method to develop models, only the best is exhibited here. The model’s performance was evaluated using the site-level flux measurement

ISSN: 1750-0680